Original Research

A randomized controlled field trial of a multifaceted intervention to improve preventive care tailored to the needs of participating family practices was conducted in Southern Ontario and delivered by nurses trained in the facilitation of prevention. We focus on the process evaluation and complement the outcome evaluation¹ by describing how the program was implemented in the intervention practices.

Improving preventive performance is both important and necessary. There is substantial room to improve the rates of appropriate preventive practice.² The Canadian Task Force on the Periodic Health xamination^3,4 has established guidelines for the delivery of preventive care that are supported by clinical evidence as effective in decreasing the impact of disease. However, evidence-based guidelines are not self-implementing.^5-7 Changing physicians’ long-held patterns of behavior and the environments in which they work is complex and difficult. Unless the barriers to change can be overcome and actions taken to put preventive care guidelines into practice, evidence-based guideline development efforts will be wasted, and the quality of preventive care will not improve.⁸

Several reviews have focussed on the effectiveness of different interventions for implementing guidelines and improving care.^6,7,9-13 Multifaceted interventions employing trained individuals who meet with providers in their practice settings to provide information and assist the practice in implementing evidence-based guidelines have been shown to be more effective than single interventions.^11-14 Tailoring interventions to the requirements of the practice has also been proposed as important in supporting practice changes and in attaining more successful outcomes in preventive care performance compared with interventions that are fixed and lack this flexibility.^15-17

As important as knowing what interventions work to improve preventive care performance is understanding why they work. The techniques of process evaluation allow the investigator to determine the extent to which the intervention designed to change practice patterns was actually implemented as planned. Adequate documentation of process facilitates replication and fine-tuning of the intervention.

Intervention Description

Our study built on the work of Fullard and colleagues18 and used a tailored multifaceted approach to getting evidence into action by focusing on the educational, attitudinal, and organizational barriers to change and tailoring interventions to the needs of the practice.^17-24 The intervention employed 3 prevention facilitators (PFs) with both master’s degrees in community nursing and skills and previous experience in facilitation. Each PF had primary responsibility for up to 8 primary care practices with up to 6 physicians per practice.

The PFs underwent 30 weeks of intensive training before being assigned to randomly selected intervention practices. The training covered an orientation session, medical office computer systems, medical practice management, prevention in primary care, evidence-based medicine, and facilitation and audit skills development. Approximately 28 hours per week were spent in training and 7 hours per week in preparation and planning. Six of the 30 weeks of training were spent applying skills in a primary care office setting. Once in the field, they were instructed to offer 7 intervention strategies designed to change practice patterns and improve preventive care performance. The strategies were identified from reviews of the literature and constituted the multifaceted component of the intervention.^10,11 The PFs were permitted to tailor these strategies to the needs and unique circumstances of the practice. The strategies were: (1) audit and ongoing feedback, (2) consensus building, (3) opinion leaders and networking, (4) academic detailing and education materials, (5) reminder systems, (6) patient-mediated activities, and (7) patient education materials.

The PFs worked with all physicians and allied health staff in the practice. They provided management support to practices and followed a quality improvement framework similar to that proposed by Leininger and coworkers.²⁵ For each practice the PFs were to: (1) present baseline preventive performance rates, (2) facilitate the development of a practice policy for preventive care, (3) assist in setting goals and desirable levels of performance, (4) assist in the development of a written plan for implementing preventive care, (5) assist in the development and adaptation of tools and the strategies to implement the prevention plan, (6) facilitate meetings to assess progress and modify the plan if necessary, and (7) conduct chart audits to measure the impact of the changes made. The intervention period lasted 18 months and ended in December 1998.

The Figure is the program logic model describing each of the 7 intervention component strategies and the associated work activities, outputs, and short-term and long-term objectives associated with each component. It served as a framework for the evaluation of the intervention.^26-28 The logic model allowed us to look inside the black box of the intervention^29,30 by linking implementation activity to outcomes, and provided a framework to explore which elements worked and why.

Intervention Outcomes

The goal of the intervention was to increase the performance of 8 preventive maneuvers supported by evidence as appropriate and decrease the performance of 5 preventive maneuvers supported by evidence as inappropriate.¹ An absolute change over time of 11.51% in preventive care performance in favor of intervention practices was found (F=19.29 [df=1,43], P<.0001. In other words, the intervention practices improved preventive performance by 36% going from 31% of eligible patients having received preventive care to 43% while the control practices remained at 32%.¹

Methods

Research Questions

There were 2 objectives to our process evaluation: to document the extent to which the intervention was implemented with fidelity and to gain insight into how facilitation worked to improve preventive performance. The process evaluation was designed to answer questions concerning: (1) the time involved to deliver intervention components, (2) the quality of the delivery of intervention components, and (3) physician satisfaction with the intervention components. Quality was assessed by examining the scope or range of delivery of the intervention components and by analyzing the feedback received from practices on the usefulness of the intervention components.

Setting

The intervention arm of the trial included 22 practices with 54 physicians (Table 1). All health service organizations (HSOs) in Southwestern Ontario were approached to participate in the study. HSOs are primary care practices reimbursed primarily through capitation rather than fee for service. A total of 46 of the 100 primary care practices were recruited (response rate=46%). At follow-up only one intervention practice was lost, because the entire practice had moved. Intervention and control group practices did not differ significantly on any of the measured demographic characteristics (Table 2). Complete details on practice recruitment and attrition rates are published elsewhere.¹

The practices covered a geographic area where the greatest distance between any 2 practices was more than 600 kilometers. PFs were assigned practices within a specific region of this geographic area. They arranged times to visit and work with intervention practices and traveled by car between visits to practices. PFs worked independently at their residences and corresponded with the project team through electronic mail regularly and quarterly with scheduled meetings.

Data Collection Tools

Each intervention practice was visited regularly by the same nurse facilitator who documented her activities and progress on 2 structured forms known as the weekly activity sheet and the monthly narrative report. Weekly activity sheets noted the number of hours spent on both on-site and offsite activities. Monthly narrative reports provided detailed information on the number of visits to a practice, the activities within each practice, the outcomes of those activities, the number of participants in meetings, and the plan for the following month. The activities in the narrative reports were summarized by intervention component to provide a cumulative overview of all intervention activity within a practice.

Also during the intervention, semistructured telephone interviews of participating physicians were conducted by 2 physician members of the project team at 6 months and 17 months. Participating physicians were asked what they had been happy and unhappy with and their ideas of improvement. Close-ended questions measured overall satisfaction with the intervention. The interview at 17 months also asked physicians if they would agree to have a nurse facilitator continue to visit their practice if funding were found.

At the end of the intervention, the PFs conducted interviews with each of the physicians identified as the primary contact in the intervention practices to solicit feedback on their experience. Physicians in both the intervention and control arm were sent a questionnaire by mail to report any changes that had taken place in their practice over the preceding 18 months.

Analysis

Data were analyzed to address the 3 research questions for the process evaluation utilizing the Logic Model as the conceptual framework (Figure). To determine how often various intervention components were delivered, the total hours spent at each practice and the total number of contacts with each practice by intervention component were calculated from the PF activity sheets.

To determine intervention quality, triangulation³¹ was used to attain a complete understanding of the quality of implementation. Multiple data sources and analysis methods were used to reveal the underlying dimensions of quality. All data sources were reviewed and analyses were conducted independently by 2 members of the investigation team. The members of the team held a debriefing session to discuss their findings and seek consensus. First, the monthly narrative reports across intervention sites were summarized to qualitatively describe the type, breadth, and scope of activity for each intervention component. Second, the activity descriptions and open-ended interview responses were content analyzed³² and coded, and frequencies were generated. The goal of this analysis was to identify significant descriptions of which intervention elements worked well and which did not. Finally, intervention and control practices were compared with contingency tables, and a chi-square statistic was used to determine differences on questionnaire responses concerning practice changes over the period of the intervention.

To determine physician satisfaction, open-ended satisfaction survey responses were coded and frequencies generated for ratings of overall satisfaction with the performance of the PF and the intervention.

Results

PF Program Implementation

Table 3 shows the number of hours spent on project activities during the period of the intervention. The PFs spent the largest proportion of their time (28%) on administrative duties, such as team meetings, telephone calls, internal reporting, preparing the project newsletter, coordinating networking conferences for intervention practices, photocopying, and filing. Sixteen percent of the PFs’ time was spent on-site facilitating changes to improve preventive care in the practice. Travel accounted for an average of 12% of the PFs’ time, although this varied depending on the distance to the practices.

Table 4 provides information on the number of contacts and hours spent on-site for each component of the intervention. On average, each intervention practice was contacted 33 times by a PF, with each visit lasting an average of 1 hour and 45 minutes. The most frequent forms of contact concerned developing reminder systems, conducting chart audits and providing feedback preventive care performance, and working to achieve consensus on the adoption of preventive care strategies. Both academic detailing to physicians and supplying critically appraised patient education materials averaged approximately 20 minutes but involved a great deal of preparation time. Few practices were interested in posters in the waiting room or a patient newsletter on prevention, so fewer contacts were made for those components.

Quality of Implementation

To assess quality, the frequency of each component of the intervention was tallied, physician feedback on the usefulness of intervention components was summarized, and self-reported practice changes between intervention and control physicians was reported.

Intervention Scope

Audit and Feedback. All 22 intervention practices received a presentation by the PF on the initial audit results to raise awareness of preventive care practice patterns. This was usually done in a kick-off meeting involving both physicians and nurses and often required more than one presentation to cover the various staff in the practice. Twenty practices requested subsequent analyses of data to follow their rates of performance. In addition, 18 practices requested audits of their charts for specific maneuvers, such as influenza vaccination and mammography

Consensus Building. All practices were involved in meetings with the PF to identify opportunities for improvement, assess needs, and select priority areas and strategies for improving preventive care performance. Interviews were conducted with nurses and other staff to promote their role in preventive care delivery.

Academic Detailing. Twenty-one out of 22 sites agreed to receive and discuss critically appraised evidence for the preventive maneuvers under study, and some requested similar information on other preventive areas, such as cholesterol and osteoporosis.

Reminder Systems. All of the intervention sites implemented some form of reminder system. Eighteen sites implemented a preventive care flow sheet; 2 sites used a chart stamp; and 2 sites implemented a computerized reminder system. Nineteen sites developed recall initiatives for flu vaccine, mammography, and Papanicolaou tests. Seventeen sites implemented chart stickers for smoking counseling or mammography.

Opinion Leaders. All sites received copies of the PF project newsletter that contained articles by influential individuals describing the importance of preventive care and descriptions of colleagues’ preventive care implementation efforts. Most practices attended a workshop that included an influential keynote speaker, and 27% of the participating physicians shared their knowledge about preventive care through publishing in the newsletter and/or public speaking.

Patient Education. All sites were provided with patient education materials from credible sources on request, and all received a binder of patient education materials constructed specifically to contain materials on the appropriate preventive maneuvers under study. The binders were regularly updated.

Patient Mediated. Posters designed to prompt patients to ask about folic acid, flu vaccine, and mammography were offered to all sites. Thirteen sites implemented a patient consent form for prostate-specific antigen (PSA) testing. Eight sites received preventive care diaries for patients. Five sites had a prevention newsletter for patients. Four sites agreed to pilot a health risk appraisal software program.

Physician Feedback

At the end of the intervention the facilitator asked physicians about their experience, including what was most and least useful to them. Table 5 provides a summary of the content analysis of physician responses.

Audit and feedback, both initially and subsequently, comprised the component most frequently considered to be important in creating change. Almost as often, the preventive care flow sheet was identified as useful. The facilitator sessions designed to seek consensus on preventive care guidelines and strategies for implementation were also appreciated.

Several physicians did not agree with the evidence on PSA testing. Others did not feel that counseling for folic acid was a priority. Some found the patient education binder cumbersome, and others found the sticker system for tobacco counseling unwieldy. Thus, both were underused. Two physicians noted that the preventive care wall chart was not helpful.

Physician Self-Reported Practice Changes

Eighty-six percent (93/108) of the intervention and control physicians responded to a questionnaire at the end of the study. Due to sample size, statistical power was limited to detecting an absolute difference of approximately 0.30 between groups, assuming an alpha of 0.05 and 80% power.³³ Table 6 shows that 71% of intervention physicians compared with 28% of control physicians reported an audit of their practice for preventive services (P<.001). By the end of the study, 65% of the intervention physicians versus 48% of the control physicians indicated that they had a prevention policy or screening protocol in place, and 70% of intervention physicians compared with 58% of control physicians had created reminder systems for disease prevention.

Satisfaction with PF Intervention

At the telephone interview 6 months into the intervention, the mean satisfaction rating of intervention physicians was 4.08 on a scale of 1 (very dissatisfied) to 5 (very satisfied) with 80% satisfied with the intervention. At 17 months the mean satisfaction rating had risen to 4.5 with a 95% satisfaction rate.

At 6 months, 85% of the practices were satisfied with the frequency of visits of their assigned facilitator. At 17 months there was a 64% satisfaction rate, with the remaining 36% wanting more visits from the facilitator. The physicians commented on how the intervention had focused them on prevention in their practice. When the physicians were asked if they would agree to have a facilitator visit their practice in the future if given the opportunity, 90% agreed.

Concerns included not being able to continue the recall of patients at the end of the intervention and questioning the inappropriate maneuvers as too controversial. A physician from a large practice with 6 physicians commented that the facilitator could not easily work in the complex practice environment.

Discussion

Our study demonstrates that PFs can significantly improve the delivery of preventive services and in the process make quality contributions to a practice environment with high satisfaction rates from participating physicians.

Our intervention had a higher frequency and intensity of visits than other studies of this genre. The PFs had an average of almost 2 visits per month lasting approximately 105 minutes per visit. Dietrich and colleagues²¹ reported only 4 visits over a 3-month period lasting an average of 120 minutes, and Hulscher and coworkers²² reported approximately 25 visits with an average duration of 73 minutes. Others have been even less frequent,²⁰ and in other studies it is not reported.^34-36

The critical intervention components as evidenced by physician feedback, changes between control and intervention practices, and the amount of facilitator time spent on each component were: (1) audit and feedback, (2) sharing and discussing information to build consensus on an action plan, and (3) a reminder system. Similarly, the Cancer Prevention in Community Practice Project achieved 100% success in implementing change using customized preventive care flowsheets.³⁷ Of the 7 intervention components, patient education materials and patient-mediated interventions such as posters in the waiting room were considered to be the least useful.

Overall, physicians and nurses working within the practices were very satisfied with the intervention, and 90% were willing to have the nurse facilitator continue working with their practice.

Lessons learned from the process evaluation for improving the delivery of the outreach facilitation intervention include:

Focusing on the 3 key intervention components (audit and feedback, seeking consensus on an action plan, and implementing a reminder system) and tailoring these to the needs of the practice

Preparing patient education and patient-mediated materials only if the practice requests such materials

Developing simpler strategies to encourage physicians to counsel their patients that smoke to quit smoking

Providing the facilitators an administrative assistant to reduce the amount of their time spent on administrative duties for the practices and increase time on-site

Strengths

The strengths of the study include the completeness of the data set, the theoretical framework for data collection, the use of multiple data sources and data collection methods, and the prospective data collection methodology.

Limitations

There are several limitations to the process evaluation methods. Much of the data was provided by the facilitators themselves, and therefore the possibility of bias exists. The study population consisted of HSOs, and therefore the results may not be generalizable. There is a possibility of social desirability bias in the satisfaction rates. Finally, our analyses of the process data were descriptive and exploratory.

Conclusions

Process evaluation often identifies future areas of research. Follow-up of the few practices that were dissatisfied with facilitation should be carried out to understand why they were dissatisfied. Sustainability needs to be addressed. For example, Dietrich and colleagues³⁸ found that 5-year durability of a preventive services office system depended on the physician’s preventive care philosophy. McCowan and coworkers³⁹ found that the effect of a facilitator was not sustained for 2 years. Finally, to maximize cost-effectiveness, more research is required to determine how much of a dose of facilitation is required and how frequently facilitators should visit to achieve a positive outcome.

Acknowledgments

We wish to acknowledge the financial support of the Ontario Ministry of Health, as well as the substantial contributions of the 3 nurse facilitators (Ingrid LeClaire, Ann MacLeod, and Ruth Blochlinger). We also wish to thank the many physicians and nurses who participated in the study.

A randomized controlled field trial of a multifaceted intervention to improve preventive care tailored to the needs of participating family practices was conducted in Southern Ontario and delivered by nurses trained in the facilitation of prevention. We focus on the process evaluation and complement the outcome evaluation¹ by describing how the program was implemented in the intervention practices.

Improving preventive performance is both important and necessary. There is substantial room to improve the rates of appropriate preventive practice.² The Canadian Task Force on the Periodic Health xamination^3,4 has established guidelines for the delivery of preventive care that are supported by clinical evidence as effective in decreasing the impact of disease. However, evidence-based guidelines are not self-implementing.^5-7 Changing physicians’ long-held patterns of behavior and the environments in which they work is complex and difficult. Unless the barriers to change can be overcome and actions taken to put preventive care guidelines into practice, evidence-based guideline development efforts will be wasted, and the quality of preventive care will not improve.⁸

Several reviews have focussed on the effectiveness of different interventions for implementing guidelines and improving care.^6,7,9-13 Multifaceted interventions employing trained individuals who meet with providers in their practice settings to provide information and assist the practice in implementing evidence-based guidelines have been shown to be more effective than single interventions.^11-14 Tailoring interventions to the requirements of the practice has also been proposed as important in supporting practice changes and in attaining more successful outcomes in preventive care performance compared with interventions that are fixed and lack this flexibility.^15-17

As important as knowing what interventions work to improve preventive care performance is understanding why they work. The techniques of process evaluation allow the investigator to determine the extent to which the intervention designed to change practice patterns was actually implemented as planned. Adequate documentation of process facilitates replication and fine-tuning of the intervention.

Intervention Description

Our study built on the work of Fullard and colleagues18 and used a tailored multifaceted approach to getting evidence into action by focusing on the educational, attitudinal, and organizational barriers to change and tailoring interventions to the needs of the practice.^17-24 The intervention employed 3 prevention facilitators (PFs) with both master’s degrees in community nursing and skills and previous experience in facilitation. Each PF had primary responsibility for up to 8 primary care practices with up to 6 physicians per practice.

The PFs underwent 30 weeks of intensive training before being assigned to randomly selected intervention practices. The training covered an orientation session, medical office computer systems, medical practice management, prevention in primary care, evidence-based medicine, and facilitation and audit skills development. Approximately 28 hours per week were spent in training and 7 hours per week in preparation and planning. Six of the 30 weeks of training were spent applying skills in a primary care office setting. Once in the field, they were instructed to offer 7 intervention strategies designed to change practice patterns and improve preventive care performance. The strategies were identified from reviews of the literature and constituted the multifaceted component of the intervention.^10,11 The PFs were permitted to tailor these strategies to the needs and unique circumstances of the practice. The strategies were: (1) audit and ongoing feedback, (2) consensus building, (3) opinion leaders and networking, (4) academic detailing and education materials, (5) reminder systems, (6) patient-mediated activities, and (7) patient education materials.

The PFs worked with all physicians and allied health staff in the practice. They provided management support to practices and followed a quality improvement framework similar to that proposed by Leininger and coworkers.²⁵ For each practice the PFs were to: (1) present baseline preventive performance rates, (2) facilitate the development of a practice policy for preventive care, (3) assist in setting goals and desirable levels of performance, (4) assist in the development of a written plan for implementing preventive care, (5) assist in the development and adaptation of tools and the strategies to implement the prevention plan, (6) facilitate meetings to assess progress and modify the plan if necessary, and (7) conduct chart audits to measure the impact of the changes made. The intervention period lasted 18 months and ended in December 1998.

The Figure is the program logic model describing each of the 7 intervention component strategies and the associated work activities, outputs, and short-term and long-term objectives associated with each component. It served as a framework for the evaluation of the intervention.^26-28 The logic model allowed us to look inside the black box of the intervention^29,30 by linking implementation activity to outcomes, and provided a framework to explore which elements worked and why.

Intervention Outcomes

The goal of the intervention was to increase the performance of 8 preventive maneuvers supported by evidence as appropriate and decrease the performance of 5 preventive maneuvers supported by evidence as inappropriate.¹ An absolute change over time of 11.51% in preventive care performance in favor of intervention practices was found (F=19.29 [df=1,43], P<.0001. In other words, the intervention practices improved preventive performance by 36% going from 31% of eligible patients having received preventive care to 43% while the control practices remained at 32%.¹

Methods

Research Questions

There were 2 objectives to our process evaluation: to document the extent to which the intervention was implemented with fidelity and to gain insight into how facilitation worked to improve preventive performance. The process evaluation was designed to answer questions concerning: (1) the time involved to deliver intervention components, (2) the quality of the delivery of intervention components, and (3) physician satisfaction with the intervention components. Quality was assessed by examining the scope or range of delivery of the intervention components and by analyzing the feedback received from practices on the usefulness of the intervention components.

Setting

The intervention arm of the trial included 22 practices with 54 physicians (Table 1). All health service organizations (HSOs) in Southwestern Ontario were approached to participate in the study. HSOs are primary care practices reimbursed primarily through capitation rather than fee for service. A total of 46 of the 100 primary care practices were recruited (response rate=46%). At follow-up only one intervention practice was lost, because the entire practice had moved. Intervention and control group practices did not differ significantly on any of the measured demographic characteristics (Table 2). Complete details on practice recruitment and attrition rates are published elsewhere.¹

The practices covered a geographic area where the greatest distance between any 2 practices was more than 600 kilometers. PFs were assigned practices within a specific region of this geographic area. They arranged times to visit and work with intervention practices and traveled by car between visits to practices. PFs worked independently at their residences and corresponded with the project team through electronic mail regularly and quarterly with scheduled meetings.

Data Collection Tools

Each intervention practice was visited regularly by the same nurse facilitator who documented her activities and progress on 2 structured forms known as the weekly activity sheet and the monthly narrative report. Weekly activity sheets noted the number of hours spent on both on-site and offsite activities. Monthly narrative reports provided detailed information on the number of visits to a practice, the activities within each practice, the outcomes of those activities, the number of participants in meetings, and the plan for the following month. The activities in the narrative reports were summarized by intervention component to provide a cumulative overview of all intervention activity within a practice.

Also during the intervention, semistructured telephone interviews of participating physicians were conducted by 2 physician members of the project team at 6 months and 17 months. Participating physicians were asked what they had been happy and unhappy with and their ideas of improvement. Close-ended questions measured overall satisfaction with the intervention. The interview at 17 months also asked physicians if they would agree to have a nurse facilitator continue to visit their practice if funding were found.

At the end of the intervention, the PFs conducted interviews with each of the physicians identified as the primary contact in the intervention practices to solicit feedback on their experience. Physicians in both the intervention and control arm were sent a questionnaire by mail to report any changes that had taken place in their practice over the preceding 18 months.

Analysis

Data were analyzed to address the 3 research questions for the process evaluation utilizing the Logic Model as the conceptual framework (Figure). To determine how often various intervention components were delivered, the total hours spent at each practice and the total number of contacts with each practice by intervention component were calculated from the PF activity sheets.

To determine intervention quality, triangulation³¹ was used to attain a complete understanding of the quality of implementation. Multiple data sources and analysis methods were used to reveal the underlying dimensions of quality. All data sources were reviewed and analyses were conducted independently by 2 members of the investigation team. The members of the team held a debriefing session to discuss their findings and seek consensus. First, the monthly narrative reports across intervention sites were summarized to qualitatively describe the type, breadth, and scope of activity for each intervention component. Second, the activity descriptions and open-ended interview responses were content analyzed³² and coded, and frequencies were generated. The goal of this analysis was to identify significant descriptions of which intervention elements worked well and which did not. Finally, intervention and control practices were compared with contingency tables, and a chi-square statistic was used to determine differences on questionnaire responses concerning practice changes over the period of the intervention.

To determine physician satisfaction, open-ended satisfaction survey responses were coded and frequencies generated for ratings of overall satisfaction with the performance of the PF and the intervention.

Results

PF Program Implementation

Table 3 shows the number of hours spent on project activities during the period of the intervention. The PFs spent the largest proportion of their time (28%) on administrative duties, such as team meetings, telephone calls, internal reporting, preparing the project newsletter, coordinating networking conferences for intervention practices, photocopying, and filing. Sixteen percent of the PFs’ time was spent on-site facilitating changes to improve preventive care in the practice. Travel accounted for an average of 12% of the PFs’ time, although this varied depending on the distance to the practices.

Table 4 provides information on the number of contacts and hours spent on-site for each component of the intervention. On average, each intervention practice was contacted 33 times by a PF, with each visit lasting an average of 1 hour and 45 minutes. The most frequent forms of contact concerned developing reminder systems, conducting chart audits and providing feedback preventive care performance, and working to achieve consensus on the adoption of preventive care strategies. Both academic detailing to physicians and supplying critically appraised patient education materials averaged approximately 20 minutes but involved a great deal of preparation time. Few practices were interested in posters in the waiting room or a patient newsletter on prevention, so fewer contacts were made for those components.

Quality of Implementation

To assess quality, the frequency of each component of the intervention was tallied, physician feedback on the usefulness of intervention components was summarized, and self-reported practice changes between intervention and control physicians was reported.

Intervention Scope

Audit and Feedback. All 22 intervention practices received a presentation by the PF on the initial audit results to raise awareness of preventive care practice patterns. This was usually done in a kick-off meeting involving both physicians and nurses and often required more than one presentation to cover the various staff in the practice. Twenty practices requested subsequent analyses of data to follow their rates of performance. In addition, 18 practices requested audits of their charts for specific maneuvers, such as influenza vaccination and mammography

Consensus Building. All practices were involved in meetings with the PF to identify opportunities for improvement, assess needs, and select priority areas and strategies for improving preventive care performance. Interviews were conducted with nurses and other staff to promote their role in preventive care delivery.

Academic Detailing. Twenty-one out of 22 sites agreed to receive and discuss critically appraised evidence for the preventive maneuvers under study, and some requested similar information on other preventive areas, such as cholesterol and osteoporosis.

Reminder Systems. All of the intervention sites implemented some form of reminder system. Eighteen sites implemented a preventive care flow sheet; 2 sites used a chart stamp; and 2 sites implemented a computerized reminder system. Nineteen sites developed recall initiatives for flu vaccine, mammography, and Papanicolaou tests. Seventeen sites implemented chart stickers for smoking counseling or mammography.

Opinion Leaders. All sites received copies of the PF project newsletter that contained articles by influential individuals describing the importance of preventive care and descriptions of colleagues’ preventive care implementation efforts. Most practices attended a workshop that included an influential keynote speaker, and 27% of the participating physicians shared their knowledge about preventive care through publishing in the newsletter and/or public speaking.

Patient Education. All sites were provided with patient education materials from credible sources on request, and all received a binder of patient education materials constructed specifically to contain materials on the appropriate preventive maneuvers under study. The binders were regularly updated.

Patient Mediated. Posters designed to prompt patients to ask about folic acid, flu vaccine, and mammography were offered to all sites. Thirteen sites implemented a patient consent form for prostate-specific antigen (PSA) testing. Eight sites received preventive care diaries for patients. Five sites had a prevention newsletter for patients. Four sites agreed to pilot a health risk appraisal software program.

Physician Feedback

At the end of the intervention the facilitator asked physicians about their experience, including what was most and least useful to them. Table 5 provides a summary of the content analysis of physician responses.

Audit and feedback, both initially and subsequently, comprised the component most frequently considered to be important in creating change. Almost as often, the preventive care flow sheet was identified as useful. The facilitator sessions designed to seek consensus on preventive care guidelines and strategies for implementation were also appreciated.

Several physicians did not agree with the evidence on PSA testing. Others did not feel that counseling for folic acid was a priority. Some found the patient education binder cumbersome, and others found the sticker system for tobacco counseling unwieldy. Thus, both were underused. Two physicians noted that the preventive care wall chart was not helpful.

Physician Self-Reported Practice Changes

Eighty-six percent (93/108) of the intervention and control physicians responded to a questionnaire at the end of the study. Due to sample size, statistical power was limited to detecting an absolute difference of approximately 0.30 between groups, assuming an alpha of 0.05 and 80% power.³³ Table 6 shows that 71% of intervention physicians compared with 28% of control physicians reported an audit of their practice for preventive services (P<.001). By the end of the study, 65% of the intervention physicians versus 48% of the control physicians indicated that they had a prevention policy or screening protocol in place, and 70% of intervention physicians compared with 58% of control physicians had created reminder systems for disease prevention.

Satisfaction with PF Intervention

At the telephone interview 6 months into the intervention, the mean satisfaction rating of intervention physicians was 4.08 on a scale of 1 (very dissatisfied) to 5 (very satisfied) with 80% satisfied with the intervention. At 17 months the mean satisfaction rating had risen to 4.5 with a 95% satisfaction rate.

At 6 months, 85% of the practices were satisfied with the frequency of visits of their assigned facilitator. At 17 months there was a 64% satisfaction rate, with the remaining 36% wanting more visits from the facilitator. The physicians commented on how the intervention had focused them on prevention in their practice. When the physicians were asked if they would agree to have a facilitator visit their practice in the future if given the opportunity, 90% agreed.

Concerns included not being able to continue the recall of patients at the end of the intervention and questioning the inappropriate maneuvers as too controversial. A physician from a large practice with 6 physicians commented that the facilitator could not easily work in the complex practice environment.

Discussion

Our study demonstrates that PFs can significantly improve the delivery of preventive services and in the process make quality contributions to a practice environment with high satisfaction rates from participating physicians.

Our intervention had a higher frequency and intensity of visits than other studies of this genre. The PFs had an average of almost 2 visits per month lasting approximately 105 minutes per visit. Dietrich and colleagues²¹ reported only 4 visits over a 3-month period lasting an average of 120 minutes, and Hulscher and coworkers²² reported approximately 25 visits with an average duration of 73 minutes. Others have been even less frequent,²⁰ and in other studies it is not reported.^34-36

The critical intervention components as evidenced by physician feedback, changes between control and intervention practices, and the amount of facilitator time spent on each component were: (1) audit and feedback, (2) sharing and discussing information to build consensus on an action plan, and (3) a reminder system. Similarly, the Cancer Prevention in Community Practice Project achieved 100% success in implementing change using customized preventive care flowsheets.³⁷ Of the 7 intervention components, patient education materials and patient-mediated interventions such as posters in the waiting room were considered to be the least useful.

Overall, physicians and nurses working within the practices were very satisfied with the intervention, and 90% were willing to have the nurse facilitator continue working with their practice.

Lessons learned from the process evaluation for improving the delivery of the outreach facilitation intervention include:

Focusing on the 3 key intervention components (audit and feedback, seeking consensus on an action plan, and implementing a reminder system) and tailoring these to the needs of the practice

Preparing patient education and patient-mediated materials only if the practice requests such materials

Developing simpler strategies to encourage physicians to counsel their patients that smoke to quit smoking

Providing the facilitators an administrative assistant to reduce the amount of their time spent on administrative duties for the practices and increase time on-site

Strengths

The strengths of the study include the completeness of the data set, the theoretical framework for data collection, the use of multiple data sources and data collection methods, and the prospective data collection methodology.

Limitations

There are several limitations to the process evaluation methods. Much of the data was provided by the facilitators themselves, and therefore the possibility of bias exists. The study population consisted of HSOs, and therefore the results may not be generalizable. There is a possibility of social desirability bias in the satisfaction rates. Finally, our analyses of the process data were descriptive and exploratory.

Conclusions

Process evaluation often identifies future areas of research. Follow-up of the few practices that were dissatisfied with facilitation should be carried out to understand why they were dissatisfied. Sustainability needs to be addressed. For example, Dietrich and colleagues³⁸ found that 5-year durability of a preventive services office system depended on the physician’s preventive care philosophy. McCowan and coworkers³⁹ found that the effect of a facilitator was not sustained for 2 years. Finally, to maximize cost-effectiveness, more research is required to determine how much of a dose of facilitation is required and how frequently facilitators should visit to achieve a positive outcome.

Acknowledgments

We wish to acknowledge the financial support of the Ontario Ministry of Health, as well as the substantial contributions of the 3 nurse facilitators (Ingrid LeClaire, Ann MacLeod, and Ruth Blochlinger). We also wish to thank the many physicians and nurses who participated in the study.

A randomized controlled field trial of a multifaceted intervention to improve preventive care tailored to the needs of participating family practices was conducted in Southern Ontario and delivered by nurses trained in the facilitation of prevention. We focus on the process evaluation and complement the outcome evaluation¹ by describing how the program was implemented in the intervention practices.

Improving preventive performance is both important and necessary. There is substantial room to improve the rates of appropriate preventive practice.² The Canadian Task Force on the Periodic Health xamination^3,4 has established guidelines for the delivery of preventive care that are supported by clinical evidence as effective in decreasing the impact of disease. However, evidence-based guidelines are not self-implementing.^5-7 Changing physicians’ long-held patterns of behavior and the environments in which they work is complex and difficult. Unless the barriers to change can be overcome and actions taken to put preventive care guidelines into practice, evidence-based guideline development efforts will be wasted, and the quality of preventive care will not improve.⁸

Several reviews have focussed on the effectiveness of different interventions for implementing guidelines and improving care.^6,7,9-13 Multifaceted interventions employing trained individuals who meet with providers in their practice settings to provide information and assist the practice in implementing evidence-based guidelines have been shown to be more effective than single interventions.^11-14 Tailoring interventions to the requirements of the practice has also been proposed as important in supporting practice changes and in attaining more successful outcomes in preventive care performance compared with interventions that are fixed and lack this flexibility.^15-17

As important as knowing what interventions work to improve preventive care performance is understanding why they work. The techniques of process evaluation allow the investigator to determine the extent to which the intervention designed to change practice patterns was actually implemented as planned. Adequate documentation of process facilitates replication and fine-tuning of the intervention.

Intervention Description

Our study built on the work of Fullard and colleagues18 and used a tailored multifaceted approach to getting evidence into action by focusing on the educational, attitudinal, and organizational barriers to change and tailoring interventions to the needs of the practice.^17-24 The intervention employed 3 prevention facilitators (PFs) with both master’s degrees in community nursing and skills and previous experience in facilitation. Each PF had primary responsibility for up to 8 primary care practices with up to 6 physicians per practice.

The PFs underwent 30 weeks of intensive training before being assigned to randomly selected intervention practices. The training covered an orientation session, medical office computer systems, medical practice management, prevention in primary care, evidence-based medicine, and facilitation and audit skills development. Approximately 28 hours per week were spent in training and 7 hours per week in preparation and planning. Six of the 30 weeks of training were spent applying skills in a primary care office setting. Once in the field, they were instructed to offer 7 intervention strategies designed to change practice patterns and improve preventive care performance. The strategies were identified from reviews of the literature and constituted the multifaceted component of the intervention.^10,11 The PFs were permitted to tailor these strategies to the needs and unique circumstances of the practice. The strategies were: (1) audit and ongoing feedback, (2) consensus building, (3) opinion leaders and networking, (4) academic detailing and education materials, (5) reminder systems, (6) patient-mediated activities, and (7) patient education materials.

The PFs worked with all physicians and allied health staff in the practice. They provided management support to practices and followed a quality improvement framework similar to that proposed by Leininger and coworkers.²⁵ For each practice the PFs were to: (1) present baseline preventive performance rates, (2) facilitate the development of a practice policy for preventive care, (3) assist in setting goals and desirable levels of performance, (4) assist in the development of a written plan for implementing preventive care, (5) assist in the development and adaptation of tools and the strategies to implement the prevention plan, (6) facilitate meetings to assess progress and modify the plan if necessary, and (7) conduct chart audits to measure the impact of the changes made. The intervention period lasted 18 months and ended in December 1998.

The Figure is the program logic model describing each of the 7 intervention component strategies and the associated work activities, outputs, and short-term and long-term objectives associated with each component. It served as a framework for the evaluation of the intervention.^26-28 The logic model allowed us to look inside the black box of the intervention^29,30 by linking implementation activity to outcomes, and provided a framework to explore which elements worked and why.

Intervention Outcomes

The goal of the intervention was to increase the performance of 8 preventive maneuvers supported by evidence as appropriate and decrease the performance of 5 preventive maneuvers supported by evidence as inappropriate.¹ An absolute change over time of 11.51% in preventive care performance in favor of intervention practices was found (F=19.29 [df=1,43], P<.0001. In other words, the intervention practices improved preventive performance by 36% going from 31% of eligible patients having received preventive care to 43% while the control practices remained at 32%.¹

Methods

Research Questions

There were 2 objectives to our process evaluation: to document the extent to which the intervention was implemented with fidelity and to gain insight into how facilitation worked to improve preventive performance. The process evaluation was designed to answer questions concerning: (1) the time involved to deliver intervention components, (2) the quality of the delivery of intervention components, and (3) physician satisfaction with the intervention components. Quality was assessed by examining the scope or range of delivery of the intervention components and by analyzing the feedback received from practices on the usefulness of the intervention components.

Setting

The intervention arm of the trial included 22 practices with 54 physicians (Table 1). All health service organizations (HSOs) in Southwestern Ontario were approached to participate in the study. HSOs are primary care practices reimbursed primarily through capitation rather than fee for service. A total of 46 of the 100 primary care practices were recruited (response rate=46%). At follow-up only one intervention practice was lost, because the entire practice had moved. Intervention and control group practices did not differ significantly on any of the measured demographic characteristics (Table 2). Complete details on practice recruitment and attrition rates are published elsewhere.¹

The practices covered a geographic area where the greatest distance between any 2 practices was more than 600 kilometers. PFs were assigned practices within a specific region of this geographic area. They arranged times to visit and work with intervention practices and traveled by car between visits to practices. PFs worked independently at their residences and corresponded with the project team through electronic mail regularly and quarterly with scheduled meetings.

Data Collection Tools

Each intervention practice was visited regularly by the same nurse facilitator who documented her activities and progress on 2 structured forms known as the weekly activity sheet and the monthly narrative report. Weekly activity sheets noted the number of hours spent on both on-site and offsite activities. Monthly narrative reports provided detailed information on the number of visits to a practice, the activities within each practice, the outcomes of those activities, the number of participants in meetings, and the plan for the following month. The activities in the narrative reports were summarized by intervention component to provide a cumulative overview of all intervention activity within a practice.

Also during the intervention, semistructured telephone interviews of participating physicians were conducted by 2 physician members of the project team at 6 months and 17 months. Participating physicians were asked what they had been happy and unhappy with and their ideas of improvement. Close-ended questions measured overall satisfaction with the intervention. The interview at 17 months also asked physicians if they would agree to have a nurse facilitator continue to visit their practice if funding were found.

At the end of the intervention, the PFs conducted interviews with each of the physicians identified as the primary contact in the intervention practices to solicit feedback on their experience. Physicians in both the intervention and control arm were sent a questionnaire by mail to report any changes that had taken place in their practice over the preceding 18 months.

Analysis

Data were analyzed to address the 3 research questions for the process evaluation utilizing the Logic Model as the conceptual framework (Figure). To determine how often various intervention components were delivered, the total hours spent at each practice and the total number of contacts with each practice by intervention component were calculated from the PF activity sheets.

To determine intervention quality, triangulation³¹ was used to attain a complete understanding of the quality of implementation. Multiple data sources and analysis methods were used to reveal the underlying dimensions of quality. All data sources were reviewed and analyses were conducted independently by 2 members of the investigation team. The members of the team held a debriefing session to discuss their findings and seek consensus. First, the monthly narrative reports across intervention sites were summarized to qualitatively describe the type, breadth, and scope of activity for each intervention component. Second, the activity descriptions and open-ended interview responses were content analyzed³² and coded, and frequencies were generated. The goal of this analysis was to identify significant descriptions of which intervention elements worked well and which did not. Finally, intervention and control practices were compared with contingency tables, and a chi-square statistic was used to determine differences on questionnaire responses concerning practice changes over the period of the intervention.

To determine physician satisfaction, open-ended satisfaction survey responses were coded and frequencies generated for ratings of overall satisfaction with the performance of the PF and the intervention.

Results

PF Program Implementation

Table 3 shows the number of hours spent on project activities during the period of the intervention. The PFs spent the largest proportion of their time (28%) on administrative duties, such as team meetings, telephone calls, internal reporting, preparing the project newsletter, coordinating networking conferences for intervention practices, photocopying, and filing. Sixteen percent of the PFs’ time was spent on-site facilitating changes to improve preventive care in the practice. Travel accounted for an average of 12% of the PFs’ time, although this varied depending on the distance to the practices.

Table 4 provides information on the number of contacts and hours spent on-site for each component of the intervention. On average, each intervention practice was contacted 33 times by a PF, with each visit lasting an average of 1 hour and 45 minutes. The most frequent forms of contact concerned developing reminder systems, conducting chart audits and providing feedback preventive care performance, and working to achieve consensus on the adoption of preventive care strategies. Both academic detailing to physicians and supplying critically appraised patient education materials averaged approximately 20 minutes but involved a great deal of preparation time. Few practices were interested in posters in the waiting room or a patient newsletter on prevention, so fewer contacts were made for those components.

Quality of Implementation

To assess quality, the frequency of each component of the intervention was tallied, physician feedback on the usefulness of intervention components was summarized, and self-reported practice changes between intervention and control physicians was reported.

Intervention Scope

Audit and Feedback. All 22 intervention practices received a presentation by the PF on the initial audit results to raise awareness of preventive care practice patterns. This was usually done in a kick-off meeting involving both physicians and nurses and often required more than one presentation to cover the various staff in the practice. Twenty practices requested subsequent analyses of data to follow their rates of performance. In addition, 18 practices requested audits of their charts for specific maneuvers, such as influenza vaccination and mammography

Consensus Building. All practices were involved in meetings with the PF to identify opportunities for improvement, assess needs, and select priority areas and strategies for improving preventive care performance. Interviews were conducted with nurses and other staff to promote their role in preventive care delivery.

Academic Detailing. Twenty-one out of 22 sites agreed to receive and discuss critically appraised evidence for the preventive maneuvers under study, and some requested similar information on other preventive areas, such as cholesterol and osteoporosis.

Reminder Systems. All of the intervention sites implemented some form of reminder system. Eighteen sites implemented a preventive care flow sheet; 2 sites used a chart stamp; and 2 sites implemented a computerized reminder system. Nineteen sites developed recall initiatives for flu vaccine, mammography, and Papanicolaou tests. Seventeen sites implemented chart stickers for smoking counseling or mammography.

Opinion Leaders. All sites received copies of the PF project newsletter that contained articles by influential individuals describing the importance of preventive care and descriptions of colleagues’ preventive care implementation efforts. Most practices attended a workshop that included an influential keynote speaker, and 27% of the participating physicians shared their knowledge about preventive care through publishing in the newsletter and/or public speaking.

Patient Education. All sites were provided with patient education materials from credible sources on request, and all received a binder of patient education materials constructed specifically to contain materials on the appropriate preventive maneuvers under study. The binders were regularly updated.

Patient Mediated. Posters designed to prompt patients to ask about folic acid, flu vaccine, and mammography were offered to all sites. Thirteen sites implemented a patient consent form for prostate-specific antigen (PSA) testing. Eight sites received preventive care diaries for patients. Five sites had a prevention newsletter for patients. Four sites agreed to pilot a health risk appraisal software program.

Physician Feedback

At the end of the intervention the facilitator asked physicians about their experience, including what was most and least useful to them. Table 5 provides a summary of the content analysis of physician responses.

Audit and feedback, both initially and subsequently, comprised the component most frequently considered to be important in creating change. Almost as often, the preventive care flow sheet was identified as useful. The facilitator sessions designed to seek consensus on preventive care guidelines and strategies for implementation were also appreciated.

Several physicians did not agree with the evidence on PSA testing. Others did not feel that counseling for folic acid was a priority. Some found the patient education binder cumbersome, and others found the sticker system for tobacco counseling unwieldy. Thus, both were underused. Two physicians noted that the preventive care wall chart was not helpful.

Physician Self-Reported Practice Changes

Eighty-six percent (93/108) of the intervention and control physicians responded to a questionnaire at the end of the study. Due to sample size, statistical power was limited to detecting an absolute difference of approximately 0.30 between groups, assuming an alpha of 0.05 and 80% power.³³ Table 6 shows that 71% of intervention physicians compared with 28% of control physicians reported an audit of their practice for preventive services (P<.001). By the end of the study, 65% of the intervention physicians versus 48% of the control physicians indicated that they had a prevention policy or screening protocol in place, and 70% of intervention physicians compared with 58% of control physicians had created reminder systems for disease prevention.

Satisfaction with PF Intervention

At the telephone interview 6 months into the intervention, the mean satisfaction rating of intervention physicians was 4.08 on a scale of 1 (very dissatisfied) to 5 (very satisfied) with 80% satisfied with the intervention. At 17 months the mean satisfaction rating had risen to 4.5 with a 95% satisfaction rate.

At 6 months, 85% of the practices were satisfied with the frequency of visits of their assigned facilitator. At 17 months there was a 64% satisfaction rate, with the remaining 36% wanting more visits from the facilitator. The physicians commented on how the intervention had focused them on prevention in their practice. When the physicians were asked if they would agree to have a facilitator visit their practice in the future if given the opportunity, 90% agreed.

Concerns included not being able to continue the recall of patients at the end of the intervention and questioning the inappropriate maneuvers as too controversial. A physician from a large practice with 6 physicians commented that the facilitator could not easily work in the complex practice environment.

Discussion

Our study demonstrates that PFs can significantly improve the delivery of preventive services and in the process make quality contributions to a practice environment with high satisfaction rates from participating physicians.

Our intervention had a higher frequency and intensity of visits than other studies of this genre. The PFs had an average of almost 2 visits per month lasting approximately 105 minutes per visit. Dietrich and colleagues²¹ reported only 4 visits over a 3-month period lasting an average of 120 minutes, and Hulscher and coworkers²² reported approximately 25 visits with an average duration of 73 minutes. Others have been even less frequent,²⁰ and in other studies it is not reported.^34-36

The critical intervention components as evidenced by physician feedback, changes between control and intervention practices, and the amount of facilitator time spent on each component were: (1) audit and feedback, (2) sharing and discussing information to build consensus on an action plan, and (3) a reminder system. Similarly, the Cancer Prevention in Community Practice Project achieved 100% success in implementing change using customized preventive care flowsheets.³⁷ Of the 7 intervention components, patient education materials and patient-mediated interventions such as posters in the waiting room were considered to be the least useful.

Overall, physicians and nurses working within the practices were very satisfied with the intervention, and 90% were willing to have the nurse facilitator continue working with their practice.

Lessons learned from the process evaluation for improving the delivery of the outreach facilitation intervention include:

Focusing on the 3 key intervention components (audit and feedback, seeking consensus on an action plan, and implementing a reminder system) and tailoring these to the needs of the practice

Preparing patient education and patient-mediated materials only if the practice requests such materials

Developing simpler strategies to encourage physicians to counsel their patients that smoke to quit smoking

Providing the facilitators an administrative assistant to reduce the amount of their time spent on administrative duties for the practices and increase time on-site

Strengths

The strengths of the study include the completeness of the data set, the theoretical framework for data collection, the use of multiple data sources and data collection methods, and the prospective data collection methodology.

Limitations

There are several limitations to the process evaluation methods. Much of the data was provided by the facilitators themselves, and therefore the possibility of bias exists. The study population consisted of HSOs, and therefore the results may not be generalizable. There is a possibility of social desirability bias in the satisfaction rates. Finally, our analyses of the process data were descriptive and exploratory.

Conclusions

Process evaluation often identifies future areas of research. Follow-up of the few practices that were dissatisfied with facilitation should be carried out to understand why they were dissatisfied. Sustainability needs to be addressed. For example, Dietrich and colleagues³⁸ found that 5-year durability of a preventive services office system depended on the physician’s preventive care philosophy. McCowan and coworkers³⁹ found that the effect of a facilitator was not sustained for 2 years. Finally, to maximize cost-effectiveness, more research is required to determine how much of a dose of facilitation is required and how frequently facilitators should visit to achieve a positive outcome.

Acknowledgments

We wish to acknowledge the financial support of the Ontario Ministry of Health, as well as the substantial contributions of the 3 nurse facilitators (Ingrid LeClaire, Ann MacLeod, and Ruth Blochlinger). We also wish to thank the many physicians and nurses who participated in the study.

It is well established that nearly half of all Americans will die of cardiovascular disease. Lipid-lowering trials^1,2 using medications have resulted in 30% reductions in mortality and morbidity. Although effective, nearly 70% of the morbidity and mortality from coronary artery disease (CAD) occurs in patients receiving lipid-lowering therapy, despite highly significant 30% to 35% low-density lipoprotein (LDL) level reductions. Additional interventions beyond medication-induced LDL reductions appear warranted if our health care system is to further reduce the morbidity, mortality, and expenses associated with CAD. Nutritional choices have been shown to beneficially influence several CAD risk factors.³

Physicians need low-cost, practical, and effective dietary programs that patients with CAD are willing to follow. In particular, there is a need to explore simple dietary interventions that influence the pathophysiology behind CAD. Fortunately, there is growing interest in dietary intake that has been shown to decrease LDL oxidation^4-7 and to improve endothelial vasomotion. For example, it has been shown in patients in France with known CAD that simply switching polyunsaturated fat intake to largely oxidation-stable monounsaturated fat intake and n-3 fatty acid intake (omega-3 fats) reduces total mortality by 70% without reductions in total fat intake or changes in lipid profiles.^9,10 Other observational studies have supported the concept that the type of fat intake is more important than reducing total fat intake.^11,12

The objectives of the Dietary Intervention and Evaluation Trial (D.I.E.T.) were to add healthy foods to the diet (eg, legumes, fruits, and vegetables) and to change dietary fat intake from polyunsaturated and saturated fat to oxidation-resistant monounsaturated fat and n-3 fatty acid sources. In essence, we assessed the willingness of Americans with CAD to move toward a more Mediterranean-like diet. The subjects were counseled during group office visits. The mechanism for physicians to offer group visits as a billable service is reviewed elsewhere.¹³

Methods

Study Sample

In January 1997, patients with CAD were selected from the Heart Care Registry at Group Health Cooperative at 4 multispecialty clinics in 3 cities. Inclusion criteria were known CAD (based on hospital-generated diagnostic coding data and a subsequent chart review confirmation) and either LDL levels greater than 3.4 mmol per L (130 mg/dL) or patients without an LDL level recorded in the previous 18 months but with a total cholesterol/high-density lipoprotein (HDL) ratio greater than 5.5. We enrolled patients with high lipid levels in an attempt to choose patients at the greatest need for intervention. The Center for Health Studies contacted 234 patients by telephone and successfully recruited 132 with known CAD to participate in our study (56% willingness to enroll). Primary care physicians excluded 11 patients with terminal or end-stage medical problems who were not likely to survive the duration of the study. One patient who was following another dietary program (the Ornish Program) was excluded, resulting in 120 remaining subjects. Anticipating a 15% greater reduction in LDL levels and a 15% greater improvement in dietary intake in the experimental group than the control group, it was determined this sample size (a=0.05) would have a power of 80 (b=0.20). The subjects gave informed consent to participate in this randomized trial, signed consent forms, and received free monthly classes over the course of 1 year. There was no monetary compensation for their participation. The Human Subjects Committee of the Center for Health Studies and the University of Washington Research Committee approved this project.

After the recruitment and consent we chose specific days for fasting blood draws and offered a specific evening at each clinic when group visits would be offered. Twenty-three of the 120 patients could not attend the group visits and blood draw sessions as scheduled, and for personal or scheduling reasons withdrew from the study. The remaining 97 patients (29.9% women) were stratified according to a single entry LDL level and then assigned using an alternating table to create 2 groups with equal LDL levels. The 2 groups were then randomly assigned as experimental and control groups. After this stratification based on LDL levels and random group assignment, we compared the ages, total cholesterol/HDL ratios, hemoglobin (Hb) A1C levels, triglyceride levels, blood pressures, and body mass indexes of the 2 groups and found them to be similar at entry Table 1. During the 1-year study, 4 of the 49 patients in the experimental group and 3 of the 48 patients in the control group dropped out for scheduling or personal reasons before completing the study. Thus, 45 experimental and 45 control subjects completed the trial.

The mean LDL levels for the entire study population decreased from 3.7 mmol per L (142 mg/dL) in January 1997, when the subjects were identified to 3.1 mmol per L (118.5 mg/dL) in September 1997, when they were randomized into groups. This LDL reduction was presumably because of a health maintenance organization–directed campaign to lower LDL levels in this heart care population with lipid-lowering medications. A single medication (simvastatin) accounted for 89% of the lipid-lowering medication used by these subjects. Despite prestudy LDL reductions, the majority of the patients recruited for the D.I.E.T intervention had not yet achieved a 35% LDL reduction.

Intervention

The experimental group met for 14 90-minute group visits over 1 year: weekly for the first month and then monthly. Classes taught by a licensed practical nurse highlighted an antioxidant-rich diet with a maximum of 20% of calories from fat, and encouraged the use of monounsaturated and n-3 fatty acid types of fat in lieu of saturated and polyunsaturated fats. The intervention patients also received a textbook (The 28-Day Antioxidant Diet Program) that included information for shopping lists, menu plans, and food-monitoring sheets. Additional recipes were added at the group’s request and were reviewed during the lectures. Cooking demonstrations were performed. A gradual increase in physical activities, such as walking, was also encouraged. Significant others were strongly encouraged to participate in these classes.

Specific intervention goals were aimed at increasing fruit and vegetable intake to 7 or more servings per day, adding garlic and antioxidant-rich herbs and 1 serving of a legume or soy product daily. The program emphasized choosing low-glycemic carbohydrate sources. Also, the intervention had a goal for each subject to exercise for 30 to 45 minutes 5 to 6 days per week and to reach a target heart rate 60% to 70% of their maximum predicted rate. Within the capitated health system where this intervention occurred, the cost for materials and personnel to continue these types of group visit sessions for CAD patients on an ongoing basis would be $7 (United States) per member per month (PMPM).

Patients in the control group were given written information that included a handout to follow the National Cholesterol Education Program’s Step II-III diet, a handout on choosing dietary fats, information on increasing produce and whole grain intake, and American Heart Association sample meal plans. The control group did not meet in group visits but continued to receive care as usual from their providers.

Members of both groups had their fasting lipids levels and Hb A1C levels drawn and compared at 0 and 12 months. All laboratory results were forwarded to the patients’ primary care physicians, who were strongly encouraged to treat elevated lipid levels with lifestyle-change recommendations and lipid-lowering medications until LDL levels reached a specific target: a 30% to 35% reduction in LDL from their pre-event LDL, or if the pre-event LDL was unavailable, a final LDL less than 2.6 mmol per L (100 mg/dL) or an LDL less than 3.4 mmol per L (130 mg/dL) and with a total cholesterol/HDL ratio less than 4.0.

Data Collection

Blood samples were collected at each clinic, and fasting lipid and Hb A1C samples were obtained. Thirty-day food frequency questionnaires from the Fred Hutchinson Women’s Health Initiative (WHI) were completed by study patients at 0, 3, and 12 months. An additional food intake questionnaire was given to assess both legume intake in servings per week and the type of fat used for cooking and baking.

PMPM expense data were obtained through patient-specific billing and utilization data. All of the subjects in our study were participants in a 100% capitated health care system. All office visits, pharmacy expenses, emergency department visits, and hospital admissions and procedures were tabulated. The expense data were categorized as total expenses, in-patient hospital care, pharmacy, and outpatient care. The $7 estimated to fund this program on an ongoing basis was added to the PMPM expenses of the experimental group.

Statistical Analysis

Differences between the experimental and control groups for fasting blood levels and PMPM expense data were assessed using a Student t test, both at entry and after 12 months. No statistical differences were noted between the control and experimental groups at entry. At entry and at 12 months we analyzed food intake questionnaires and laboratory results using an independent samples t test. Statistical analysis was performed using SAS software (SAS Institute, Inc, Cary, NC).

Results

Food Intake Questionnaires

Table 2 shows the mean fruit and vegetable intake for patients in the experimental group and those in the control group at entry and after 12 months. The experimental group patients increased their vegetable and fruit intake significantly compared with the control group over 12 months (P = .0001 and .0072, respectively).

The experimental group reduced their total fat intake and their saturated fat intake after 12 months; however, these differences in change were not significant (P=.4045, P=.1049).

Separate from the WHI food intake questionnaire, subjects were asked to report their weekly intake of legumes (a serving size was equal to 0.5 cup legumes) and the type of fat used for cooking and baking. The experimental group also reported a significant 45% increase in use of monounsaturated cooking oils compared with the control group’s 1% increase (P=.0001).

Fasting Blood Levels

Using a paired comparison t test, the experimental group noted a significant reduction in LDL levels, 117 mg per dL at entry to 104 mg per dL at 12 months (P=.0035,) while the control group’s LDL reduction was not significant, 119 mg per dL at entry and 111.7 mg per dL at 12 months (P=.1475). The difference in LDL reductions between the 2 groups was not significant by an independent samples t test. The total cholesterol/HDL ratio, Hb A1C, and triglyceride levels decreased for both groups; HDL increased for both groups, but the difference between the changes was not statistically significant.*

PMPM Expenses

Both groups noted reductions in PMPM total and in-patient expenses. The total PMPM expenses decreased 38% for the experimental group and 10% for the control group. No statistically significant differences were found between the groups’ total PMPM expenses (P=.2975).†

After the 1-year intervention, there was no difference in overall pharmacy PMPM expenses between the 2 groups (P=.4578). Specifically, lipid-lowering medication use and expense was very similar in the 2 groups before and after the intervention.

Discussion

In our small group of 97 patients with known CAD and elevated lipid levels, this intervention was not powered to yield significant improvements in clinical outcomes. Our study was associated with increased fruit and vegetable intake (nearly 2 more servings per day), a small increase in legume intake, and a switch to oxidation-resistant monounsaturated cooking oils.

Our study does help to confirm that half of patients with CAD are willing to make significant lifestyle changes when offered a program that emphasizes adding healthy foods in a group visit format. We also targeted patients with known CAD and elevated lipid levels and demonstrated that patients in the greatest need of therapy were willing to try lifestyle changes.

The experimental group noted significant reductions in LDL levels, but they were not statistically greater reductions than the control group. A bigger trend toward improvements in lipid and Hb A1C levels and reductions in total and saturated fat intake occurred in the experimental group than in the control group, yet these differences were not statistically significant. A 10% dropout rate was anticipated over the study; however, the additional 19% dropout rate that occurred after recruitment and before randomization was unexpected and limited the power to assess some of the trends we noted. The difference in LDL reductions between groups would have reached statistical significance if a limited reduction in LDL levels had occurred in the control group (<2%), or if the study had achieved the original target size planned and the changes noted persisted in the missing subjects.

The lipid reductions that occurred during our study may seem limited. However, this study reflects the ongoing implementation of various strategies over several years to improve lipid levels in this cohort that initially had LDL levels well above their target. In the 9-month time interval between identifying the patients and starting the intervention, LDL levels decreased by 16.5% in both groups, from a mean of 3.7 mmol per L (142 mg/dL) to 3.1 mmol per L (118.5 mg/dL).

PMPM statistical data is difficult to evaluate because of the well-known high variability in the cost of providing care to high-risk patients. No statistically significant differences were noted in total PMPM expenses between the groups in our study. The Cooperative Health Care Clinic Study²⁰ supports the idea that group visit interventions may reduce health care expenses while improving clinical outcomes. That study in Colorado was conducted using seniors with multiple medical problems and noted reduced PMPM expenses, as well as enhanced patient satisfaction, improved immunization rates, and reduced hospital admission rates.

Limitations

A limitation of this and many lifestyle intervention studies that are taught on the group level is that we cannot distinguish between the direct benefits of the lifestyle interventions and the indirect benefits of meeting within a group. A group visit itself is an intervention that may provide clinical benefits. These attributes include group support and improved adherence to lifestyle changes. More studies are needed to clarify the direct benefits of combining cohorts of patients with specific illnesses with the same intervention taught on an individual and a group visit level.

Additional limitations to our study are that the dietary results were self-reported and that all study participants would be expected to show a healthy participant effect. Although the control group did show a reduction in lipid profiles during our study, that group noted only a 3% decrease in total fat intake, no change in saturated fat intake, and a reduction in their vegetable, fruit, and legume intake. Thus, no significant dietary improvements were noted in the control group.

Conclusions

Patients with known CAD who are already being treated with lipid-lowering medication are willing to make dietary changes that are taught during group visits. More than 50% of inadequately controlled patients with known CAD who were offered our program were willing to enroll, and we achieved significant improvements in these patients in increased fruit and vegetable intake, legume intake, and in changing the type of fat use for cooking. In larger studies these improvements may prove to be associated with reductions in total health care expenses and in clinical events. Further studies are needed to test this type of group visit program with other patient populations in larger clinical settings.

Related Resources

• American Diabetes Association www.diabetes.org
• American Heart Association www.americanheart.org

Acknowledgments

This study was funded by the South Region Executive Committee at Group Health Cooperative of Puget Sound. The patient data registry was provided by Group Health’s Heart Care Team. Patient recruitment and sample size calculations were provided by Group Health’s Center for Health Studies. Fred Hutchinson’s Cancer Research Center provided food frequency questionnaires and performed the associated data analysis. The Geriatric Research Team at Morton Plant Mease Health Care in Clearwater, Florida, performed the remaining statistical data analysis.

It is well established that nearly half of all Americans will die of cardiovascular disease. Lipid-lowering trials^1,2 using medications have resulted in 30% reductions in mortality and morbidity. Although effective, nearly 70% of the morbidity and mortality from coronary artery disease (CAD) occurs in patients receiving lipid-lowering therapy, despite highly significant 30% to 35% low-density lipoprotein (LDL) level reductions. Additional interventions beyond medication-induced LDL reductions appear warranted if our health care system is to further reduce the morbidity, mortality, and expenses associated with CAD. Nutritional choices have been shown to beneficially influence several CAD risk factors.³

Physicians need low-cost, practical, and effective dietary programs that patients with CAD are willing to follow. In particular, there is a need to explore simple dietary interventions that influence the pathophysiology behind CAD. Fortunately, there is growing interest in dietary intake that has been shown to decrease LDL oxidation^4-7 and to improve endothelial vasomotion. For example, it has been shown in patients in France with known CAD that simply switching polyunsaturated fat intake to largely oxidation-stable monounsaturated fat intake and n-3 fatty acid intake (omega-3 fats) reduces total mortality by 70% without reductions in total fat intake or changes in lipid profiles.^9,10 Other observational studies have supported the concept that the type of fat intake is more important than reducing total fat intake.^11,12

The objectives of the Dietary Intervention and Evaluation Trial (D.I.E.T.) were to add healthy foods to the diet (eg, legumes, fruits, and vegetables) and to change dietary fat intake from polyunsaturated and saturated fat to oxidation-resistant monounsaturated fat and n-3 fatty acid sources. In essence, we assessed the willingness of Americans with CAD to move toward a more Mediterranean-like diet. The subjects were counseled during group office visits. The mechanism for physicians to offer group visits as a billable service is reviewed elsewhere.¹³

Methods

Study Sample

In January 1997, patients with CAD were selected from the Heart Care Registry at Group Health Cooperative at 4 multispecialty clinics in 3 cities. Inclusion criteria were known CAD (based on hospital-generated diagnostic coding data and a subsequent chart review confirmation) and either LDL levels greater than 3.4 mmol per L (130 mg/dL) or patients without an LDL level recorded in the previous 18 months but with a total cholesterol/high-density lipoprotein (HDL) ratio greater than 5.5. We enrolled patients with high lipid levels in an attempt to choose patients at the greatest need for intervention. The Center for Health Studies contacted 234 patients by telephone and successfully recruited 132 with known CAD to participate in our study (56% willingness to enroll). Primary care physicians excluded 11 patients with terminal or end-stage medical problems who were not likely to survive the duration of the study. One patient who was following another dietary program (the Ornish Program) was excluded, resulting in 120 remaining subjects. Anticipating a 15% greater reduction in LDL levels and a 15% greater improvement in dietary intake in the experimental group than the control group, it was determined this sample size (a=0.05) would have a power of 80 (b=0.20). The subjects gave informed consent to participate in this randomized trial, signed consent forms, and received free monthly classes over the course of 1 year. There was no monetary compensation for their participation. The Human Subjects Committee of the Center for Health Studies and the University of Washington Research Committee approved this project.

After the recruitment and consent we chose specific days for fasting blood draws and offered a specific evening at each clinic when group visits would be offered. Twenty-three of the 120 patients could not attend the group visits and blood draw sessions as scheduled, and for personal or scheduling reasons withdrew from the study. The remaining 97 patients (29.9% women) were stratified according to a single entry LDL level and then assigned using an alternating table to create 2 groups with equal LDL levels. The 2 groups were then randomly assigned as experimental and control groups. After this stratification based on LDL levels and random group assignment, we compared the ages, total cholesterol/HDL ratios, hemoglobin (Hb) A1C levels, triglyceride levels, blood pressures, and body mass indexes of the 2 groups and found them to be similar at entry Table 1. During the 1-year study, 4 of the 49 patients in the experimental group and 3 of the 48 patients in the control group dropped out for scheduling or personal reasons before completing the study. Thus, 45 experimental and 45 control subjects completed the trial.

The mean LDL levels for the entire study population decreased from 3.7 mmol per L (142 mg/dL) in January 1997, when the subjects were identified to 3.1 mmol per L (118.5 mg/dL) in September 1997, when they were randomized into groups. This LDL reduction was presumably because of a health maintenance organization–directed campaign to lower LDL levels in this heart care population with lipid-lowering medications. A single medication (simvastatin) accounted for 89% of the lipid-lowering medication used by these subjects. Despite prestudy LDL reductions, the majority of the patients recruited for the D.I.E.T intervention had not yet achieved a 35% LDL reduction.

Intervention

The experimental group met for 14 90-minute group visits over 1 year: weekly for the first month and then monthly. Classes taught by a licensed practical nurse highlighted an antioxidant-rich diet with a maximum of 20% of calories from fat, and encouraged the use of monounsaturated and n-3 fatty acid types of fat in lieu of saturated and polyunsaturated fats. The intervention patients also received a textbook (The 28-Day Antioxidant Diet Program) that included information for shopping lists, menu plans, and food-monitoring sheets. Additional recipes were added at the group’s request and were reviewed during the lectures. Cooking demonstrations were performed. A gradual increase in physical activities, such as walking, was also encouraged. Significant others were strongly encouraged to participate in these classes.

Specific intervention goals were aimed at increasing fruit and vegetable intake to 7 or more servings per day, adding garlic and antioxidant-rich herbs and 1 serving of a legume or soy product daily. The program emphasized choosing low-glycemic carbohydrate sources. Also, the intervention had a goal for each subject to exercise for 30 to 45 minutes 5 to 6 days per week and to reach a target heart rate 60% to 70% of their maximum predicted rate. Within the capitated health system where this intervention occurred, the cost for materials and personnel to continue these types of group visit sessions for CAD patients on an ongoing basis would be $7 (United States) per member per month (PMPM).

Patients in the control group were given written information that included a handout to follow the National Cholesterol Education Program’s Step II-III diet, a handout on choosing dietary fats, information on increasing produce and whole grain intake, and American Heart Association sample meal plans. The control group did not meet in group visits but continued to receive care as usual from their providers.

Members of both groups had their fasting lipids levels and Hb A1C levels drawn and compared at 0 and 12 months. All laboratory results were forwarded to the patients’ primary care physicians, who were strongly encouraged to treat elevated lipid levels with lifestyle-change recommendations and lipid-lowering medications until LDL levels reached a specific target: a 30% to 35% reduction in LDL from their pre-event LDL, or if the pre-event LDL was unavailable, a final LDL less than 2.6 mmol per L (100 mg/dL) or an LDL less than 3.4 mmol per L (130 mg/dL) and with a total cholesterol/HDL ratio less than 4.0.

Data Collection

Blood samples were collected at each clinic, and fasting lipid and Hb A1C samples were obtained. Thirty-day food frequency questionnaires from the Fred Hutchinson Women’s Health Initiative (WHI) were completed by study patients at 0, 3, and 12 months. An additional food intake questionnaire was given to assess both legume intake in servings per week and the type of fat used for cooking and baking.

PMPM expense data were obtained through patient-specific billing and utilization data. All of the subjects in our study were participants in a 100% capitated health care system. All office visits, pharmacy expenses, emergency department visits, and hospital admissions and procedures were tabulated. The expense data were categorized as total expenses, in-patient hospital care, pharmacy, and outpatient care. The $7 estimated to fund this program on an ongoing basis was added to the PMPM expenses of the experimental group.

Statistical Analysis

Differences between the experimental and control groups for fasting blood levels and PMPM expense data were assessed using a Student t test, both at entry and after 12 months. No statistical differences were noted between the control and experimental groups at entry. At entry and at 12 months we analyzed food intake questionnaires and laboratory results using an independent samples t test. Statistical analysis was performed using SAS software (SAS Institute, Inc, Cary, NC).

Results

Food Intake Questionnaires

Table 2 shows the mean fruit and vegetable intake for patients in the experimental group and those in the control group at entry and after 12 months. The experimental group patients increased their vegetable and fruit intake significantly compared with the control group over 12 months (P = .0001 and .0072, respectively).

The experimental group reduced their total fat intake and their saturated fat intake after 12 months; however, these differences in change were not significant (P=.4045, P=.1049).

Separate from the WHI food intake questionnaire, subjects were asked to report their weekly intake of legumes (a serving size was equal to 0.5 cup legumes) and the type of fat used for cooking and baking. The experimental group also reported a significant 45% increase in use of monounsaturated cooking oils compared with the control group’s 1% increase (P=.0001).

Fasting Blood Levels

Using a paired comparison t test, the experimental group noted a significant reduction in LDL levels, 117 mg per dL at entry to 104 mg per dL at 12 months (P=.0035,) while the control group’s LDL reduction was not significant, 119 mg per dL at entry and 111.7 mg per dL at 12 months (P=.1475). The difference in LDL reductions between the 2 groups was not significant by an independent samples t test. The total cholesterol/HDL ratio, Hb A1C, and triglyceride levels decreased for both groups; HDL increased for both groups, but the difference between the changes was not statistically significant.*

PMPM Expenses

Both groups noted reductions in PMPM total and in-patient expenses. The total PMPM expenses decreased 38% for the experimental group and 10% for the control group. No statistically significant differences were found between the groups’ total PMPM expenses (P=.2975).†

After the 1-year intervention, there was no difference in overall pharmacy PMPM expenses between the 2 groups (P=.4578). Specifically, lipid-lowering medication use and expense was very similar in the 2 groups before and after the intervention.

Discussion

In our small group of 97 patients with known CAD and elevated lipid levels, this intervention was not powered to yield significant improvements in clinical outcomes. Our study was associated with increased fruit and vegetable intake (nearly 2 more servings per day), a small increase in legume intake, and a switch to oxidation-resistant monounsaturated cooking oils.

Our study does help to confirm that half of patients with CAD are willing to make significant lifestyle changes when offered a program that emphasizes adding healthy foods in a group visit format. We also targeted patients with known CAD and elevated lipid levels and demonstrated that patients in the greatest need of therapy were willing to try lifestyle changes.

The experimental group noted significant reductions in LDL levels, but they were not statistically greater reductions than the control group. A bigger trend toward improvements in lipid and Hb A1C levels and reductions in total and saturated fat intake occurred in the experimental group than in the control group, yet these differences were not statistically significant. A 10% dropout rate was anticipated over the study; however, the additional 19% dropout rate that occurred after recruitment and before randomization was unexpected and limited the power to assess some of the trends we noted. The difference in LDL reductions between groups would have reached statistical significance if a limited reduction in LDL levels had occurred in the control group (<2%), or if the study had achieved the original target size planned and the changes noted persisted in the missing subjects.

The lipid reductions that occurred during our study may seem limited. However, this study reflects the ongoing implementation of various strategies over several years to improve lipid levels in this cohort that initially had LDL levels well above their target. In the 9-month time interval between identifying the patients and starting the intervention, LDL levels decreased by 16.5% in both groups, from a mean of 3.7 mmol per L (142 mg/dL) to 3.1 mmol per L (118.5 mg/dL).

PMPM statistical data is difficult to evaluate because of the well-known high variability in the cost of providing care to high-risk patients. No statistically significant differences were noted in total PMPM expenses between the groups in our study. The Cooperative Health Care Clinic Study²⁰ supports the idea that group visit interventions may reduce health care expenses while improving clinical outcomes. That study in Colorado was conducted using seniors with multiple medical problems and noted reduced PMPM expenses, as well as enhanced patient satisfaction, improved immunization rates, and reduced hospital admission rates.

Limitations

A limitation of this and many lifestyle intervention studies that are taught on the group level is that we cannot distinguish between the direct benefits of the lifestyle interventions and the indirect benefits of meeting within a group. A group visit itself is an intervention that may provide clinical benefits. These attributes include group support and improved adherence to lifestyle changes. More studies are needed to clarify the direct benefits of combining cohorts of patients with specific illnesses with the same intervention taught on an individual and a group visit level.

Additional limitations to our study are that the dietary results were self-reported and that all study participants would be expected to show a healthy participant effect. Although the control group did show a reduction in lipid profiles during our study, that group noted only a 3% decrease in total fat intake, no change in saturated fat intake, and a reduction in their vegetable, fruit, and legume intake. Thus, no significant dietary improvements were noted in the control group.

Conclusions

Patients with known CAD who are already being treated with lipid-lowering medication are willing to make dietary changes that are taught during group visits. More than 50% of inadequately controlled patients with known CAD who were offered our program were willing to enroll, and we achieved significant improvements in these patients in increased fruit and vegetable intake, legume intake, and in changing the type of fat use for cooking. In larger studies these improvements may prove to be associated with reductions in total health care expenses and in clinical events. Further studies are needed to test this type of group visit program with other patient populations in larger clinical settings.

Related Resources

• American Diabetes Association www.diabetes.org
• American Heart Association www.americanheart.org

Acknowledgments

This study was funded by the South Region Executive Committee at Group Health Cooperative of Puget Sound. The patient data registry was provided by Group Health’s Heart Care Team. Patient recruitment and sample size calculations were provided by Group Health’s Center for Health Studies. Fred Hutchinson’s Cancer Research Center provided food frequency questionnaires and performed the associated data analysis. The Geriatric Research Team at Morton Plant Mease Health Care in Clearwater, Florida, performed the remaining statistical data analysis.

It is well established that nearly half of all Americans will die of cardiovascular disease. Lipid-lowering trials^1,2 using medications have resulted in 30% reductions in mortality and morbidity. Although effective, nearly 70% of the morbidity and mortality from coronary artery disease (CAD) occurs in patients receiving lipid-lowering therapy, despite highly significant 30% to 35% low-density lipoprotein (LDL) level reductions. Additional interventions beyond medication-induced LDL reductions appear warranted if our health care system is to further reduce the morbidity, mortality, and expenses associated with CAD. Nutritional choices have been shown to beneficially influence several CAD risk factors.³

Physicians need low-cost, practical, and effective dietary programs that patients with CAD are willing to follow. In particular, there is a need to explore simple dietary interventions that influence the pathophysiology behind CAD. Fortunately, there is growing interest in dietary intake that has been shown to decrease LDL oxidation^4-7 and to improve endothelial vasomotion. For example, it has been shown in patients in France with known CAD that simply switching polyunsaturated fat intake to largely oxidation-stable monounsaturated fat intake and n-3 fatty acid intake (omega-3 fats) reduces total mortality by 70% without reductions in total fat intake or changes in lipid profiles.^9,10 Other observational studies have supported the concept that the type of fat intake is more important than reducing total fat intake.^11,12

The objectives of the Dietary Intervention and Evaluation Trial (D.I.E.T.) were to add healthy foods to the diet (eg, legumes, fruits, and vegetables) and to change dietary fat intake from polyunsaturated and saturated fat to oxidation-resistant monounsaturated fat and n-3 fatty acid sources. In essence, we assessed the willingness of Americans with CAD to move toward a more Mediterranean-like diet. The subjects were counseled during group office visits. The mechanism for physicians to offer group visits as a billable service is reviewed elsewhere.¹³

Methods

Study Sample

In January 1997, patients with CAD were selected from the Heart Care Registry at Group Health Cooperative at 4 multispecialty clinics in 3 cities. Inclusion criteria were known CAD (based on hospital-generated diagnostic coding data and a subsequent chart review confirmation) and either LDL levels greater than 3.4 mmol per L (130 mg/dL) or patients without an LDL level recorded in the previous 18 months but with a total cholesterol/high-density lipoprotein (HDL) ratio greater than 5.5. We enrolled patients with high lipid levels in an attempt to choose patients at the greatest need for intervention. The Center for Health Studies contacted 234 patients by telephone and successfully recruited 132 with known CAD to participate in our study (56% willingness to enroll). Primary care physicians excluded 11 patients with terminal or end-stage medical problems who were not likely to survive the duration of the study. One patient who was following another dietary program (the Ornish Program) was excluded, resulting in 120 remaining subjects. Anticipating a 15% greater reduction in LDL levels and a 15% greater improvement in dietary intake in the experimental group than the control group, it was determined this sample size (a=0.05) would have a power of 80 (b=0.20). The subjects gave informed consent to participate in this randomized trial, signed consent forms, and received free monthly classes over the course of 1 year. There was no monetary compensation for their participation. The Human Subjects Committee of the Center for Health Studies and the University of Washington Research Committee approved this project.

After the recruitment and consent we chose specific days for fasting blood draws and offered a specific evening at each clinic when group visits would be offered. Twenty-three of the 120 patients could not attend the group visits and blood draw sessions as scheduled, and for personal or scheduling reasons withdrew from the study. The remaining 97 patients (29.9% women) were stratified according to a single entry LDL level and then assigned using an alternating table to create 2 groups with equal LDL levels. The 2 groups were then randomly assigned as experimental and control groups. After this stratification based on LDL levels and random group assignment, we compared the ages, total cholesterol/HDL ratios, hemoglobin (Hb) A1C levels, triglyceride levels, blood pressures, and body mass indexes of the 2 groups and found them to be similar at entry Table 1. During the 1-year study, 4 of the 49 patients in the experimental group and 3 of the 48 patients in the control group dropped out for scheduling or personal reasons before completing the study. Thus, 45 experimental and 45 control subjects completed the trial.

The mean LDL levels for the entire study population decreased from 3.7 mmol per L (142 mg/dL) in January 1997, when the subjects were identified to 3.1 mmol per L (118.5 mg/dL) in September 1997, when they were randomized into groups. This LDL reduction was presumably because of a health maintenance organization–directed campaign to lower LDL levels in this heart care population with lipid-lowering medications. A single medication (simvastatin) accounted for 89% of the lipid-lowering medication used by these subjects. Despite prestudy LDL reductions, the majority of the patients recruited for the D.I.E.T intervention had not yet achieved a 35% LDL reduction.

Intervention

The experimental group met for 14 90-minute group visits over 1 year: weekly for the first month and then monthly. Classes taught by a licensed practical nurse highlighted an antioxidant-rich diet with a maximum of 20% of calories from fat, and encouraged the use of monounsaturated and n-3 fatty acid types of fat in lieu of saturated and polyunsaturated fats. The intervention patients also received a textbook (The 28-Day Antioxidant Diet Program) that included information for shopping lists, menu plans, and food-monitoring sheets. Additional recipes were added at the group’s request and were reviewed during the lectures. Cooking demonstrations were performed. A gradual increase in physical activities, such as walking, was also encouraged. Significant others were strongly encouraged to participate in these classes.

Specific intervention goals were aimed at increasing fruit and vegetable intake to 7 or more servings per day, adding garlic and antioxidant-rich herbs and 1 serving of a legume or soy product daily. The program emphasized choosing low-glycemic carbohydrate sources. Also, the intervention had a goal for each subject to exercise for 30 to 45 minutes 5 to 6 days per week and to reach a target heart rate 60% to 70% of their maximum predicted rate. Within the capitated health system where this intervention occurred, the cost for materials and personnel to continue these types of group visit sessions for CAD patients on an ongoing basis would be $7 (United States) per member per month (PMPM).

Patients in the control group were given written information that included a handout to follow the National Cholesterol Education Program’s Step II-III diet, a handout on choosing dietary fats, information on increasing produce and whole grain intake, and American Heart Association sample meal plans. The control group did not meet in group visits but continued to receive care as usual from their providers.

Members of both groups had their fasting lipids levels and Hb A1C levels drawn and compared at 0 and 12 months. All laboratory results were forwarded to the patients’ primary care physicians, who were strongly encouraged to treat elevated lipid levels with lifestyle-change recommendations and lipid-lowering medications until LDL levels reached a specific target: a 30% to 35% reduction in LDL from their pre-event LDL, or if the pre-event LDL was unavailable, a final LDL less than 2.6 mmol per L (100 mg/dL) or an LDL less than 3.4 mmol per L (130 mg/dL) and with a total cholesterol/HDL ratio less than 4.0.

Data Collection

Blood samples were collected at each clinic, and fasting lipid and Hb A1C samples were obtained. Thirty-day food frequency questionnaires from the Fred Hutchinson Women’s Health Initiative (WHI) were completed by study patients at 0, 3, and 12 months. An additional food intake questionnaire was given to assess both legume intake in servings per week and the type of fat used for cooking and baking.

PMPM expense data were obtained through patient-specific billing and utilization data. All of the subjects in our study were participants in a 100% capitated health care system. All office visits, pharmacy expenses, emergency department visits, and hospital admissions and procedures were tabulated. The expense data were categorized as total expenses, in-patient hospital care, pharmacy, and outpatient care. The $7 estimated to fund this program on an ongoing basis was added to the PMPM expenses of the experimental group.

Statistical Analysis

Differences between the experimental and control groups for fasting blood levels and PMPM expense data were assessed using a Student t test, both at entry and after 12 months. No statistical differences were noted between the control and experimental groups at entry. At entry and at 12 months we analyzed food intake questionnaires and laboratory results using an independent samples t test. Statistical analysis was performed using SAS software (SAS Institute, Inc, Cary, NC).

Results

Food Intake Questionnaires

Table 2 shows the mean fruit and vegetable intake for patients in the experimental group and those in the control group at entry and after 12 months. The experimental group patients increased their vegetable and fruit intake significantly compared with the control group over 12 months (P = .0001 and .0072, respectively).

The experimental group reduced their total fat intake and their saturated fat intake after 12 months; however, these differences in change were not significant (P=.4045, P=.1049).

Separate from the WHI food intake questionnaire, subjects were asked to report their weekly intake of legumes (a serving size was equal to 0.5 cup legumes) and the type of fat used for cooking and baking. The experimental group also reported a significant 45% increase in use of monounsaturated cooking oils compared with the control group’s 1% increase (P=.0001).

Fasting Blood Levels

Using a paired comparison t test, the experimental group noted a significant reduction in LDL levels, 117 mg per dL at entry to 104 mg per dL at 12 months (P=.0035,) while the control group’s LDL reduction was not significant, 119 mg per dL at entry and 111.7 mg per dL at 12 months (P=.1475). The difference in LDL reductions between the 2 groups was not significant by an independent samples t test. The total cholesterol/HDL ratio, Hb A1C, and triglyceride levels decreased for both groups; HDL increased for both groups, but the difference between the changes was not statistically significant.*

PMPM Expenses

Both groups noted reductions in PMPM total and in-patient expenses. The total PMPM expenses decreased 38% for the experimental group and 10% for the control group. No statistically significant differences were found between the groups’ total PMPM expenses (P=.2975).†

After the 1-year intervention, there was no difference in overall pharmacy PMPM expenses between the 2 groups (P=.4578). Specifically, lipid-lowering medication use and expense was very similar in the 2 groups before and after the intervention.

Discussion

In our small group of 97 patients with known CAD and elevated lipid levels, this intervention was not powered to yield significant improvements in clinical outcomes. Our study was associated with increased fruit and vegetable intake (nearly 2 more servings per day), a small increase in legume intake, and a switch to oxidation-resistant monounsaturated cooking oils.

Our study does help to confirm that half of patients with CAD are willing to make significant lifestyle changes when offered a program that emphasizes adding healthy foods in a group visit format. We also targeted patients with known CAD and elevated lipid levels and demonstrated that patients in the greatest need of therapy were willing to try lifestyle changes.

The experimental group noted significant reductions in LDL levels, but they were not statistically greater reductions than the control group. A bigger trend toward improvements in lipid and Hb A1C levels and reductions in total and saturated fat intake occurred in the experimental group than in the control group, yet these differences were not statistically significant. A 10% dropout rate was anticipated over the study; however, the additional 19% dropout rate that occurred after recruitment and before randomization was unexpected and limited the power to assess some of the trends we noted. The difference in LDL reductions between groups would have reached statistical significance if a limited reduction in LDL levels had occurred in the control group (<2%), or if the study had achieved the original target size planned and the changes noted persisted in the missing subjects.

The lipid reductions that occurred during our study may seem limited. However, this study reflects the ongoing implementation of various strategies over several years to improve lipid levels in this cohort that initially had LDL levels well above their target. In the 9-month time interval between identifying the patients and starting the intervention, LDL levels decreased by 16.5% in both groups, from a mean of 3.7 mmol per L (142 mg/dL) to 3.1 mmol per L (118.5 mg/dL).

PMPM statistical data is difficult to evaluate because of the well-known high variability in the cost of providing care to high-risk patients. No statistically significant differences were noted in total PMPM expenses between the groups in our study. The Cooperative Health Care Clinic Study²⁰ supports the idea that group visit interventions may reduce health care expenses while improving clinical outcomes. That study in Colorado was conducted using seniors with multiple medical problems and noted reduced PMPM expenses, as well as enhanced patient satisfaction, improved immunization rates, and reduced hospital admission rates.

Limitations

A limitation of this and many lifestyle intervention studies that are taught on the group level is that we cannot distinguish between the direct benefits of the lifestyle interventions and the indirect benefits of meeting within a group. A group visit itself is an intervention that may provide clinical benefits. These attributes include group support and improved adherence to lifestyle changes. More studies are needed to clarify the direct benefits of combining cohorts of patients with specific illnesses with the same intervention taught on an individual and a group visit level.

Additional limitations to our study are that the dietary results were self-reported and that all study participants would be expected to show a healthy participant effect. Although the control group did show a reduction in lipid profiles during our study, that group noted only a 3% decrease in total fat intake, no change in saturated fat intake, and a reduction in their vegetable, fruit, and legume intake. Thus, no significant dietary improvements were noted in the control group.

Conclusions

Patients with known CAD who are already being treated with lipid-lowering medication are willing to make dietary changes that are taught during group visits. More than 50% of inadequately controlled patients with known CAD who were offered our program were willing to enroll, and we achieved significant improvements in these patients in increased fruit and vegetable intake, legume intake, and in changing the type of fat use for cooking. In larger studies these improvements may prove to be associated with reductions in total health care expenses and in clinical events. Further studies are needed to test this type of group visit program with other patient populations in larger clinical settings.

Related Resources

• American Diabetes Association www.diabetes.org
• American Heart Association www.americanheart.org

Acknowledgments

This study was funded by the South Region Executive Committee at Group Health Cooperative of Puget Sound. The patient data registry was provided by Group Health’s Heart Care Team. Patient recruitment and sample size calculations were provided by Group Health’s Center for Health Studies. Fred Hutchinson’s Cancer Research Center provided food frequency questionnaires and performed the associated data analysis. The Geriatric Research Team at Morton Plant Mease Health Care in Clearwater, Florida, performed the remaining statistical data analysis.

Prostate cancer (the most common cancer in men) has a considerable impact on the quality of patients’ lives. After a frightening diagnosis, the treatments cause a variety of unpleasant side effects. In the case of metastatic disease, hormonal treatment through either orchiectomy or a permanent regimen of medications to suppress testosterone causes nausea, hot flashes, loss of muscle tone, and erectile dysfunction.¹ The complex psychosocial effects of diagnosis and treatment, such as changes in body and self-image, masculinity, sexuality, and uncertainty, are gradually recognized.² However, most investigations of quality of life have focused on the patients, increasing our understanding of how men respond to prostate cancer, but telling us little about how this disease affects the shared lives of men and their significant others. Prostate cancer threatens men’s survival and families’ futures. Sexual dysfunction following treatment has implications for both men and their intimate partners. Therefore, we explored the perceptions of men who have been treated for prostate cancer, and those of their wives, regarding the changes that were caused by prostate cancer and its treatment.

The effect of illness on marital quality is largely an unresolved area.³ Some researchers argue that there is no change—or if there is a change, it is a positive association between illness and the spousal relationship, because it brings spouses closer together.⁴ Others point to negative impacts on the marital relationship.⁵ It has been suggested that a worsening of marital quality can be explained by the financial implications of illness, the problematic behavior of the sick person, and the changed division of labor in the couple, including the loss of shared activities.⁵ A decline in marital quality is often perceived by spouses who suffer from increased stress and depression because of their spouses’ disease.^5,6 Research on couples coping with illness has highlighted the importance of open communication for positive adjustment by patients and their spouses.^7,8 But couples’ ability to communicate varies,⁸ and many avoid truthful and open communication with each other⁹ and with other friends and family members.¹⁰

Only a few prostate cancer studies have considered the patient’s social context. Studies that include spouses point to the interrelationship of their reactions to prostate cancer.^3,11-17 We focussed on couples’ communication about prostate cancer and its treatment effects.

Methods

Our study is part of a larger investigation of the psychosocial effects of metastatic prostate cancer, that had the development of quality-of-life scales as its goal.² We recruited a convenience sample of men who had undergone treatment and some of their wives. After we obtained informed consent, 15 focus groups were conducted with patients and 2 were conducted with the wives. The wives participated in focus groups immediately after their husbands, which minimized the opportunities for them to discuss the content of their respective focus groups. A medical sociologist and a urologist (both men) conducted the sessions, which were videotaped and transcribed for analysis.

We examined what the 20 married participants said about their experiences with prostate cancer. The husbands had an average age of 69 years; 11 were white, 9 were African American; 13 had undergone orchiectomy, and 7 were receiving monthly injections to suppress testosterone. Of the 7 wives, 5 were white and 2 African American.

We analyzed these data according to grounded theory methods of qualitative analysis.¹⁸ Thus, we developed codes to interpret the content of focus group participants’ accounts and then inductively developed summary themes characterizing their expressed experiences of prostate cancer. This method involves constant comparison of analytic codes. Within the set of 20 husbands, we compared how they described their communication with their wives in the context of discussions with other men. Then we compared within the set of 7 husband-wife pairs how each party described their communication with their spouse. The first author analyzed all data, and the second author independently analyzed a random selection of the data for validation. Disagreements in interpretation were resolved by continued comparative usage of the data, revising the codes and themes.

Results

The focus groups used an open-ended unstructured protocol. However, all discussions included questions about the perception of changes, including physical changes, sexuality, oneself or one’s spouse.

Physical Changes

Both husbands and wives readily answered questions about the occurrence of physical changes after prostate cancer treatment. Most men were forthcoming with descriptions of physical changes such as fatigue, hair loss, hot flashes, and weight gain. In their focus groups wives confirmed their husbands’ experience of these changes. Wives learned about these complications in 2 ways. Most men told their wives about their changed physical conditions. Some men, however, were silent about their physical well-being. In these cases, wives relied on nonverbal communication and close observation of their husbands. One man said, “I’ve slowed quite a bit. My wife, she worries about me more than I do myself. She watches every move I make when I’m around there. I don’t know why, but she keeps her eyes on me.”

Coping with Physical Changes

There was variation in couples’ communication. It ranged from sharing information to dealing with feelings in complete isolation. Two men whose wives were not interviewed had not told their wives they have prostate cancer. They reasoned that their wives would not be able to handle the information because they would associate cancer with death. In the Table 1 we present responses from one couple that express how both spouses noticed and had feelings about physical changes but avoided talking with each other about them.

Wives’ perspectives on couple communication were that verbal exchanges about feelings do not generally occur, but the reasons for this vary. Wives admit to not asking based on a fear they might stir things up in their husbands or create problems that were nonexistent for their husbands. Other wives expected their husbands to hide feelings, thereby admitting that their mental well-being is interrelated. For example, one wife said, “I don’t know if he worries about his death because if he would get down, I would get down. So he hides it very well.” Others stated that men refuse to talk about their feelings even when they are asked about their well-being. Despite this emphasis that the men were the ones who were not communicating, other women revealed that they were participants, by hiding their emotions or avoiding questions to protect their husbands.

The men’s statements indicated that they had a hard time adjusting to their physical changes and that they were not comfortable disclosing their feelings about these changes. They expressed feelings of embarrassment and shame about their physical changes. Also, the men stated their fears and how they withhold them from their wives. One man said:

The main thing is that you worry. A normal pain that you normally wouldn’t pay any attention to before, now you say, well maybe the cancer has spread to there…. My wife worries about that if I have a little ache or pain. I don’t normally tell her, because if I do, she goes crazy. So you have to be very careful with your wife, too.

Men generally pull themselves together. Reasons for the men’s reticence in showing emotions were either that they were protecting others, such as their wife and children, or that it was self-protection from the reactions of their wives.

The wives were more verbal about the despair they felt. Most of the men downplayed the implications of the cancer diagnosis or cancer treatment on their lives. They made statements such as “nothing has changed” and “life goes on as before.” This might partially explain the couples’ lack of communication about feelings and fears about cancer and how it alters men’s lives. Couples in general, but particularly the men, wanted to get their lives back together and move beyond prostate cancer. This desire may be so strong that it undercut the couples’ communication about fears and other feelings. One outcome of avoiding conversations is uncertainty about each other’s feelings or thoughts. This became most apparent when one of the wives questioned the interviewer about what their husbands shared in their interviews, wanting the interviewer to indicate what is burdensome to their husbands. She reasoned that knowing this would help them to support their husbands better.

Although not communicating openly, many of the couples used jokes and teasing to make each other comfortable. At the core of the jokes were important emotions, such as fears, that they seemed not to dare voice otherwise. Although the jokes transmitted some feelings or fears, they did not facilitate couple communication. The wives told jokes to build up their husbands’ morale, while the men used jokes to couch their fears. For example, one man indirectly revealed his fear of dying. His wife stated, “He is always asking me to lose weight, and he’ll say, ‘Now [wife’s name] you’re going to have a hard time getting another husband, if you keep putting that weight on.’”

The men’s communication with people other than their wives was even more restricted. Not all men feel comfortable enough to reveal their diagnosis to family and friends. Some of the men relied on their wives to inform others about the diagnosis, and the men who told others reported talking predominantly to other men. If they had children they talked to their sons, while not necessarily telling their daughters. If the men had conversations with others, the information was limited to stating the diagnosis and the outcome of check-ups. No man reported ever turning to other men to deal with feelings or fears about cancer, even though they assumed that some of their friends might have it as well. At best, men received factual information from other men but never emotional support or information about others’ experiences and adjustment to the disease. The wives’ communication network was larger and the quality of their support network different. Generally wives reported that others, such as daughters, relatives, church members, and neighbors who were friends, were available to them for support.

Perceptions of Changes in Spouse

Despite the complicated communication patterns, including an avoidance to disclose feelings about the effects of protate cancer, most couples denied that cancer caused other changes in their mates or in their marital relationship. Most of the couples denied that cancer caused other than physical changes in their mates or their marital relationship. The men generally said that their wives treated them just the same. Most of the wives confirmed this, stating that they have not increased their attention. However, the accounts by the husbands and the wives make references to changes that did occur. One man told the interviewer that his wife finds him difficult to get along with since his cancer treatment. Another man admitted being frequently irritable and having angry outbursts. Another change that was reported involved the men’s altered social network. Some wives talked about a decrease in their husbands’ interactions with friends. Some husbands confirmed that they underwent certain social changes. One husband stated that they socialize more as a couple, since he believes it to be important to his wife.

A few wives reported changes, such as their husbands’ jealousy. The women said that there was no reason for their husbands to be jealous since their commitments to their husbands remained unchanged. Jealousy was also brought up in the men’s focus groups. Both husbands and wives told the interviewer about the breakups of other marriages. The women referred to wives who left their husbands because they had prostate cancer, and the men reported that other men left their wives because they feared their wives were betraying them. One man explained:

I don’t try to worry about it, because really if you go to worrying about it, a person doing something, you can’t do it and with the going out or something like that, but I will just let her go, you know, if I’m going to sit there and worry, cause I’m not going to try to hurt myself. Now I’m concerned where she goes, but I won’t think of that…. I don’t want it on my mind. If it ever got on my mind, she may as well be doing it, see what I mean. I don’t even think about it.

Although it was a small minority that voiced worries and fears about betrayal and jealousy, it was an indication that this emergence of insecurity might have been triggered by the most radical change in couples’ lives: their changed sexuality.

Sexuality

In the interviews, husbands and wives described the implications of prostate cancer on their sex lives. With regard to communication in couples, however, most communication about sexuality took place when men were presented with treatment options that had impotence as a side effect. Husbands and wives described how they discussed the threat of giving up their sex lives. One man said, “we had quite a few discussions…. We just cried a little bit and that was it.” The same man revealed that his decision to choose hormone shots over surgery was also influenced by the belief that if he were to stop taking the hormones his sexual functioning would return, and surgery was permanent. Another man disclosed that after surgery he was depressed for months because he lost his ability to have sexual relations.

Several women let their husbands know that when faced with a choice between prolonging his life and having sexual relations, they would opt for his prolonged life expectancy. One woman stated, “I would rather have him than sex until I am 100.” Some men described having been motivated by the same tradeoff. The men’s descriptions, however, are dominated by references to the difficulty of consenting to a treatment that deprived them of their sexuality. One man explained its importance by stating, “I wouldn’t care if I had to walk around straddle-legged or not if I could have sex. This testicle removal hurt the desire for my sex life. It took it, and it dissolved my self away.”

After the decision about treatment, however, when the treatments take their toll on men’s sexual functioning, most couples’ communication about their sex life discontinues. But the men were outspoken with the interviewer in their descriptions of impotence. They stated their inability to achieve an erection, their loss of a sexual urge, and that lack of sexual functioning threatened their masculinity. Also, the men’s accounts indicated that their definition of sexuality is intercourse. The men’s inability to achieve an erection caused them to restrain from any type of sexual interactions. Generally, the men described their wives as understanding and accommodating to their loss of sexual functioning. One man told the interviewer that his wife made sexual advances that he could not reciprocate. Many men held the belief that the loss of sexuality had little or no effect on their wives. They portrayed their wives as adapting well to the loss of sexuality, because in men’s opinions women know how to cope with this change. Religiosity was one coping style of wives that men mentioned to the interviewer, “My wife goes to church a lot. She found the Lord, I guess. It don’t bother her much. She don’t act like it does.”

Men’s accounts reveal that there is little knowledge about their wives’ feelings about the sexual loss. Men’s assessment that sexual loss means little to their wives was based on reasoning that it must mean little to them since they are not complaining. As one man put it: “As far as I know [it does not bother her]… Now if she had said something…but if it is, she hasn’t said anything. It could be bothering her, but she hasn’t said anything.”

The importance of sexuality was mostly proclaimed in focus groups with men. Some men confided in the interview situation that their sexuality was diminished before prostate cancer. With increasing age, sex became less frequent. Others admitted to problems with impotence before prostate cancer. One man stated that being with his wife has always been more important than sexual activity. The men’s coping with a lack of sexuality after prostate cancer included avoidance (by putting it out of their minds), as well as through resignation by accepting that sexuality belongs to their past. These coping styles preclude seeking conversations with their wives.

Overwhelmingly, wives are focused on the importance of sexuality for their husbands without ever indicating the importance they give to sexuality. Wives’ reports center on ways of accommodating to the lack of sexual relations. One wife said:

It doesn’t bother me. I just made up my mind that it can’t be, so I just keep it off of my mind. I do other things to keep it off of my mind. I just try to stay active doing other things. It hadn’t bothered me. It used to, it would bother me to hug and kiss him, but now it doesn’t since it is no longer that way. It can’t be. I just made up my mind. Consistently, wives stated how devastating the loss of sex is for their husbands, paying little attention to their own needs. One wife wished her husband could have an erection, because she believed it was crucial for his masculinity. Wives accommodated the lack of sex and made it known that they would never seek sex outside of marriage. Instead, they did everything they could to build up their husband’s self-esteem by reassuring him that his masculinity was not tied to sexual performance. Wives confirmed that there is no sexual activity of any form; one wife even stated that her husband is less caring or romantic since his sexual dysfunction. Another wife reacted surprised to the interviewer raising the possibility of sexual activity. The avoidance of communication about sexuality leaves spouses on their own when adjusting to sexual loss.

Discussion

The focus group interviews with men and some of their wives provided us with insight into couples’ experiences with prostate cancer. Their perspectives contribute toward understanding the implications of prostate cancer on the marital relationship. Our analysis is greatly influenced by the suggestion of previous studies that open communication among spouses is important for a positive adjustment by patients and their spouses.^7,8 We find that most of the men are capable of communicating the factual physical changes, while they appear unwilling, or perhaps unable, to communicate with their wives about their feelings regarding these changes.

Limitations

The inclusion of spouses in studies that focus on the well-being of men with prostate cancer is imperative. However, we are aware of the limitations of the wives’ focus group data. In addition to the small number, our study’s focus restricted the interviews with wives mostly to their assessment of their husbands’ well-being. The women had little opportunity to discuss their subjective experiences. Also, the focus groups with wives were conducted by 2 men. One can speculate that discussions of sensitive matters such as sexuality may be constrained when the interviewers are of the opposite sex. The same-sex environment of the men’s focus groups, however, may have elicited frank and honest accounts. Our limited sample prevents us from exploring specific cultural differences in couples of different racial or ethnic background. The conclusions we reached were affected by the preliminary nature of our study. In particular, we have no information about the couples’ marital relationship and communication styles before the diagnosis of prostate cancer. Other research indicates that couples’ communication patterns after a diagnosis are similar to the style they had before the diagnosis.⁸ Also, our study’s findings may be affected by the stage of disease. Future prospective studies that measure psychological well-being of patients with prostate cancer and their partners need to consider the important issues raised by our study. In particular, the apparent contradiction of not communicating about fears and death while claiming marital satisfaction warrants further research. Larger studies will have to demonstrate the benefits of disclosure of feelings on adjustment before care providers are asked to contribute to open communication between spouses. Nevertheless, the findings of our research indicate that the inclusion of spouses in research expands our understanding of the effects of prostate cancer on well-being.

Lack of Communication

The data give no indication that men deal with cancer by discussing fears of death and dependency with their wives. Instead they express a desire to put the cancer behind them, “to be done with the disease.” The men presented themselves as having adjusted to the disease with a self-identity that is unchanged by prostate cancer. Their lack of communication about emotions received little challenge from their wives. Moreover, the wives collaborated in that they hid their own fears and despair instead of pursuing spousal communication about emotions. Thus, protective buffering is the prevalent coping style by both spouses, confirming the findings of earlier research that spouses’ coping styles and emotions are interrelated.^7,19,20

This is problematic for 2 reasons. Earlier research indicates that there is an inverse relationship between protective buffering and marital satisfaction, as well as a strong positive association between protective buffering and patient and spouse distress.⁷ The couples with metastatic prostate cancer claim that there is little change in their mates. They overwhelmingly portray themselves as happy and content with their marital relationship. Contrary to this depiction, references to moodiness and the emergence of jealousy point to strains on the marital relationship. Also, lack of communication leads to uncertainty about their spouse’s feelings and thoughts, another potential strain on the marital relationship. Although our study does not entail an assessment of men’s and wives individual or joint adjustment to prostate cancer, we certainly find evidence of the risk factors for poor adjustment which previous studies indicate: lack of communicating and protective buffering. Also, losing or abandoning one’s sex life, as in our sample, is a significant change that most likely has a strong association with psychological well-being, as found in earlier studies.²¹ This appears to suggest that the couples with metastatic prostate cancer may be at risk for distress and poor adjustment.

Conclusions

These insights into couples’ coping styles are of special relevance to physicians who care for men with prostate cancer. Generally, physicians are open to married men involving their wives in the management of prostate cancer. However, our findings raise questions about how involved wives really are. Although women are interested in their husbands’ prostate cancer, the lack of communication within the couple suggests that their active involvement may be less than is commonly assumed. Care providers may positively influence patients’ adaptation and quality of life by facilitating involvement by the patient’s wife, rather than assume that her presence alone signals active involvement. Also, physicians can suspect that patients who choose to go through treatment by themselves may be at risk for poor adjustment to their diagnosis.

Related Resources

• National Prostate Cancer Coalition http://www.4npcc.org/Advocates.htm/
• US TOO, International http://www.ustoo.com/
• Online Decision Support for Prostate Cancer http://www.cancerfacts.com/

· Acknowledgments ·

Support for our research was provided by US Army Medical Research and Materiel Command, Fort Detrick, Maryland, grant number DAMD17-99-1-9052, Principle Investigator, Dr Boehmer. The data collection was supported by the Department of Veterans Affairs Health Services Research and Development Service grant SDR-93-007.

Prostate cancer (the most common cancer in men) has a considerable impact on the quality of patients’ lives. After a frightening diagnosis, the treatments cause a variety of unpleasant side effects. In the case of metastatic disease, hormonal treatment through either orchiectomy or a permanent regimen of medications to suppress testosterone causes nausea, hot flashes, loss of muscle tone, and erectile dysfunction.¹ The complex psychosocial effects of diagnosis and treatment, such as changes in body and self-image, masculinity, sexuality, and uncertainty, are gradually recognized.² However, most investigations of quality of life have focused on the patients, increasing our understanding of how men respond to prostate cancer, but telling us little about how this disease affects the shared lives of men and their significant others. Prostate cancer threatens men’s survival and families’ futures. Sexual dysfunction following treatment has implications for both men and their intimate partners. Therefore, we explored the perceptions of men who have been treated for prostate cancer, and those of their wives, regarding the changes that were caused by prostate cancer and its treatment.

The effect of illness on marital quality is largely an unresolved area.³ Some researchers argue that there is no change—or if there is a change, it is a positive association between illness and the spousal relationship, because it brings spouses closer together.⁴ Others point to negative impacts on the marital relationship.⁵ It has been suggested that a worsening of marital quality can be explained by the financial implications of illness, the problematic behavior of the sick person, and the changed division of labor in the couple, including the loss of shared activities.⁵ A decline in marital quality is often perceived by spouses who suffer from increased stress and depression because of their spouses’ disease.^5,6 Research on couples coping with illness has highlighted the importance of open communication for positive adjustment by patients and their spouses.^7,8 But couples’ ability to communicate varies,⁸ and many avoid truthful and open communication with each other⁹ and with other friends and family members.¹⁰

Only a few prostate cancer studies have considered the patient’s social context. Studies that include spouses point to the interrelationship of their reactions to prostate cancer.^3,11-17 We focussed on couples’ communication about prostate cancer and its treatment effects.

Methods

Our study is part of a larger investigation of the psychosocial effects of metastatic prostate cancer, that had the development of quality-of-life scales as its goal.² We recruited a convenience sample of men who had undergone treatment and some of their wives. After we obtained informed consent, 15 focus groups were conducted with patients and 2 were conducted with the wives. The wives participated in focus groups immediately after their husbands, which minimized the opportunities for them to discuss the content of their respective focus groups. A medical sociologist and a urologist (both men) conducted the sessions, which were videotaped and transcribed for analysis.

We examined what the 20 married participants said about their experiences with prostate cancer. The husbands had an average age of 69 years; 11 were white, 9 were African American; 13 had undergone orchiectomy, and 7 were receiving monthly injections to suppress testosterone. Of the 7 wives, 5 were white and 2 African American.

We analyzed these data according to grounded theory methods of qualitative analysis.¹⁸ Thus, we developed codes to interpret the content of focus group participants’ accounts and then inductively developed summary themes characterizing their expressed experiences of prostate cancer. This method involves constant comparison of analytic codes. Within the set of 20 husbands, we compared how they described their communication with their wives in the context of discussions with other men. Then we compared within the set of 7 husband-wife pairs how each party described their communication with their spouse. The first author analyzed all data, and the second author independently analyzed a random selection of the data for validation. Disagreements in interpretation were resolved by continued comparative usage of the data, revising the codes and themes.

Results

The focus groups used an open-ended unstructured protocol. However, all discussions included questions about the perception of changes, including physical changes, sexuality, oneself or one’s spouse.

Physical Changes

Both husbands and wives readily answered questions about the occurrence of physical changes after prostate cancer treatment. Most men were forthcoming with descriptions of physical changes such as fatigue, hair loss, hot flashes, and weight gain. In their focus groups wives confirmed their husbands’ experience of these changes. Wives learned about these complications in 2 ways. Most men told their wives about their changed physical conditions. Some men, however, were silent about their physical well-being. In these cases, wives relied on nonverbal communication and close observation of their husbands. One man said, “I’ve slowed quite a bit. My wife, she worries about me more than I do myself. She watches every move I make when I’m around there. I don’t know why, but she keeps her eyes on me.”

Coping with Physical Changes

There was variation in couples’ communication. It ranged from sharing information to dealing with feelings in complete isolation. Two men whose wives were not interviewed had not told their wives they have prostate cancer. They reasoned that their wives would not be able to handle the information because they would associate cancer with death. In the Table 1 we present responses from one couple that express how both spouses noticed and had feelings about physical changes but avoided talking with each other about them.

Wives’ perspectives on couple communication were that verbal exchanges about feelings do not generally occur, but the reasons for this vary. Wives admit to not asking based on a fear they might stir things up in their husbands or create problems that were nonexistent for their husbands. Other wives expected their husbands to hide feelings, thereby admitting that their mental well-being is interrelated. For example, one wife said, “I don’t know if he worries about his death because if he would get down, I would get down. So he hides it very well.” Others stated that men refuse to talk about their feelings even when they are asked about their well-being. Despite this emphasis that the men were the ones who were not communicating, other women revealed that they were participants, by hiding their emotions or avoiding questions to protect their husbands.

The men’s statements indicated that they had a hard time adjusting to their physical changes and that they were not comfortable disclosing their feelings about these changes. They expressed feelings of embarrassment and shame about their physical changes. Also, the men stated their fears and how they withhold them from their wives. One man said:

The main thing is that you worry. A normal pain that you normally wouldn’t pay any attention to before, now you say, well maybe the cancer has spread to there…. My wife worries about that if I have a little ache or pain. I don’t normally tell her, because if I do, she goes crazy. So you have to be very careful with your wife, too.

Men generally pull themselves together. Reasons for the men’s reticence in showing emotions were either that they were protecting others, such as their wife and children, or that it was self-protection from the reactions of their wives.

The wives were more verbal about the despair they felt. Most of the men downplayed the implications of the cancer diagnosis or cancer treatment on their lives. They made statements such as “nothing has changed” and “life goes on as before.” This might partially explain the couples’ lack of communication about feelings and fears about cancer and how it alters men’s lives. Couples in general, but particularly the men, wanted to get their lives back together and move beyond prostate cancer. This desire may be so strong that it undercut the couples’ communication about fears and other feelings. One outcome of avoiding conversations is uncertainty about each other’s feelings or thoughts. This became most apparent when one of the wives questioned the interviewer about what their husbands shared in their interviews, wanting the interviewer to indicate what is burdensome to their husbands. She reasoned that knowing this would help them to support their husbands better.

Although not communicating openly, many of the couples used jokes and teasing to make each other comfortable. At the core of the jokes were important emotions, such as fears, that they seemed not to dare voice otherwise. Although the jokes transmitted some feelings or fears, they did not facilitate couple communication. The wives told jokes to build up their husbands’ morale, while the men used jokes to couch their fears. For example, one man indirectly revealed his fear of dying. His wife stated, “He is always asking me to lose weight, and he’ll say, ‘Now [wife’s name] you’re going to have a hard time getting another husband, if you keep putting that weight on.’”

The men’s communication with people other than their wives was even more restricted. Not all men feel comfortable enough to reveal their diagnosis to family and friends. Some of the men relied on their wives to inform others about the diagnosis, and the men who told others reported talking predominantly to other men. If they had children they talked to their sons, while not necessarily telling their daughters. If the men had conversations with others, the information was limited to stating the diagnosis and the outcome of check-ups. No man reported ever turning to other men to deal with feelings or fears about cancer, even though they assumed that some of their friends might have it as well. At best, men received factual information from other men but never emotional support or information about others’ experiences and adjustment to the disease. The wives’ communication network was larger and the quality of their support network different. Generally wives reported that others, such as daughters, relatives, church members, and neighbors who were friends, were available to them for support.

Perceptions of Changes in Spouse

Despite the complicated communication patterns, including an avoidance to disclose feelings about the effects of protate cancer, most couples denied that cancer caused other changes in their mates or in their marital relationship. Most of the couples denied that cancer caused other than physical changes in their mates or their marital relationship. The men generally said that their wives treated them just the same. Most of the wives confirmed this, stating that they have not increased their attention. However, the accounts by the husbands and the wives make references to changes that did occur. One man told the interviewer that his wife finds him difficult to get along with since his cancer treatment. Another man admitted being frequently irritable and having angry outbursts. Another change that was reported involved the men’s altered social network. Some wives talked about a decrease in their husbands’ interactions with friends. Some husbands confirmed that they underwent certain social changes. One husband stated that they socialize more as a couple, since he believes it to be important to his wife.

A few wives reported changes, such as their husbands’ jealousy. The women said that there was no reason for their husbands to be jealous since their commitments to their husbands remained unchanged. Jealousy was also brought up in the men’s focus groups. Both husbands and wives told the interviewer about the breakups of other marriages. The women referred to wives who left their husbands because they had prostate cancer, and the men reported that other men left their wives because they feared their wives were betraying them. One man explained:

I don’t try to worry about it, because really if you go to worrying about it, a person doing something, you can’t do it and with the going out or something like that, but I will just let her go, you know, if I’m going to sit there and worry, cause I’m not going to try to hurt myself. Now I’m concerned where she goes, but I won’t think of that…. I don’t want it on my mind. If it ever got on my mind, she may as well be doing it, see what I mean. I don’t even think about it.

Although it was a small minority that voiced worries and fears about betrayal and jealousy, it was an indication that this emergence of insecurity might have been triggered by the most radical change in couples’ lives: their changed sexuality.

Sexuality

In the interviews, husbands and wives described the implications of prostate cancer on their sex lives. With regard to communication in couples, however, most communication about sexuality took place when men were presented with treatment options that had impotence as a side effect. Husbands and wives described how they discussed the threat of giving up their sex lives. One man said, “we had quite a few discussions…. We just cried a little bit and that was it.” The same man revealed that his decision to choose hormone shots over surgery was also influenced by the belief that if he were to stop taking the hormones his sexual functioning would return, and surgery was permanent. Another man disclosed that after surgery he was depressed for months because he lost his ability to have sexual relations.

Several women let their husbands know that when faced with a choice between prolonging his life and having sexual relations, they would opt for his prolonged life expectancy. One woman stated, “I would rather have him than sex until I am 100.” Some men described having been motivated by the same tradeoff. The men’s descriptions, however, are dominated by references to the difficulty of consenting to a treatment that deprived them of their sexuality. One man explained its importance by stating, “I wouldn’t care if I had to walk around straddle-legged or not if I could have sex. This testicle removal hurt the desire for my sex life. It took it, and it dissolved my self away.”

After the decision about treatment, however, when the treatments take their toll on men’s sexual functioning, most couples’ communication about their sex life discontinues. But the men were outspoken with the interviewer in their descriptions of impotence. They stated their inability to achieve an erection, their loss of a sexual urge, and that lack of sexual functioning threatened their masculinity. Also, the men’s accounts indicated that their definition of sexuality is intercourse. The men’s inability to achieve an erection caused them to restrain from any type of sexual interactions. Generally, the men described their wives as understanding and accommodating to their loss of sexual functioning. One man told the interviewer that his wife made sexual advances that he could not reciprocate. Many men held the belief that the loss of sexuality had little or no effect on their wives. They portrayed their wives as adapting well to the loss of sexuality, because in men’s opinions women know how to cope with this change. Religiosity was one coping style of wives that men mentioned to the interviewer, “My wife goes to church a lot. She found the Lord, I guess. It don’t bother her much. She don’t act like it does.”

Men’s accounts reveal that there is little knowledge about their wives’ feelings about the sexual loss. Men’s assessment that sexual loss means little to their wives was based on reasoning that it must mean little to them since they are not complaining. As one man put it: “As far as I know [it does not bother her]… Now if she had said something…but if it is, she hasn’t said anything. It could be bothering her, but she hasn’t said anything.”

The importance of sexuality was mostly proclaimed in focus groups with men. Some men confided in the interview situation that their sexuality was diminished before prostate cancer. With increasing age, sex became less frequent. Others admitted to problems with impotence before prostate cancer. One man stated that being with his wife has always been more important than sexual activity. The men’s coping with a lack of sexuality after prostate cancer included avoidance (by putting it out of their minds), as well as through resignation by accepting that sexuality belongs to their past. These coping styles preclude seeking conversations with their wives.

Overwhelmingly, wives are focused on the importance of sexuality for their husbands without ever indicating the importance they give to sexuality. Wives’ reports center on ways of accommodating to the lack of sexual relations. One wife said:

It doesn’t bother me. I just made up my mind that it can’t be, so I just keep it off of my mind. I do other things to keep it off of my mind. I just try to stay active doing other things. It hadn’t bothered me. It used to, it would bother me to hug and kiss him, but now it doesn’t since it is no longer that way. It can’t be. I just made up my mind. Consistently, wives stated how devastating the loss of sex is for their husbands, paying little attention to their own needs. One wife wished her husband could have an erection, because she believed it was crucial for his masculinity. Wives accommodated the lack of sex and made it known that they would never seek sex outside of marriage. Instead, they did everything they could to build up their husband’s self-esteem by reassuring him that his masculinity was not tied to sexual performance. Wives confirmed that there is no sexual activity of any form; one wife even stated that her husband is less caring or romantic since his sexual dysfunction. Another wife reacted surprised to the interviewer raising the possibility of sexual activity. The avoidance of communication about sexuality leaves spouses on their own when adjusting to sexual loss.

Discussion

The focus group interviews with men and some of their wives provided us with insight into couples’ experiences with prostate cancer. Their perspectives contribute toward understanding the implications of prostate cancer on the marital relationship. Our analysis is greatly influenced by the suggestion of previous studies that open communication among spouses is important for a positive adjustment by patients and their spouses.^7,8 We find that most of the men are capable of communicating the factual physical changes, while they appear unwilling, or perhaps unable, to communicate with their wives about their feelings regarding these changes.

Limitations

The inclusion of spouses in studies that focus on the well-being of men with prostate cancer is imperative. However, we are aware of the limitations of the wives’ focus group data. In addition to the small number, our study’s focus restricted the interviews with wives mostly to their assessment of their husbands’ well-being. The women had little opportunity to discuss their subjective experiences. Also, the focus groups with wives were conducted by 2 men. One can speculate that discussions of sensitive matters such as sexuality may be constrained when the interviewers are of the opposite sex. The same-sex environment of the men’s focus groups, however, may have elicited frank and honest accounts. Our limited sample prevents us from exploring specific cultural differences in couples of different racial or ethnic background. The conclusions we reached were affected by the preliminary nature of our study. In particular, we have no information about the couples’ marital relationship and communication styles before the diagnosis of prostate cancer. Other research indicates that couples’ communication patterns after a diagnosis are similar to the style they had before the diagnosis.⁸ Also, our study’s findings may be affected by the stage of disease. Future prospective studies that measure psychological well-being of patients with prostate cancer and their partners need to consider the important issues raised by our study. In particular, the apparent contradiction of not communicating about fears and death while claiming marital satisfaction warrants further research. Larger studies will have to demonstrate the benefits of disclosure of feelings on adjustment before care providers are asked to contribute to open communication between spouses. Nevertheless, the findings of our research indicate that the inclusion of spouses in research expands our understanding of the effects of prostate cancer on well-being.

Lack of Communication

The data give no indication that men deal with cancer by discussing fears of death and dependency with their wives. Instead they express a desire to put the cancer behind them, “to be done with the disease.” The men presented themselves as having adjusted to the disease with a self-identity that is unchanged by prostate cancer. Their lack of communication about emotions received little challenge from their wives. Moreover, the wives collaborated in that they hid their own fears and despair instead of pursuing spousal communication about emotions. Thus, protective buffering is the prevalent coping style by both spouses, confirming the findings of earlier research that spouses’ coping styles and emotions are interrelated.^7,19,20

This is problematic for 2 reasons. Earlier research indicates that there is an inverse relationship between protective buffering and marital satisfaction, as well as a strong positive association between protective buffering and patient and spouse distress.⁷ The couples with metastatic prostate cancer claim that there is little change in their mates. They overwhelmingly portray themselves as happy and content with their marital relationship. Contrary to this depiction, references to moodiness and the emergence of jealousy point to strains on the marital relationship. Also, lack of communication leads to uncertainty about their spouse’s feelings and thoughts, another potential strain on the marital relationship. Although our study does not entail an assessment of men’s and wives individual or joint adjustment to prostate cancer, we certainly find evidence of the risk factors for poor adjustment which previous studies indicate: lack of communicating and protective buffering. Also, losing or abandoning one’s sex life, as in our sample, is a significant change that most likely has a strong association with psychological well-being, as found in earlier studies.²¹ This appears to suggest that the couples with metastatic prostate cancer may be at risk for distress and poor adjustment.

Conclusions

These insights into couples’ coping styles are of special relevance to physicians who care for men with prostate cancer. Generally, physicians are open to married men involving their wives in the management of prostate cancer. However, our findings raise questions about how involved wives really are. Although women are interested in their husbands’ prostate cancer, the lack of communication within the couple suggests that their active involvement may be less than is commonly assumed. Care providers may positively influence patients’ adaptation and quality of life by facilitating involvement by the patient’s wife, rather than assume that her presence alone signals active involvement. Also, physicians can suspect that patients who choose to go through treatment by themselves may be at risk for poor adjustment to their diagnosis.

Related Resources

• National Prostate Cancer Coalition http://www.4npcc.org/Advocates.htm/
• US TOO, International http://www.ustoo.com/
• Online Decision Support for Prostate Cancer http://www.cancerfacts.com/

· Acknowledgments ·

Support for our research was provided by US Army Medical Research and Materiel Command, Fort Detrick, Maryland, grant number DAMD17-99-1-9052, Principle Investigator, Dr Boehmer. The data collection was supported by the Department of Veterans Affairs Health Services Research and Development Service grant SDR-93-007.

Prostate cancer (the most common cancer in men) has a considerable impact on the quality of patients’ lives. After a frightening diagnosis, the treatments cause a variety of unpleasant side effects. In the case of metastatic disease, hormonal treatment through either orchiectomy or a permanent regimen of medications to suppress testosterone causes nausea, hot flashes, loss of muscle tone, and erectile dysfunction.¹ The complex psychosocial effects of diagnosis and treatment, such as changes in body and self-image, masculinity, sexuality, and uncertainty, are gradually recognized.² However, most investigations of quality of life have focused on the patients, increasing our understanding of how men respond to prostate cancer, but telling us little about how this disease affects the shared lives of men and their significant others. Prostate cancer threatens men’s survival and families’ futures. Sexual dysfunction following treatment has implications for both men and their intimate partners. Therefore, we explored the perceptions of men who have been treated for prostate cancer, and those of their wives, regarding the changes that were caused by prostate cancer and its treatment.

The effect of illness on marital quality is largely an unresolved area.³ Some researchers argue that there is no change—or if there is a change, it is a positive association between illness and the spousal relationship, because it brings spouses closer together.⁴ Others point to negative impacts on the marital relationship.⁵ It has been suggested that a worsening of marital quality can be explained by the financial implications of illness, the problematic behavior of the sick person, and the changed division of labor in the couple, including the loss of shared activities.⁵ A decline in marital quality is often perceived by spouses who suffer from increased stress and depression because of their spouses’ disease.^5,6 Research on couples coping with illness has highlighted the importance of open communication for positive adjustment by patients and their spouses.^7,8 But couples’ ability to communicate varies,⁸ and many avoid truthful and open communication with each other⁹ and with other friends and family members.¹⁰

Only a few prostate cancer studies have considered the patient’s social context. Studies that include spouses point to the interrelationship of their reactions to prostate cancer.^3,11-17 We focussed on couples’ communication about prostate cancer and its treatment effects.

Methods

Our study is part of a larger investigation of the psychosocial effects of metastatic prostate cancer, that had the development of quality-of-life scales as its goal.² We recruited a convenience sample of men who had undergone treatment and some of their wives. After we obtained informed consent, 15 focus groups were conducted with patients and 2 were conducted with the wives. The wives participated in focus groups immediately after their husbands, which minimized the opportunities for them to discuss the content of their respective focus groups. A medical sociologist and a urologist (both men) conducted the sessions, which were videotaped and transcribed for analysis.

We examined what the 20 married participants said about their experiences with prostate cancer. The husbands had an average age of 69 years; 11 were white, 9 were African American; 13 had undergone orchiectomy, and 7 were receiving monthly injections to suppress testosterone. Of the 7 wives, 5 were white and 2 African American.

We analyzed these data according to grounded theory methods of qualitative analysis.¹⁸ Thus, we developed codes to interpret the content of focus group participants’ accounts and then inductively developed summary themes characterizing their expressed experiences of prostate cancer. This method involves constant comparison of analytic codes. Within the set of 20 husbands, we compared how they described their communication with their wives in the context of discussions with other men. Then we compared within the set of 7 husband-wife pairs how each party described their communication with their spouse. The first author analyzed all data, and the second author independently analyzed a random selection of the data for validation. Disagreements in interpretation were resolved by continued comparative usage of the data, revising the codes and themes.

Results

The focus groups used an open-ended unstructured protocol. However, all discussions included questions about the perception of changes, including physical changes, sexuality, oneself or one’s spouse.

Physical Changes

Both husbands and wives readily answered questions about the occurrence of physical changes after prostate cancer treatment. Most men were forthcoming with descriptions of physical changes such as fatigue, hair loss, hot flashes, and weight gain. In their focus groups wives confirmed their husbands’ experience of these changes. Wives learned about these complications in 2 ways. Most men told their wives about their changed physical conditions. Some men, however, were silent about their physical well-being. In these cases, wives relied on nonverbal communication and close observation of their husbands. One man said, “I’ve slowed quite a bit. My wife, she worries about me more than I do myself. She watches every move I make when I’m around there. I don’t know why, but she keeps her eyes on me.”

Coping with Physical Changes

There was variation in couples’ communication. It ranged from sharing information to dealing with feelings in complete isolation. Two men whose wives were not interviewed had not told their wives they have prostate cancer. They reasoned that their wives would not be able to handle the information because they would associate cancer with death. In the Table 1 we present responses from one couple that express how both spouses noticed and had feelings about physical changes but avoided talking with each other about them.

Wives’ perspectives on couple communication were that verbal exchanges about feelings do not generally occur, but the reasons for this vary. Wives admit to not asking based on a fear they might stir things up in their husbands or create problems that were nonexistent for their husbands. Other wives expected their husbands to hide feelings, thereby admitting that their mental well-being is interrelated. For example, one wife said, “I don’t know if he worries about his death because if he would get down, I would get down. So he hides it very well.” Others stated that men refuse to talk about their feelings even when they are asked about their well-being. Despite this emphasis that the men were the ones who were not communicating, other women revealed that they were participants, by hiding their emotions or avoiding questions to protect their husbands.

The men’s statements indicated that they had a hard time adjusting to their physical changes and that they were not comfortable disclosing their feelings about these changes. They expressed feelings of embarrassment and shame about their physical changes. Also, the men stated their fears and how they withhold them from their wives. One man said:

The main thing is that you worry. A normal pain that you normally wouldn’t pay any attention to before, now you say, well maybe the cancer has spread to there…. My wife worries about that if I have a little ache or pain. I don’t normally tell her, because if I do, she goes crazy. So you have to be very careful with your wife, too.

Men generally pull themselves together. Reasons for the men’s reticence in showing emotions were either that they were protecting others, such as their wife and children, or that it was self-protection from the reactions of their wives.

The wives were more verbal about the despair they felt. Most of the men downplayed the implications of the cancer diagnosis or cancer treatment on their lives. They made statements such as “nothing has changed” and “life goes on as before.” This might partially explain the couples’ lack of communication about feelings and fears about cancer and how it alters men’s lives. Couples in general, but particularly the men, wanted to get their lives back together and move beyond prostate cancer. This desire may be so strong that it undercut the couples’ communication about fears and other feelings. One outcome of avoiding conversations is uncertainty about each other’s feelings or thoughts. This became most apparent when one of the wives questioned the interviewer about what their husbands shared in their interviews, wanting the interviewer to indicate what is burdensome to their husbands. She reasoned that knowing this would help them to support their husbands better.

Although not communicating openly, many of the couples used jokes and teasing to make each other comfortable. At the core of the jokes were important emotions, such as fears, that they seemed not to dare voice otherwise. Although the jokes transmitted some feelings or fears, they did not facilitate couple communication. The wives told jokes to build up their husbands’ morale, while the men used jokes to couch their fears. For example, one man indirectly revealed his fear of dying. His wife stated, “He is always asking me to lose weight, and he’ll say, ‘Now [wife’s name] you’re going to have a hard time getting another husband, if you keep putting that weight on.’”

The men’s communication with people other than their wives was even more restricted. Not all men feel comfortable enough to reveal their diagnosis to family and friends. Some of the men relied on their wives to inform others about the diagnosis, and the men who told others reported talking predominantly to other men. If they had children they talked to their sons, while not necessarily telling their daughters. If the men had conversations with others, the information was limited to stating the diagnosis and the outcome of check-ups. No man reported ever turning to other men to deal with feelings or fears about cancer, even though they assumed that some of their friends might have it as well. At best, men received factual information from other men but never emotional support or information about others’ experiences and adjustment to the disease. The wives’ communication network was larger and the quality of their support network different. Generally wives reported that others, such as daughters, relatives, church members, and neighbors who were friends, were available to them for support.

Perceptions of Changes in Spouse

Despite the complicated communication patterns, including an avoidance to disclose feelings about the effects of protate cancer, most couples denied that cancer caused other changes in their mates or in their marital relationship. Most of the couples denied that cancer caused other than physical changes in their mates or their marital relationship. The men generally said that their wives treated them just the same. Most of the wives confirmed this, stating that they have not increased their attention. However, the accounts by the husbands and the wives make references to changes that did occur. One man told the interviewer that his wife finds him difficult to get along with since his cancer treatment. Another man admitted being frequently irritable and having angry outbursts. Another change that was reported involved the men’s altered social network. Some wives talked about a decrease in their husbands’ interactions with friends. Some husbands confirmed that they underwent certain social changes. One husband stated that they socialize more as a couple, since he believes it to be important to his wife.

A few wives reported changes, such as their husbands’ jealousy. The women said that there was no reason for their husbands to be jealous since their commitments to their husbands remained unchanged. Jealousy was also brought up in the men’s focus groups. Both husbands and wives told the interviewer about the breakups of other marriages. The women referred to wives who left their husbands because they had prostate cancer, and the men reported that other men left their wives because they feared their wives were betraying them. One man explained:

I don’t try to worry about it, because really if you go to worrying about it, a person doing something, you can’t do it and with the going out or something like that, but I will just let her go, you know, if I’m going to sit there and worry, cause I’m not going to try to hurt myself. Now I’m concerned where she goes, but I won’t think of that…. I don’t want it on my mind. If it ever got on my mind, she may as well be doing it, see what I mean. I don’t even think about it.

Although it was a small minority that voiced worries and fears about betrayal and jealousy, it was an indication that this emergence of insecurity might have been triggered by the most radical change in couples’ lives: their changed sexuality.

Sexuality

In the interviews, husbands and wives described the implications of prostate cancer on their sex lives. With regard to communication in couples, however, most communication about sexuality took place when men were presented with treatment options that had impotence as a side effect. Husbands and wives described how they discussed the threat of giving up their sex lives. One man said, “we had quite a few discussions…. We just cried a little bit and that was it.” The same man revealed that his decision to choose hormone shots over surgery was also influenced by the belief that if he were to stop taking the hormones his sexual functioning would return, and surgery was permanent. Another man disclosed that after surgery he was depressed for months because he lost his ability to have sexual relations.

Several women let their husbands know that when faced with a choice between prolonging his life and having sexual relations, they would opt for his prolonged life expectancy. One woman stated, “I would rather have him than sex until I am 100.” Some men described having been motivated by the same tradeoff. The men’s descriptions, however, are dominated by references to the difficulty of consenting to a treatment that deprived them of their sexuality. One man explained its importance by stating, “I wouldn’t care if I had to walk around straddle-legged or not if I could have sex. This testicle removal hurt the desire for my sex life. It took it, and it dissolved my self away.”

After the decision about treatment, however, when the treatments take their toll on men’s sexual functioning, most couples’ communication about their sex life discontinues. But the men were outspoken with the interviewer in their descriptions of impotence. They stated their inability to achieve an erection, their loss of a sexual urge, and that lack of sexual functioning threatened their masculinity. Also, the men’s accounts indicated that their definition of sexuality is intercourse. The men’s inability to achieve an erection caused them to restrain from any type of sexual interactions. Generally, the men described their wives as understanding and accommodating to their loss of sexual functioning. One man told the interviewer that his wife made sexual advances that he could not reciprocate. Many men held the belief that the loss of sexuality had little or no effect on their wives. They portrayed their wives as adapting well to the loss of sexuality, because in men’s opinions women know how to cope with this change. Religiosity was one coping style of wives that men mentioned to the interviewer, “My wife goes to church a lot. She found the Lord, I guess. It don’t bother her much. She don’t act like it does.”

Men’s accounts reveal that there is little knowledge about their wives’ feelings about the sexual loss. Men’s assessment that sexual loss means little to their wives was based on reasoning that it must mean little to them since they are not complaining. As one man put it: “As far as I know [it does not bother her]… Now if she had said something…but if it is, she hasn’t said anything. It could be bothering her, but she hasn’t said anything.”

The importance of sexuality was mostly proclaimed in focus groups with men. Some men confided in the interview situation that their sexuality was diminished before prostate cancer. With increasing age, sex became less frequent. Others admitted to problems with impotence before prostate cancer. One man stated that being with his wife has always been more important than sexual activity. The men’s coping with a lack of sexuality after prostate cancer included avoidance (by putting it out of their minds), as well as through resignation by accepting that sexuality belongs to their past. These coping styles preclude seeking conversations with their wives.

Overwhelmingly, wives are focused on the importance of sexuality for their husbands without ever indicating the importance they give to sexuality. Wives’ reports center on ways of accommodating to the lack of sexual relations. One wife said:

It doesn’t bother me. I just made up my mind that it can’t be, so I just keep it off of my mind. I do other things to keep it off of my mind. I just try to stay active doing other things. It hadn’t bothered me. It used to, it would bother me to hug and kiss him, but now it doesn’t since it is no longer that way. It can’t be. I just made up my mind. Consistently, wives stated how devastating the loss of sex is for their husbands, paying little attention to their own needs. One wife wished her husband could have an erection, because she believed it was crucial for his masculinity. Wives accommodated the lack of sex and made it known that they would never seek sex outside of marriage. Instead, they did everything they could to build up their husband’s self-esteem by reassuring him that his masculinity was not tied to sexual performance. Wives confirmed that there is no sexual activity of any form; one wife even stated that her husband is less caring or romantic since his sexual dysfunction. Another wife reacted surprised to the interviewer raising the possibility of sexual activity. The avoidance of communication about sexuality leaves spouses on their own when adjusting to sexual loss.

Discussion

The focus group interviews with men and some of their wives provided us with insight into couples’ experiences with prostate cancer. Their perspectives contribute toward understanding the implications of prostate cancer on the marital relationship. Our analysis is greatly influenced by the suggestion of previous studies that open communication among spouses is important for a positive adjustment by patients and their spouses.^7,8 We find that most of the men are capable of communicating the factual physical changes, while they appear unwilling, or perhaps unable, to communicate with their wives about their feelings regarding these changes.

Limitations

The inclusion of spouses in studies that focus on the well-being of men with prostate cancer is imperative. However, we are aware of the limitations of the wives’ focus group data. In addition to the small number, our study’s focus restricted the interviews with wives mostly to their assessment of their husbands’ well-being. The women had little opportunity to discuss their subjective experiences. Also, the focus groups with wives were conducted by 2 men. One can speculate that discussions of sensitive matters such as sexuality may be constrained when the interviewers are of the opposite sex. The same-sex environment of the men’s focus groups, however, may have elicited frank and honest accounts. Our limited sample prevents us from exploring specific cultural differences in couples of different racial or ethnic background. The conclusions we reached were affected by the preliminary nature of our study. In particular, we have no information about the couples’ marital relationship and communication styles before the diagnosis of prostate cancer. Other research indicates that couples’ communication patterns after a diagnosis are similar to the style they had before the diagnosis.⁸ Also, our study’s findings may be affected by the stage of disease. Future prospective studies that measure psychological well-being of patients with prostate cancer and their partners need to consider the important issues raised by our study. In particular, the apparent contradiction of not communicating about fears and death while claiming marital satisfaction warrants further research. Larger studies will have to demonstrate the benefits of disclosure of feelings on adjustment before care providers are asked to contribute to open communication between spouses. Nevertheless, the findings of our research indicate that the inclusion of spouses in research expands our understanding of the effects of prostate cancer on well-being.

Lack of Communication

The data give no indication that men deal with cancer by discussing fears of death and dependency with their wives. Instead they express a desire to put the cancer behind them, “to be done with the disease.” The men presented themselves as having adjusted to the disease with a self-identity that is unchanged by prostate cancer. Their lack of communication about emotions received little challenge from their wives. Moreover, the wives collaborated in that they hid their own fears and despair instead of pursuing spousal communication about emotions. Thus, protective buffering is the prevalent coping style by both spouses, confirming the findings of earlier research that spouses’ coping styles and emotions are interrelated.^7,19,20

This is problematic for 2 reasons. Earlier research indicates that there is an inverse relationship between protective buffering and marital satisfaction, as well as a strong positive association between protective buffering and patient and spouse distress.⁷ The couples with metastatic prostate cancer claim that there is little change in their mates. They overwhelmingly portray themselves as happy and content with their marital relationship. Contrary to this depiction, references to moodiness and the emergence of jealousy point to strains on the marital relationship. Also, lack of communication leads to uncertainty about their spouse’s feelings and thoughts, another potential strain on the marital relationship. Although our study does not entail an assessment of men’s and wives individual or joint adjustment to prostate cancer, we certainly find evidence of the risk factors for poor adjustment which previous studies indicate: lack of communicating and protective buffering. Also, losing or abandoning one’s sex life, as in our sample, is a significant change that most likely has a strong association with psychological well-being, as found in earlier studies.²¹ This appears to suggest that the couples with metastatic prostate cancer may be at risk for distress and poor adjustment.

Conclusions

These insights into couples’ coping styles are of special relevance to physicians who care for men with prostate cancer. Generally, physicians are open to married men involving their wives in the management of prostate cancer. However, our findings raise questions about how involved wives really are. Although women are interested in their husbands’ prostate cancer, the lack of communication within the couple suggests that their active involvement may be less than is commonly assumed. Care providers may positively influence patients’ adaptation and quality of life by facilitating involvement by the patient’s wife, rather than assume that her presence alone signals active involvement. Also, physicians can suspect that patients who choose to go through treatment by themselves may be at risk for poor adjustment to their diagnosis.

Related Resources

• National Prostate Cancer Coalition http://www.4npcc.org/Advocates.htm/
• US TOO, International http://www.ustoo.com/
• Online Decision Support for Prostate Cancer http://www.cancerfacts.com/

· Acknowledgments ·

Support for our research was provided by US Army Medical Research and Materiel Command, Fort Detrick, Maryland, grant number DAMD17-99-1-9052, Principle Investigator, Dr Boehmer. The data collection was supported by the Department of Veterans Affairs Health Services Research and Development Service grant SDR-93-007.

Although the seemingly relentless increase in cesarean delivery rates in North America in the 1970s and 1980s has stabilized in recent years and even reversed slightly in some regions,¹ rates are still generally considered to be too high, and efforts to diminish them continue.² In spite of accumulating evidence about the preventable causes of cesarean delivery, individual institutions have had difficulty changing practice.³ At BC Women’s Hospital, the largest maternity facility in Canada with 7500 births annually, the cesarean delivery rate in 1995 was 22.9%.⁴ Although BC Women’s is the referral/tertiary center for the province of British Columbia, 80% to 85% of the births occur to low-risk women in the surrounding catchment area of Vancouver. Approximately 30 obstetricians and 100 family practice physicians have admitting privileges. At nearby Burnaby Hospital, a community general hospital providing primary- and secondary-level maternity care, cesarean delivery rates were 10.3%.⁴ At Burnaby, 3 obstetricians and 38 family practice physicians deliver 2000 babies annually. At the time of our study, neither hospital had practicing midwives. The discrepancy in cesarean delivery rates between hospitals only 20 minutes’ driving distance apart afforded us the opportunity to look at differences in practice between the institutions.

Methods

Phase I

In phase I we completed a retrospective cohort study using hospital records. Records were considered eligible if parturient women met the following criteria: nulliparous, maternal age 16 to 35 years, singleton gestation, and gestation of 37 completed weeks or longer. We excluded women with known pregnancy complications.

Charts were selected from a computer-generated list that incorporated inclusion and exclusion criteria and were then reviewed by a research nurse. Blinding as to hospital was not possible. Using this system, 430 nulliparous women from each hospital were selected consecutively. The sample size was calculated to be 430 per group to have 80% power with type I error set at 0.05, to determine a difference in cesarean delivery rate of 30% from a baseline of 23%. We examined the role of demographic and obstetric factors in the association between hospital and cesarean delivery rate. Maternal and fetal outcomes were measured, including postpartum hemorrhage and infection, as were APGAR scores at 5 minutes. We used logistic regression to simultaneously adjust for confounding factors.

Phase II

Since we were not able to randomly allocate women to hospitals, it was important to determine if women giving birth at BC Women’s compared with Burnaby selfselected to either of the hospitals because of differing expectations for pain management in labor. We also wanted to discern if there were aspects of the care given during the intrapartum period that might explain differing rates of use of epidural analgesia. In the phase II observation study, a research assistant was assigned to observe consecutive women giving birth 2 days per week at each hospital. Our research assistant was a medical student who had not practiced in either hospital. Structured observations were recorded along with times for each observation period and the stage of labor at which the observation took place. The same inclusion and exclusion criteria were used as in our phase I retrospective study to ensure that we were observing comparable women without preexisting risk factors for cesarean delivery at the onset of labor. The phase II study did, however, include multiparous women to maximize use of the time the student was available to assist with the study.

Analysis

We compared demographic characteristics and rates of selected outcomes between hospitals, using the chi-square statistic for categoric variables and the Student t test statistic for continuous variables. When expected cell frequencies were less than 2, we used the Fisher exact test for categoric variables instead of the chi-square statistic. The type I error, 2-sided a, was set at 0.05. Multivariate analysis was undertaken using unconditional logistic regression. We obtained maximum likelihood estimates of the odds ratios using the logistic model. We calculated 95% confidence intervals using the estimates of the standard error derived from the model.

Results

Phase I

Comparing cesarean delivery rates from our retrospective study of 857 births, the crude nulliparous rate at BC Women’s was 20.7%, while at Burnaby it was 6.7% (P <.0001). We examined a number of demographic and obstetric factors to determine if they would explain the reason for the increased rate at BC Women’s.

Women at Burnaby were younger, more likely to be single, and white (P <.0001, Table 1). A larger proportion of women at BC Women’s were Asian. Cesarean delivery rates did not differ between white and Asian women (9.1% vs 9.3%) at Burnaby, but there was a marked difference (18% vs 28.6%) at BC Women’s (data not shown). Women arrived earlier in labor at BC Women’s Table 2. The lengths of the first and second stages were longer at BC Women’s, as was the length of time membranes were ruptured. Time from admission to delivery was also significantly increased at BC Women’s.

Rates of obstetric interventions were compared between the 2 settings Table 2. Rate of induction of labor was not different between the 2 hospitals; however, augmentation of labor with oxytocin occurred significantly more frequently at Burnaby than at BC Women’s. Artificial rupture of membranes during labor was performed significantly more often at BC Women’s than at Burnaby. BC Women’s primarily used epidural analgesia, while Burnaby used intramuscular or intravenous administration of meperidine (Demerol). Use of Entonox (nitrous oxygen and oxygen) was also significantly different. Electronic fetal monitoring (EFM) was used for almost all patients at both hospitals; however, at Burnaby nurses were more likely to obtain only a baseline or admission fetal monitoring record, or to monitor intermittently. BC Women’s, on the other hand, was more likely to employ EFM continuously. Increased use of EFM both with and without epidural analgesia resulted in women being confined to bed more often, since remote methods (telemetry) of monitoring fetal heart rates were rarely used.

Overall rates of assisted vaginal deliveries at the 2 hospitals were not different, although forceps were used more frequently at BC Women’s and vacuum extraction at Burnaby. More frequent use of cesarean delivery at BC Women’s was not associated with an improvement in selected maternal/newborn outcomes (postpartum hemorrhage and 5-minute APGAR score <7). Postpartum infections were significantly more common at BC Women’s (P <.0001).

Data were analyzed for each hospital separately to determine all the factors contributing to cesarean delivery in each Table 3. After regressing all the demographic factors in a multiple regression model with cesarean delivery as the dependant variable for each hospital, only age remained significant. To evaluate the role of obstetric interventions for each hospital each intervention was entered separately, the most significant retained, and the process repeated with the next most significant retained until no other variables remained significantly associated with cesarean delivery. Age was retained in all the models.

In Burnaby, patient age, cervical dilatation on admission to the hospital, induction, epidural analgesia, and augmentation with oxytocin were important. At BC Women’s the same factors were important, as well as race/ethnicity (white vs nonwhite). Use of electronic fetal monitoring, Entonox, and intravenous or intramuscular narcotics did not predict cesarean delivery in either setting.

To compare the 2 hospitals in cesarean delivery rate we created a model combining data from both institutions. With both in the model, we could quantify the risk for cesarean delivery in one hospital compared with the other. First we examined the role of demographic factors in accounting for differences in cesarean delivery rate. Adjusting for age, the odds ratio (OR) for cesarean delivery at BC Women’s compared with Burnaby was 3.4 (95% confidence interval [CI], 2.1-5.4).

Cervical dilatation on admission, oxytocin augmentation, and epidural analgesia remained statistically significant in addition to age for predicting cesarean delivery. When age, augmentation, and cervical dilatation were retained in the model there remained a statistically significant difference between the 2 hospitals; the OR for BC Women’s versus Burnaby for cesarean delivery was 3.8 (95% CI, 2.3- 6.3). When we adjusted for epidural in addition, the OR for BC Women’s versus Burnaby dropped to 2.0 (95% CI, 1.09-3.5); the difference became only marginally statistically significant. The principal factor associated with the difference in cesarean delivery rate between the 2 hospitals, therefore, was use of epidural analgesia.

Phase II

Our sample consisted of 24 laboring women at Burnaby and 26 laboring women at BC Women’s. The groups were comparable in mean age, marital status, employment, and parity. There were more white women: 18 (75%) in the Burnaby group compared with 11 (42.3%) in the BC Women’s group.

Almost all the women were accompanied by support persons, 24 (100%) at Burnaby and 25 (96.2%) at BC Women’s. The majority at Burnaby had been to childbirth education classes (18 [78.3%] vs 14 [58.3%] at BC Women’s).

It was not possible to observe all the women throughout the entire length of labor. Given this reality and the fact that length of labor differed between women, observations related to management of labor were expressed as rates (ie, the number of minutes spent in bed divided by the total observation time in hours). Women in the 2 hospitals spent similar amounts of time in bed and in the shower. Women at Burnaby spent significantly more time walking, 12 minutes per hour compared with 4.7 minutes per hour at BC Women’s (P=.01, Table 4). There was no difference between the 2 hospital groups in the amount of physical contact with the laboring woman and her nurse or support person. At BC Women’s, women were more often accompanied by a nurse, physician, medical student, or resident and less often left alone than at Burnaby. Women at BC Women’s were exposed to greater numbers of different caregivers.

There were no differences between the groups in the number of couples who expressed specific expectations of labor Table 5. Ten women in Burnaby and 8 in BC Women’s were observed to ask for help with pain management, and responses to these requests included general encouragement or specific verbal advice as to what method they should use. Women were more frequently offered epidural analgesia at BC Women’s; there were no differences in rates of offers for other pain management options. This question related to a preemptive discussion about pain management. In terms of the pain management that actually took place, there was a significant difference in the number of doses of narcotic, in that no women at BC Women’s received narcotics. There was a trend toward more exposure to Entonox and epidural at BC Women’s, although these differences were not significant.

Discussion

Institutional differences in cesarean birth rates have previously been associated with differences in maternity care practices.⁵ The time during labor (state of cervical dilatation) at which women are admitted to the hospital has been associated with cesarean delivery rate.⁶

Age as a risk factor for cesarean delivery has also been well documented. The influence of age appears to be important independent of risk and remains important even among women without risk factors.^7-9 The reason for this is unknown but may point to an association with physician bias or age-related biological factors in the labor process that remain unmeasured.^10,11

The need for oxytocin augmentation has been associated with cesarean delivery.¹² Randomized controlled trials, however, have not demonstrated an association of early use of oxytocin augmentation with cesarean delivery reduction.¹³ Augmentation did not contribute significantly to the model examining differences in cesarean delivery between hospitals when an epidural was included.

A meta-analysis by Halpern and colleagues¹⁴ evaluated the association of epidural analgesia and cesarean delivery. In that report there was no increase in risk of cesarean delivery associated with use of epidural versus narcotic analgesia (OR=1.28; 95% CI, 0.55-2.93 for nulliparous women; OR=0.83; 95% CI, 0.22-3.15 for multiparous women). Findings from the intention-to-treat analysis that could not overcome the limitation of high rates of crossover or noncompliance among some of the trials^15-17 are in contrast with a protocol-compliant analysis in some of the studies that did show an effect of epidural on cesarean delivery rates.¹⁵ Although the protocol-compliant analysis provides important information, it does not necessarily indicate that epidural analgesia is the cause of higher cesarean rates. Subjects having more difficult labors may be more likely to cross over to epidural analgesia from the narcotic arm of the study, increasing the potential need for cesarean delivery in the epidural arm. In the meta-analysis the cesarean delivery rate among women receiving epidurals was 8.2%,¹⁴ much lower than at BC Women’s and more in line with those at Burnaby. The results, therefore, may not be generalizable to BC Women’s, at which high cesarean delivery rates would indicate greater opportunity for changes in intrapartum management, including use of epidural analgesia, to influence cesarean delivery rate.

An important aspect of our work addressed in phase II, our observation study, considered characteristics of women choosing delivery at either hospital that might also influence their choice to use an epidural. Although there were differences in ethnic background between the 2 study groups, these differences did not explain the difference in cesarean delivery rate between the 2 hospitals in our larger study. Although the differences were not statistically significant, more women at Burnaby appeared to have attended prenatal classes. The literature to date has not demonstrated an effect of prenatal education on cesarean delivery.¹⁸

Our observational study failed to suggest differences in client expectations between the 2 hospitals. Features of labor management did differ, however, in that women in Burnaby ambulated more, had fewer types and numbers of caregivers in labor and were less often offered an epidural. Nurses in Burnaby may have offered epidural less often because they were aware that anesthetists had to come to the delivery suite from elsewhere in the hospital or from outside the hospital. Alternatively, a cohort study has reported that nurses grouped according to cesarean rate quintiles differed in their recording of psychosocial data and other aspects of nursing care.¹⁹ Aspects of nursing practice may influence use of epidural analgesia. Our phase II study is limited by its small size, and other additional effects of nursing practice on labor cannot be ruled out. A large randomized controlled study is under way to examine the association of aspects of nursing care (including consistency of nursing caregiver) with cesarean delivery rates.²⁰ Ultimately, the decision to undertake a cesarean delivery resides with the obstetrician. It is possible that in the smaller community hospital physicians may practice with a greater degree of cohesion, perhaps influenced more readily by the philosophy of opinion leaders who advocate a more conservative approach to cesarean delivery. In addition, the presence of obstetric and family practice residency programs at BC Women’s may have encouraged use of interventions, including cesarean the delivery.

The role of ambulation in cesarean delivery remains controversial. A randomized trial failed to demonstrate an association of ambulation with cesarean delivery, but this study had a 22% crossover rate and enrolled women after they had attained a cervical dilatation rate of 3 to 5 cm, when early ambulation may have already exerted a positive effect.²¹ Among low-risk parturients cared for by midwives and not requiring either augmentation of labor or epidural analgesia, ambulation has been associated with a reduction in cesarean delivery rates.²²

Limitations

Our study is limited by lack of knowledge of women’s levels of anxiety and pain during their labor. This would have allowed us to gauge whether less frequent use of epidural analgesia at Burnaby was associated with a cost of diminished satisfaction with the childbirth experience.

We were fortunate in being able to study 2 institutions that were only 20 minutes’ driving distance apart and that served populations from which we were able to sample women who were demographically comparable. Our retrospective cohort analysis identified use of an epidural as the main predictive factor differentiating cesarean delivery rates between the hospitals. It is not possible to determine cause and effect from this retrospective study design, however, and use of epidural analgesia may be a proxy for other unmeasured variables, such as physician practice, anxiety level among patients, or level of education. Direct observation of intrapartum care in a follow-up study failed to differentiate clients attending either hospital in terms of preparation for or expectations of the labor experience. Differences in some aspects of caregiving, however, including more frequent offering of epidural for pain management, may explain the increased use of epidural at BC Women’s. Factors influencing use of epidural need to be studied more thoroughly to support its appropriate place in an overall strategy for pain management in labor.

Related Resources

• Society of Obstetricians and Gynaecologists of Canada (SOGC) http://www.sogc.org/SOGCnet
• College of Family Physicians of Canada www.cfpc.ca
• Canadian Paediatric Society www.cps.ca

Although the seemingly relentless increase in cesarean delivery rates in North America in the 1970s and 1980s has stabilized in recent years and even reversed slightly in some regions,¹ rates are still generally considered to be too high, and efforts to diminish them continue.² In spite of accumulating evidence about the preventable causes of cesarean delivery, individual institutions have had difficulty changing practice.³ At BC Women’s Hospital, the largest maternity facility in Canada with 7500 births annually, the cesarean delivery rate in 1995 was 22.9%.⁴ Although BC Women’s is the referral/tertiary center for the province of British Columbia, 80% to 85% of the births occur to low-risk women in the surrounding catchment area of Vancouver. Approximately 30 obstetricians and 100 family practice physicians have admitting privileges. At nearby Burnaby Hospital, a community general hospital providing primary- and secondary-level maternity care, cesarean delivery rates were 10.3%.⁴ At Burnaby, 3 obstetricians and 38 family practice physicians deliver 2000 babies annually. At the time of our study, neither hospital had practicing midwives. The discrepancy in cesarean delivery rates between hospitals only 20 minutes’ driving distance apart afforded us the opportunity to look at differences in practice between the institutions.

Methods

Phase I

In phase I we completed a retrospective cohort study using hospital records. Records were considered eligible if parturient women met the following criteria: nulliparous, maternal age 16 to 35 years, singleton gestation, and gestation of 37 completed weeks or longer. We excluded women with known pregnancy complications.

Charts were selected from a computer-generated list that incorporated inclusion and exclusion criteria and were then reviewed by a research nurse. Blinding as to hospital was not possible. Using this system, 430 nulliparous women from each hospital were selected consecutively. The sample size was calculated to be 430 per group to have 80% power with type I error set at 0.05, to determine a difference in cesarean delivery rate of 30% from a baseline of 23%. We examined the role of demographic and obstetric factors in the association between hospital and cesarean delivery rate. Maternal and fetal outcomes were measured, including postpartum hemorrhage and infection, as were APGAR scores at 5 minutes. We used logistic regression to simultaneously adjust for confounding factors.

Phase II

Since we were not able to randomly allocate women to hospitals, it was important to determine if women giving birth at BC Women’s compared with Burnaby selfselected to either of the hospitals because of differing expectations for pain management in labor. We also wanted to discern if there were aspects of the care given during the intrapartum period that might explain differing rates of use of epidural analgesia. In the phase II observation study, a research assistant was assigned to observe consecutive women giving birth 2 days per week at each hospital. Our research assistant was a medical student who had not practiced in either hospital. Structured observations were recorded along with times for each observation period and the stage of labor at which the observation took place. The same inclusion and exclusion criteria were used as in our phase I retrospective study to ensure that we were observing comparable women without preexisting risk factors for cesarean delivery at the onset of labor. The phase II study did, however, include multiparous women to maximize use of the time the student was available to assist with the study.

Analysis

We compared demographic characteristics and rates of selected outcomes between hospitals, using the chi-square statistic for categoric variables and the Student t test statistic for continuous variables. When expected cell frequencies were less than 2, we used the Fisher exact test for categoric variables instead of the chi-square statistic. The type I error, 2-sided a, was set at 0.05. Multivariate analysis was undertaken using unconditional logistic regression. We obtained maximum likelihood estimates of the odds ratios using the logistic model. We calculated 95% confidence intervals using the estimates of the standard error derived from the model.

Results

Phase I

Comparing cesarean delivery rates from our retrospective study of 857 births, the crude nulliparous rate at BC Women’s was 20.7%, while at Burnaby it was 6.7% (P <.0001). We examined a number of demographic and obstetric factors to determine if they would explain the reason for the increased rate at BC Women’s.

Women at Burnaby were younger, more likely to be single, and white (P <.0001, Table 1). A larger proportion of women at BC Women’s were Asian. Cesarean delivery rates did not differ between white and Asian women (9.1% vs 9.3%) at Burnaby, but there was a marked difference (18% vs 28.6%) at BC Women’s (data not shown). Women arrived earlier in labor at BC Women’s Table 2. The lengths of the first and second stages were longer at BC Women’s, as was the length of time membranes were ruptured. Time from admission to delivery was also significantly increased at BC Women’s.

Rates of obstetric interventions were compared between the 2 settings Table 2. Rate of induction of labor was not different between the 2 hospitals; however, augmentation of labor with oxytocin occurred significantly more frequently at Burnaby than at BC Women’s. Artificial rupture of membranes during labor was performed significantly more often at BC Women’s than at Burnaby. BC Women’s primarily used epidural analgesia, while Burnaby used intramuscular or intravenous administration of meperidine (Demerol). Use of Entonox (nitrous oxygen and oxygen) was also significantly different. Electronic fetal monitoring (EFM) was used for almost all patients at both hospitals; however, at Burnaby nurses were more likely to obtain only a baseline or admission fetal monitoring record, or to monitor intermittently. BC Women’s, on the other hand, was more likely to employ EFM continuously. Increased use of EFM both with and without epidural analgesia resulted in women being confined to bed more often, since remote methods (telemetry) of monitoring fetal heart rates were rarely used.

Overall rates of assisted vaginal deliveries at the 2 hospitals were not different, although forceps were used more frequently at BC Women’s and vacuum extraction at Burnaby. More frequent use of cesarean delivery at BC Women’s was not associated with an improvement in selected maternal/newborn outcomes (postpartum hemorrhage and 5-minute APGAR score <7). Postpartum infections were significantly more common at BC Women’s (P <.0001).

Data were analyzed for each hospital separately to determine all the factors contributing to cesarean delivery in each Table 3. After regressing all the demographic factors in a multiple regression model with cesarean delivery as the dependant variable for each hospital, only age remained significant. To evaluate the role of obstetric interventions for each hospital each intervention was entered separately, the most significant retained, and the process repeated with the next most significant retained until no other variables remained significantly associated with cesarean delivery. Age was retained in all the models.

In Burnaby, patient age, cervical dilatation on admission to the hospital, induction, epidural analgesia, and augmentation with oxytocin were important. At BC Women’s the same factors were important, as well as race/ethnicity (white vs nonwhite). Use of electronic fetal monitoring, Entonox, and intravenous or intramuscular narcotics did not predict cesarean delivery in either setting.

To compare the 2 hospitals in cesarean delivery rate we created a model combining data from both institutions. With both in the model, we could quantify the risk for cesarean delivery in one hospital compared with the other. First we examined the role of demographic factors in accounting for differences in cesarean delivery rate. Adjusting for age, the odds ratio (OR) for cesarean delivery at BC Women’s compared with Burnaby was 3.4 (95% confidence interval [CI], 2.1-5.4).

Cervical dilatation on admission, oxytocin augmentation, and epidural analgesia remained statistically significant in addition to age for predicting cesarean delivery. When age, augmentation, and cervical dilatation were retained in the model there remained a statistically significant difference between the 2 hospitals; the OR for BC Women’s versus Burnaby for cesarean delivery was 3.8 (95% CI, 2.3- 6.3). When we adjusted for epidural in addition, the OR for BC Women’s versus Burnaby dropped to 2.0 (95% CI, 1.09-3.5); the difference became only marginally statistically significant. The principal factor associated with the difference in cesarean delivery rate between the 2 hospitals, therefore, was use of epidural analgesia.

Phase II

Our sample consisted of 24 laboring women at Burnaby and 26 laboring women at BC Women’s. The groups were comparable in mean age, marital status, employment, and parity. There were more white women: 18 (75%) in the Burnaby group compared with 11 (42.3%) in the BC Women’s group.

Almost all the women were accompanied by support persons, 24 (100%) at Burnaby and 25 (96.2%) at BC Women’s. The majority at Burnaby had been to childbirth education classes (18 [78.3%] vs 14 [58.3%] at BC Women’s).

It was not possible to observe all the women throughout the entire length of labor. Given this reality and the fact that length of labor differed between women, observations related to management of labor were expressed as rates (ie, the number of minutes spent in bed divided by the total observation time in hours). Women in the 2 hospitals spent similar amounts of time in bed and in the shower. Women at Burnaby spent significantly more time walking, 12 minutes per hour compared with 4.7 minutes per hour at BC Women’s (P=.01, Table 4). There was no difference between the 2 hospital groups in the amount of physical contact with the laboring woman and her nurse or support person. At BC Women’s, women were more often accompanied by a nurse, physician, medical student, or resident and less often left alone than at Burnaby. Women at BC Women’s were exposed to greater numbers of different caregivers.

There were no differences between the groups in the number of couples who expressed specific expectations of labor Table 5. Ten women in Burnaby and 8 in BC Women’s were observed to ask for help with pain management, and responses to these requests included general encouragement or specific verbal advice as to what method they should use. Women were more frequently offered epidural analgesia at BC Women’s; there were no differences in rates of offers for other pain management options. This question related to a preemptive discussion about pain management. In terms of the pain management that actually took place, there was a significant difference in the number of doses of narcotic, in that no women at BC Women’s received narcotics. There was a trend toward more exposure to Entonox and epidural at BC Women’s, although these differences were not significant.

Discussion

Institutional differences in cesarean birth rates have previously been associated with differences in maternity care practices.⁵ The time during labor (state of cervical dilatation) at which women are admitted to the hospital has been associated with cesarean delivery rate.⁶

Age as a risk factor for cesarean delivery has also been well documented. The influence of age appears to be important independent of risk and remains important even among women without risk factors.^7-9 The reason for this is unknown but may point to an association with physician bias or age-related biological factors in the labor process that remain unmeasured.^10,11

The need for oxytocin augmentation has been associated with cesarean delivery.¹² Randomized controlled trials, however, have not demonstrated an association of early use of oxytocin augmentation with cesarean delivery reduction.¹³ Augmentation did not contribute significantly to the model examining differences in cesarean delivery between hospitals when an epidural was included.

A meta-analysis by Halpern and colleagues¹⁴ evaluated the association of epidural analgesia and cesarean delivery. In that report there was no increase in risk of cesarean delivery associated with use of epidural versus narcotic analgesia (OR=1.28; 95% CI, 0.55-2.93 for nulliparous women; OR=0.83; 95% CI, 0.22-3.15 for multiparous women). Findings from the intention-to-treat analysis that could not overcome the limitation of high rates of crossover or noncompliance among some of the trials^15-17 are in contrast with a protocol-compliant analysis in some of the studies that did show an effect of epidural on cesarean delivery rates.¹⁵ Although the protocol-compliant analysis provides important information, it does not necessarily indicate that epidural analgesia is the cause of higher cesarean rates. Subjects having more difficult labors may be more likely to cross over to epidural analgesia from the narcotic arm of the study, increasing the potential need for cesarean delivery in the epidural arm. In the meta-analysis the cesarean delivery rate among women receiving epidurals was 8.2%,¹⁴ much lower than at BC Women’s and more in line with those at Burnaby. The results, therefore, may not be generalizable to BC Women’s, at which high cesarean delivery rates would indicate greater opportunity for changes in intrapartum management, including use of epidural analgesia, to influence cesarean delivery rate.

An important aspect of our work addressed in phase II, our observation study, considered characteristics of women choosing delivery at either hospital that might also influence their choice to use an epidural. Although there were differences in ethnic background between the 2 study groups, these differences did not explain the difference in cesarean delivery rate between the 2 hospitals in our larger study. Although the differences were not statistically significant, more women at Burnaby appeared to have attended prenatal classes. The literature to date has not demonstrated an effect of prenatal education on cesarean delivery.¹⁸

Our observational study failed to suggest differences in client expectations between the 2 hospitals. Features of labor management did differ, however, in that women in Burnaby ambulated more, had fewer types and numbers of caregivers in labor and were less often offered an epidural. Nurses in Burnaby may have offered epidural less often because they were aware that anesthetists had to come to the delivery suite from elsewhere in the hospital or from outside the hospital. Alternatively, a cohort study has reported that nurses grouped according to cesarean rate quintiles differed in their recording of psychosocial data and other aspects of nursing care.¹⁹ Aspects of nursing practice may influence use of epidural analgesia. Our phase II study is limited by its small size, and other additional effects of nursing practice on labor cannot be ruled out. A large randomized controlled study is under way to examine the association of aspects of nursing care (including consistency of nursing caregiver) with cesarean delivery rates.²⁰ Ultimately, the decision to undertake a cesarean delivery resides with the obstetrician. It is possible that in the smaller community hospital physicians may practice with a greater degree of cohesion, perhaps influenced more readily by the philosophy of opinion leaders who advocate a more conservative approach to cesarean delivery. In addition, the presence of obstetric and family practice residency programs at BC Women’s may have encouraged use of interventions, including cesarean the delivery.

The role of ambulation in cesarean delivery remains controversial. A randomized trial failed to demonstrate an association of ambulation with cesarean delivery, but this study had a 22% crossover rate and enrolled women after they had attained a cervical dilatation rate of 3 to 5 cm, when early ambulation may have already exerted a positive effect.²¹ Among low-risk parturients cared for by midwives and not requiring either augmentation of labor or epidural analgesia, ambulation has been associated with a reduction in cesarean delivery rates.²²

Limitations

Our study is limited by lack of knowledge of women’s levels of anxiety and pain during their labor. This would have allowed us to gauge whether less frequent use of epidural analgesia at Burnaby was associated with a cost of diminished satisfaction with the childbirth experience.

We were fortunate in being able to study 2 institutions that were only 20 minutes’ driving distance apart and that served populations from which we were able to sample women who were demographically comparable. Our retrospective cohort analysis identified use of an epidural as the main predictive factor differentiating cesarean delivery rates between the hospitals. It is not possible to determine cause and effect from this retrospective study design, however, and use of epidural analgesia may be a proxy for other unmeasured variables, such as physician practice, anxiety level among patients, or level of education. Direct observation of intrapartum care in a follow-up study failed to differentiate clients attending either hospital in terms of preparation for or expectations of the labor experience. Differences in some aspects of caregiving, however, including more frequent offering of epidural for pain management, may explain the increased use of epidural at BC Women’s. Factors influencing use of epidural need to be studied more thoroughly to support its appropriate place in an overall strategy for pain management in labor.

Related Resources

• Society of Obstetricians and Gynaecologists of Canada (SOGC) http://www.sogc.org/SOGCnet
• College of Family Physicians of Canada www.cfpc.ca
• Canadian Paediatric Society www.cps.ca

Although the seemingly relentless increase in cesarean delivery rates in North America in the 1970s and 1980s has stabilized in recent years and even reversed slightly in some regions,¹ rates are still generally considered to be too high, and efforts to diminish them continue.² In spite of accumulating evidence about the preventable causes of cesarean delivery, individual institutions have had difficulty changing practice.³ At BC Women’s Hospital, the largest maternity facility in Canada with 7500 births annually, the cesarean delivery rate in 1995 was 22.9%.⁴ Although BC Women’s is the referral/tertiary center for the province of British Columbia, 80% to 85% of the births occur to low-risk women in the surrounding catchment area of Vancouver. Approximately 30 obstetricians and 100 family practice physicians have admitting privileges. At nearby Burnaby Hospital, a community general hospital providing primary- and secondary-level maternity care, cesarean delivery rates were 10.3%.⁴ At Burnaby, 3 obstetricians and 38 family practice physicians deliver 2000 babies annually. At the time of our study, neither hospital had practicing midwives. The discrepancy in cesarean delivery rates between hospitals only 20 minutes’ driving distance apart afforded us the opportunity to look at differences in practice between the institutions.

Methods

Phase I

In phase I we completed a retrospective cohort study using hospital records. Records were considered eligible if parturient women met the following criteria: nulliparous, maternal age 16 to 35 years, singleton gestation, and gestation of 37 completed weeks or longer. We excluded women with known pregnancy complications.

Charts were selected from a computer-generated list that incorporated inclusion and exclusion criteria and were then reviewed by a research nurse. Blinding as to hospital was not possible. Using this system, 430 nulliparous women from each hospital were selected consecutively. The sample size was calculated to be 430 per group to have 80% power with type I error set at 0.05, to determine a difference in cesarean delivery rate of 30% from a baseline of 23%. We examined the role of demographic and obstetric factors in the association between hospital and cesarean delivery rate. Maternal and fetal outcomes were measured, including postpartum hemorrhage and infection, as were APGAR scores at 5 minutes. We used logistic regression to simultaneously adjust for confounding factors.

Phase II

Since we were not able to randomly allocate women to hospitals, it was important to determine if women giving birth at BC Women’s compared with Burnaby selfselected to either of the hospitals because of differing expectations for pain management in labor. We also wanted to discern if there were aspects of the care given during the intrapartum period that might explain differing rates of use of epidural analgesia. In the phase II observation study, a research assistant was assigned to observe consecutive women giving birth 2 days per week at each hospital. Our research assistant was a medical student who had not practiced in either hospital. Structured observations were recorded along with times for each observation period and the stage of labor at which the observation took place. The same inclusion and exclusion criteria were used as in our phase I retrospective study to ensure that we were observing comparable women without preexisting risk factors for cesarean delivery at the onset of labor. The phase II study did, however, include multiparous women to maximize use of the time the student was available to assist with the study.

Analysis

We compared demographic characteristics and rates of selected outcomes between hospitals, using the chi-square statistic for categoric variables and the Student t test statistic for continuous variables. When expected cell frequencies were less than 2, we used the Fisher exact test for categoric variables instead of the chi-square statistic. The type I error, 2-sided a, was set at 0.05. Multivariate analysis was undertaken using unconditional logistic regression. We obtained maximum likelihood estimates of the odds ratios using the logistic model. We calculated 95% confidence intervals using the estimates of the standard error derived from the model.

Results

Phase I

Comparing cesarean delivery rates from our retrospective study of 857 births, the crude nulliparous rate at BC Women’s was 20.7%, while at Burnaby it was 6.7% (P <.0001). We examined a number of demographic and obstetric factors to determine if they would explain the reason for the increased rate at BC Women’s.

Women at Burnaby were younger, more likely to be single, and white (P <.0001, Table 1). A larger proportion of women at BC Women’s were Asian. Cesarean delivery rates did not differ between white and Asian women (9.1% vs 9.3%) at Burnaby, but there was a marked difference (18% vs 28.6%) at BC Women’s (data not shown). Women arrived earlier in labor at BC Women’s Table 2. The lengths of the first and second stages were longer at BC Women’s, as was the length of time membranes were ruptured. Time from admission to delivery was also significantly increased at BC Women’s.

Rates of obstetric interventions were compared between the 2 settings Table 2. Rate of induction of labor was not different between the 2 hospitals; however, augmentation of labor with oxytocin occurred significantly more frequently at Burnaby than at BC Women’s. Artificial rupture of membranes during labor was performed significantly more often at BC Women’s than at Burnaby. BC Women’s primarily used epidural analgesia, while Burnaby used intramuscular or intravenous administration of meperidine (Demerol). Use of Entonox (nitrous oxygen and oxygen) was also significantly different. Electronic fetal monitoring (EFM) was used for almost all patients at both hospitals; however, at Burnaby nurses were more likely to obtain only a baseline or admission fetal monitoring record, or to monitor intermittently. BC Women’s, on the other hand, was more likely to employ EFM continuously. Increased use of EFM both with and without epidural analgesia resulted in women being confined to bed more often, since remote methods (telemetry) of monitoring fetal heart rates were rarely used.

Overall rates of assisted vaginal deliveries at the 2 hospitals were not different, although forceps were used more frequently at BC Women’s and vacuum extraction at Burnaby. More frequent use of cesarean delivery at BC Women’s was not associated with an improvement in selected maternal/newborn outcomes (postpartum hemorrhage and 5-minute APGAR score <7). Postpartum infections were significantly more common at BC Women’s (P <.0001).

Data were analyzed for each hospital separately to determine all the factors contributing to cesarean delivery in each Table 3. After regressing all the demographic factors in a multiple regression model with cesarean delivery as the dependant variable for each hospital, only age remained significant. To evaluate the role of obstetric interventions for each hospital each intervention was entered separately, the most significant retained, and the process repeated with the next most significant retained until no other variables remained significantly associated with cesarean delivery. Age was retained in all the models.

In Burnaby, patient age, cervical dilatation on admission to the hospital, induction, epidural analgesia, and augmentation with oxytocin were important. At BC Women’s the same factors were important, as well as race/ethnicity (white vs nonwhite). Use of electronic fetal monitoring, Entonox, and intravenous or intramuscular narcotics did not predict cesarean delivery in either setting.

To compare the 2 hospitals in cesarean delivery rate we created a model combining data from both institutions. With both in the model, we could quantify the risk for cesarean delivery in one hospital compared with the other. First we examined the role of demographic factors in accounting for differences in cesarean delivery rate. Adjusting for age, the odds ratio (OR) for cesarean delivery at BC Women’s compared with Burnaby was 3.4 (95% confidence interval [CI], 2.1-5.4).

Cervical dilatation on admission, oxytocin augmentation, and epidural analgesia remained statistically significant in addition to age for predicting cesarean delivery. When age, augmentation, and cervical dilatation were retained in the model there remained a statistically significant difference between the 2 hospitals; the OR for BC Women’s versus Burnaby for cesarean delivery was 3.8 (95% CI, 2.3- 6.3). When we adjusted for epidural in addition, the OR for BC Women’s versus Burnaby dropped to 2.0 (95% CI, 1.09-3.5); the difference became only marginally statistically significant. The principal factor associated with the difference in cesarean delivery rate between the 2 hospitals, therefore, was use of epidural analgesia.

Phase II

Our sample consisted of 24 laboring women at Burnaby and 26 laboring women at BC Women’s. The groups were comparable in mean age, marital status, employment, and parity. There were more white women: 18 (75%) in the Burnaby group compared with 11 (42.3%) in the BC Women’s group.

Almost all the women were accompanied by support persons, 24 (100%) at Burnaby and 25 (96.2%) at BC Women’s. The majority at Burnaby had been to childbirth education classes (18 [78.3%] vs 14 [58.3%] at BC Women’s).

It was not possible to observe all the women throughout the entire length of labor. Given this reality and the fact that length of labor differed between women, observations related to management of labor were expressed as rates (ie, the number of minutes spent in bed divided by the total observation time in hours). Women in the 2 hospitals spent similar amounts of time in bed and in the shower. Women at Burnaby spent significantly more time walking, 12 minutes per hour compared with 4.7 minutes per hour at BC Women’s (P=.01, Table 4). There was no difference between the 2 hospital groups in the amount of physical contact with the laboring woman and her nurse or support person. At BC Women’s, women were more often accompanied by a nurse, physician, medical student, or resident and less often left alone than at Burnaby. Women at BC Women’s were exposed to greater numbers of different caregivers.

There were no differences between the groups in the number of couples who expressed specific expectations of labor Table 5. Ten women in Burnaby and 8 in BC Women’s were observed to ask for help with pain management, and responses to these requests included general encouragement or specific verbal advice as to what method they should use. Women were more frequently offered epidural analgesia at BC Women’s; there were no differences in rates of offers for other pain management options. This question related to a preemptive discussion about pain management. In terms of the pain management that actually took place, there was a significant difference in the number of doses of narcotic, in that no women at BC Women’s received narcotics. There was a trend toward more exposure to Entonox and epidural at BC Women’s, although these differences were not significant.

Discussion

Institutional differences in cesarean birth rates have previously been associated with differences in maternity care practices.⁵ The time during labor (state of cervical dilatation) at which women are admitted to the hospital has been associated with cesarean delivery rate.⁶

Age as a risk factor for cesarean delivery has also been well documented. The influence of age appears to be important independent of risk and remains important even among women without risk factors.^7-9 The reason for this is unknown but may point to an association with physician bias or age-related biological factors in the labor process that remain unmeasured.^10,11

The need for oxytocin augmentation has been associated with cesarean delivery.¹² Randomized controlled trials, however, have not demonstrated an association of early use of oxytocin augmentation with cesarean delivery reduction.¹³ Augmentation did not contribute significantly to the model examining differences in cesarean delivery between hospitals when an epidural was included.

A meta-analysis by Halpern and colleagues¹⁴ evaluated the association of epidural analgesia and cesarean delivery. In that report there was no increase in risk of cesarean delivery associated with use of epidural versus narcotic analgesia (OR=1.28; 95% CI, 0.55-2.93 for nulliparous women; OR=0.83; 95% CI, 0.22-3.15 for multiparous women). Findings from the intention-to-treat analysis that could not overcome the limitation of high rates of crossover or noncompliance among some of the trials^15-17 are in contrast with a protocol-compliant analysis in some of the studies that did show an effect of epidural on cesarean delivery rates.¹⁵ Although the protocol-compliant analysis provides important information, it does not necessarily indicate that epidural analgesia is the cause of higher cesarean rates. Subjects having more difficult labors may be more likely to cross over to epidural analgesia from the narcotic arm of the study, increasing the potential need for cesarean delivery in the epidural arm. In the meta-analysis the cesarean delivery rate among women receiving epidurals was 8.2%,¹⁴ much lower than at BC Women’s and more in line with those at Burnaby. The results, therefore, may not be generalizable to BC Women’s, at which high cesarean delivery rates would indicate greater opportunity for changes in intrapartum management, including use of epidural analgesia, to influence cesarean delivery rate.

An important aspect of our work addressed in phase II, our observation study, considered characteristics of women choosing delivery at either hospital that might also influence their choice to use an epidural. Although there were differences in ethnic background between the 2 study groups, these differences did not explain the difference in cesarean delivery rate between the 2 hospitals in our larger study. Although the differences were not statistically significant, more women at Burnaby appeared to have attended prenatal classes. The literature to date has not demonstrated an effect of prenatal education on cesarean delivery.¹⁸

Our observational study failed to suggest differences in client expectations between the 2 hospitals. Features of labor management did differ, however, in that women in Burnaby ambulated more, had fewer types and numbers of caregivers in labor and were less often offered an epidural. Nurses in Burnaby may have offered epidural less often because they were aware that anesthetists had to come to the delivery suite from elsewhere in the hospital or from outside the hospital. Alternatively, a cohort study has reported that nurses grouped according to cesarean rate quintiles differed in their recording of psychosocial data and other aspects of nursing care.¹⁹ Aspects of nursing practice may influence use of epidural analgesia. Our phase II study is limited by its small size, and other additional effects of nursing practice on labor cannot be ruled out. A large randomized controlled study is under way to examine the association of aspects of nursing care (including consistency of nursing caregiver) with cesarean delivery rates.²⁰ Ultimately, the decision to undertake a cesarean delivery resides with the obstetrician. It is possible that in the smaller community hospital physicians may practice with a greater degree of cohesion, perhaps influenced more readily by the philosophy of opinion leaders who advocate a more conservative approach to cesarean delivery. In addition, the presence of obstetric and family practice residency programs at BC Women’s may have encouraged use of interventions, including cesarean the delivery.

The role of ambulation in cesarean delivery remains controversial. A randomized trial failed to demonstrate an association of ambulation with cesarean delivery, but this study had a 22% crossover rate and enrolled women after they had attained a cervical dilatation rate of 3 to 5 cm, when early ambulation may have already exerted a positive effect.²¹ Among low-risk parturients cared for by midwives and not requiring either augmentation of labor or epidural analgesia, ambulation has been associated with a reduction in cesarean delivery rates.²²

Limitations

Our study is limited by lack of knowledge of women’s levels of anxiety and pain during their labor. This would have allowed us to gauge whether less frequent use of epidural analgesia at Burnaby was associated with a cost of diminished satisfaction with the childbirth experience.

We were fortunate in being able to study 2 institutions that were only 20 minutes’ driving distance apart and that served populations from which we were able to sample women who were demographically comparable. Our retrospective cohort analysis identified use of an epidural as the main predictive factor differentiating cesarean delivery rates between the hospitals. It is not possible to determine cause and effect from this retrospective study design, however, and use of epidural analgesia may be a proxy for other unmeasured variables, such as physician practice, anxiety level among patients, or level of education. Direct observation of intrapartum care in a follow-up study failed to differentiate clients attending either hospital in terms of preparation for or expectations of the labor experience. Differences in some aspects of caregiving, however, including more frequent offering of epidural for pain management, may explain the increased use of epidural at BC Women’s. Factors influencing use of epidural need to be studied more thoroughly to support its appropriate place in an overall strategy for pain management in labor.

Related Resources

• Society of Obstetricians and Gynaecologists of Canada (SOGC) http://www.sogc.org/SOGCnet
• College of Family Physicians of Canada www.cfpc.ca
• Canadian Paediatric Society www.cps.ca

Primary care disciplines continue to have a central role in the health care of Americans. They provide breadth of care within an ongoing relationship, bridging the boundaries between health and illness and guiding access to more narrowly focused care when needed.¹ The ability to orchestrate a broad health agenda during a visit is central to primary care, but this ability is challenged by competing demands for time.²

Attempts to influence provision of care and treatment decisions by primary care physicians, such as financial incentives, administrative restrictions, and the implementation of evidence-based clinical guidelines add to the demands on physicians’ time and may affect how time is allocated during the day and with each patient. Within this context a primary care physician must prioritize the agenda for each patient visit. This may include providing services beyond the patient’s primary reason for the visit as time permits, such as including preventive services,³ follow-up of acute or chronic illnesses,¹ mental health⁴ or family issues,^5-7 or investigating “by the way” patient comments that may indicate serious medical issues.

The competing demands for time are compounded by patient requests during the visit. Based on an audiotape of 139 patient encounters, Kravitz and colleagues⁸ reported that on average a patient makes 5 requests for physician action or information per visit, and the number of unfulfilled requests was negatively associated with patient satisfaction. Such findings may fuel a sense of pressure to address patient requests. Also, another recent report indicates that the majority of patients do not have the opportunity to express all of their concerns before the physician redirects the interview; once redirected, additional patient concerns are rarely elicited.⁹ Fitting both the physician’s and patient’s agenda into the time allotted for an outpatient visit has important implications for the duration of the visit, physician productivity, and possibly patient outcomes.

Data on the number of problems raised and addressed have been limited by the lack of appropriate collection methods. Primarily audio and video technology have been used for the study of physician-patient communication.^10-12 Direct observation of patient encounters^12,13 and incorporation of ethnographic approaches have more recently been employed to fill a large void in the understanding of the content, context, and complexity of primary care.^13-15 Findings from the Direct Observation of Primary Care study, which employed such methods, indicate that among 4454 patient visits care was provided to a secondary patient during 18% of the visits and preventive services were addressed during 32% of the illness visits.³ Data from that study provide a glimpse into some types of problems addressed in addition to the main reason for the visit; however, data about the number of problems addressed during patient encounters were not specifically collected by the nurse observer.

When additional issues are raised during a patient encounter, little is known about the nature of these problems, how additional problems affect the duration of the visit, and how well additional problems are reflected in the billing record. This led us to conduct an observational study to ask: How many problems are addressed during family practice outpatient visits, and who is raising additional problems? How much work and time is associated with addressing problems raised beyond the initial problem? How well does the billing list represent the number of problems addressed during the outpatient visit? Our study was designed to directly observe and record how many problems were raised and addressed during outpatient visits to family physicians.

Methods

Seven first-year medical students observed patient care provided by their summer fellowship family physician preceptor and other physicians in the preceptor’s practice from June through August 1999. Six of the sites were located in Northeast Ohio, and one was in Tulsa, Oklahoma.

Each student collected data on one randomly selected adult patient encounter for each half day of precepting. At the beginning of each half-day of patient care the student rolled a die to generate a random number to select a patient from the patient schedule. To ensure random selection of encounters within each half-day session, on alternating days the random number was counted from the beginning or the end of the half-day schedule. If the selected patient was aged younger than 18 years, the patient or physician preferred the encounter not be observed, or the patient did not show up for the scheduled appointment, the next scheduled appointment was selected as a replacement. Patient age and sex were collected for those who were no-shows or chose not to be observed, so they could be compared with those patients who were observed. Each student was to collect data on approximately 50 patient encounters during the 6-week summer fellowship. The physicians were blinded to the study purpose and were not told which patient encounter would be included in the study.

A problem was operationalized as an issue requiring physician action in the form of a decision, diagnosis, treatment, or monitoring. Each item was listed as it was raised, and the type of problem, who raised it, and what physician actions were involved to address it were coded. Each problem was coded as 1 of 14 categories: acute, acute follow-up, chronic, chronic follow-up, prevention, prevention follow-up, psychosocial, psychosocial follow-up, work-related administrative, health care system-related administrative, other family member’s problem, pregnancy, emergent, and other. The person who raised the problem was coded as 1 of 3 options: the physician, the patient or another person in the room. Multiple physician actions could be coded for how the problem was addressed. The 19 physician action categories included: question, reassurance, examination, procedure, referral, return visit, advice, review tests, order laboratory testing, prescription, provide written material, imaging, admits uncertainty, counseling, return to work/time off work letter, defer, complementary/alternative medicine, ignored or lost, and other.

Patient characteristics, the duration of the visit, and the billing diagnoses for each visit were also recorded on the data collection form. Videotaped encounters were used to pilot test the data collection form, to allow the observers to practice using the form in real time, and to calibrate the observers before data collection in the field.

We used descriptive statistics to address most research questions. Student t tests and chi-square tests were used to compare age and sex differences between participants and nonparticipants. We tested the association of the number of problems with the duration of the visit with analysis of variance and a test for linear trend. A difference score of the number of problems observed and the number of problems recorded on the billing sheet for the encounter was computed and summarized graphically.

Results

We collected usable data on 266 encounters representing 37 physicians. Patient and visit characteristics are displayed in Table 1. The patients had an average age of 48 years, and 69% were women. They were predominately white. A large proportion was observed visiting their regular primary care physician (83%), and 85% were established patients of the practice. Most of the observed patients had some kind of commercial health care insurance, 19% had Medicare, and a small proportion had Medicaid or no insurance. The visit duration ranged from 2 to 65 minutes; the median was 15 minutes with a mean of 19.3 (standard deviation [SD]=12.7). The first problem raised was most commonly an acute problem (49%); prevention and chronic illness were the first problem raised during 21% and 19% of encounters, respectively. Patients who were randomly selected but were not observed (n=52, primarily no-shows) were similar in sex (67% women, c2 =0.119, P=.73 ) but were younger than those patients who were observed (mean age=32.1 years, t=3.79, P=.001).

On average, 2.7 problems were raised during an encounter Table 2. Forty-four percent of all problems were classified as acute, 30% chronic, 14% prevention, 4% administrative, 2% psychosocial, and 6% were classified as other. Of the observed encounters, 73% had more than one problem addressed. The physician raised 36% of these additional problems, and patients raised 58%. The problems raised by physicians were most frequently pertaining to chronic illness, prevention, and follow-up issues. The problems raised by patients were most likely to be acute illness problems. Additional problems were least likely to arise when the first problem addressed was an acute problem (61%) compared with visits during which the first problem addressed was chronic or prevention focused, where 88% and 87%, respectively, included additional problems during the visit (c2=21.2, P <.001).

On average, 8 (SD=4.5) physician actions were observed per encounter Table 2. Physicians performed an average of 3.3 (SD=1.2) actions per problem. The most common physician actions were questioning (77%), physical examination (49%), prescription writing (32%), providing advice (31%), and reassurance (25%). Of the 452 additional problems raised, only 3% of problems were ignored, and 6% were deferred to another visit.

The association of the number of problems addressed with the duration of the visit was assessed by analysis of variance and a test for linear trend. As shown in Figure 1, the duration of the visit increased approximately 2.5 minutes for each additional problem addressed (P <.001 for linear trend). The visit duration within each of the number of problem groups varied greatly as indicated by the large range for each group; however, the SD for each of the groups as indicated by the shaded bars are a similar size for each of the groups (Levene’s test of equality of error variance=1.48, P=.195).

The concordance between the number of problems observed and the number of problems on the billing sheet was modest, with a trend toward billing for fewer problems than were observed. As shown in Figure 2, 29% of encounters represented a match between the number of problems observed and the number of problems on the billing sheet. Fifty-eight percent of the encounters had more problems observed than recorded on the billing sheet. A much smaller proportion of encounters recorded more problems on the billing sheet than were observed during the encounter.

Discussion

Our exploratory study suggests that it is common for multiple problems to be addressed during visits to a family physician regardless of the initial reason for the visit. Additional problems are raised by both physicians and patients and are rarely deferred or ignored by the physician. Although the phenomenon of integrating a broad health agenda and addressing multiple problems during a single outpatient visit may be well known by practicing community-based family physicians, it may not be recognized by policymakers or health services researchers whose window into the process of outpatient care is provided by the medical record and billing data.

Addressing the majority of a patient’s health care needs and providing comprehensive care is a core feature of quality primary care.^16-20 Previous work has documented the wide range of diagnoses and clusters of diagnoses that family physicians commonly address during outpatient care.^13,21 However, truly comprehensive care goes beyond providing a broad array of services; it also involves the integration of care in a physician-patient relationship context. Prioritizing, providing, and orchestrating care for acute and undifferentiated illness, chronic disease, preventive services, and mental health care represents a key feature of primary care practice such that the care is greater than the sum of its individual commodities.¹ These data suggest that single visits often address a broad agenda of health care.

Overall, as the number of problems increase so does the length of the visit. Others have found that ordering or performing more tests, providing preventive services, and conducting ambulatory surgical procedures increase the length of the visit.²² It is not surprising that doing more is associated with a longer visit. However, the findings from our study suggest that longer visits and more physician actions are associated with addressing multiple unrelated problems during the patient encounter, which provides a different perspective on the intensity of the physician’s work.^23-26

Factors that affect the duration of the visit are of interest to those who use physician productivity as a measure for making policy and management decisions. Primary care physician productivity is commonly defined as the number of patients seen per hour.^27,28 Such indicators of productivity would rate a physician who saw many patients in a short time productive, while a physician who provided care to fewer patients but addressed multiple problems would be viewed as less productive. This viewpoint overlooks the cost savings that may result from the reduced number of future visits the patient may require to address these problems, the enhanced quality of care that may be attributable to follow-up of previously identified health concerns, and the enhanced patient satisfaction that may result from the physician’s expanded approach. The current measures of productivity are crude and possibly misleading indicators of the work involved with providing comprehensive primary care to patients. Perhaps health service researchers and policymakers should reconsider the definition of productivity in light of the number of problems addressed or the number of physician actions necessary to address the problems during a patient visit.

Our findings also have implications for evaluating the quality of care provided by family physicians. The current narrowly diseased-focused assessments of quality care are limited because they neglect to take into account the wide range of competing multiple illnesses, prevention, and psychosocial and family context issues confronting family physicians. Quality indicators for primary care should also assess the degree to which family physicians are making the right choices about how to prioritize among the multiple problems that could be addressed during an outpatient visit.

In combination with other reports,²⁹ these data should caution the use of billing records as an indicator of the content of the visit. These data indicate that the billing record generally underrepresents the number of problems addressed during the visit. The lack of concordance between what was observed and what was billed may have several explanations. Underrecording on the billing sheet may be due to the lack of an adequate way to code some problems addressed. Some physicians may approach the completion of the billing sheet by documenting just enough to justify the time spent. Also, the mode of recording the billing (forms or computer programs) may limit the number of problems that can be recorded per visit. Nonconcordance may have also occurred if the physician made decisions about management of ongoing illnesses that were not overtly apparent to the observer.

Limitations

The generalizability of our findings is limited by the modest-sized convenience sample of physicians observed. The higher no-show rate by younger patients may have increased the number of problems seen per visit, since older patients tend to have more problems. However, the patient visits included in our study were randomly selected from all adult patient visits during the 6-week data collection period and were similar in sex to the few patients who were not observed and are likely to be reflective of the patients presenting for care. Although not assessed directly, inter-rater reliability among the 7 students was maximized through the use of videotaped patient encounters for practicing completing the data collection form and for calibrating the observers before data collection in the field.

Conclusions

Prioritizing and delivering a diverse array of services within a relationship context is a hallmark of family practice. Our data suggest that addressing multiple problems during a single outpatient visit is one important mechanism family physicians use to provide comprehensive care. The value of addressing multiple problems per visit in terms of patient satisfaction, cost, and quality of care deserves further investigation.

Acknowledgments

We are grateful to Catharine Symmonds, Catherine Bettcher, Elizabeth Welsh, Tracy Lemonovich, Robin Baines, and Sarah Younkin who contributed to the study design and data collection phase and without whose participation our study would not have been possible. William R. Phillips, MD, MPH, and Kurt C. Stange, MD, PhD, provided valuable suggestions on an earlier draft of this paper.

Related Resources

• Center for Research in Family Practice and Primary Care http://mediswww.cwru.edu/dept/CRFPPC.
• American Academy of Family Practice policy studies in family practice and primary care http://www.aafppolicy.org

Primary care disciplines continue to have a central role in the health care of Americans. They provide breadth of care within an ongoing relationship, bridging the boundaries between health and illness and guiding access to more narrowly focused care when needed.¹ The ability to orchestrate a broad health agenda during a visit is central to primary care, but this ability is challenged by competing demands for time.²

Attempts to influence provision of care and treatment decisions by primary care physicians, such as financial incentives, administrative restrictions, and the implementation of evidence-based clinical guidelines add to the demands on physicians’ time and may affect how time is allocated during the day and with each patient. Within this context a primary care physician must prioritize the agenda for each patient visit. This may include providing services beyond the patient’s primary reason for the visit as time permits, such as including preventive services,³ follow-up of acute or chronic illnesses,¹ mental health⁴ or family issues,^5-7 or investigating “by the way” patient comments that may indicate serious medical issues.

The competing demands for time are compounded by patient requests during the visit. Based on an audiotape of 139 patient encounters, Kravitz and colleagues⁸ reported that on average a patient makes 5 requests for physician action or information per visit, and the number of unfulfilled requests was negatively associated with patient satisfaction. Such findings may fuel a sense of pressure to address patient requests. Also, another recent report indicates that the majority of patients do not have the opportunity to express all of their concerns before the physician redirects the interview; once redirected, additional patient concerns are rarely elicited.⁹ Fitting both the physician’s and patient’s agenda into the time allotted for an outpatient visit has important implications for the duration of the visit, physician productivity, and possibly patient outcomes.

Data on the number of problems raised and addressed have been limited by the lack of appropriate collection methods. Primarily audio and video technology have been used for the study of physician-patient communication.^10-12 Direct observation of patient encounters^12,13 and incorporation of ethnographic approaches have more recently been employed to fill a large void in the understanding of the content, context, and complexity of primary care.^13-15 Findings from the Direct Observation of Primary Care study, which employed such methods, indicate that among 4454 patient visits care was provided to a secondary patient during 18% of the visits and preventive services were addressed during 32% of the illness visits.³ Data from that study provide a glimpse into some types of problems addressed in addition to the main reason for the visit; however, data about the number of problems addressed during patient encounters were not specifically collected by the nurse observer.

When additional issues are raised during a patient encounter, little is known about the nature of these problems, how additional problems affect the duration of the visit, and how well additional problems are reflected in the billing record. This led us to conduct an observational study to ask: How many problems are addressed during family practice outpatient visits, and who is raising additional problems? How much work and time is associated with addressing problems raised beyond the initial problem? How well does the billing list represent the number of problems addressed during the outpatient visit? Our study was designed to directly observe and record how many problems were raised and addressed during outpatient visits to family physicians.

Methods

Seven first-year medical students observed patient care provided by their summer fellowship family physician preceptor and other physicians in the preceptor’s practice from June through August 1999. Six of the sites were located in Northeast Ohio, and one was in Tulsa, Oklahoma.

Each student collected data on one randomly selected adult patient encounter for each half day of precepting. At the beginning of each half-day of patient care the student rolled a die to generate a random number to select a patient from the patient schedule. To ensure random selection of encounters within each half-day session, on alternating days the random number was counted from the beginning or the end of the half-day schedule. If the selected patient was aged younger than 18 years, the patient or physician preferred the encounter not be observed, or the patient did not show up for the scheduled appointment, the next scheduled appointment was selected as a replacement. Patient age and sex were collected for those who were no-shows or chose not to be observed, so they could be compared with those patients who were observed. Each student was to collect data on approximately 50 patient encounters during the 6-week summer fellowship. The physicians were blinded to the study purpose and were not told which patient encounter would be included in the study.

A problem was operationalized as an issue requiring physician action in the form of a decision, diagnosis, treatment, or monitoring. Each item was listed as it was raised, and the type of problem, who raised it, and what physician actions were involved to address it were coded. Each problem was coded as 1 of 14 categories: acute, acute follow-up, chronic, chronic follow-up, prevention, prevention follow-up, psychosocial, psychosocial follow-up, work-related administrative, health care system-related administrative, other family member’s problem, pregnancy, emergent, and other. The person who raised the problem was coded as 1 of 3 options: the physician, the patient or another person in the room. Multiple physician actions could be coded for how the problem was addressed. The 19 physician action categories included: question, reassurance, examination, procedure, referral, return visit, advice, review tests, order laboratory testing, prescription, provide written material, imaging, admits uncertainty, counseling, return to work/time off work letter, defer, complementary/alternative medicine, ignored or lost, and other.

Patient characteristics, the duration of the visit, and the billing diagnoses for each visit were also recorded on the data collection form. Videotaped encounters were used to pilot test the data collection form, to allow the observers to practice using the form in real time, and to calibrate the observers before data collection in the field.

We used descriptive statistics to address most research questions. Student t tests and chi-square tests were used to compare age and sex differences between participants and nonparticipants. We tested the association of the number of problems with the duration of the visit with analysis of variance and a test for linear trend. A difference score of the number of problems observed and the number of problems recorded on the billing sheet for the encounter was computed and summarized graphically.

Results

We collected usable data on 266 encounters representing 37 physicians. Patient and visit characteristics are displayed in Table 1. The patients had an average age of 48 years, and 69% were women. They were predominately white. A large proportion was observed visiting their regular primary care physician (83%), and 85% were established patients of the practice. Most of the observed patients had some kind of commercial health care insurance, 19% had Medicare, and a small proportion had Medicaid or no insurance. The visit duration ranged from 2 to 65 minutes; the median was 15 minutes with a mean of 19.3 (standard deviation [SD]=12.7). The first problem raised was most commonly an acute problem (49%); prevention and chronic illness were the first problem raised during 21% and 19% of encounters, respectively. Patients who were randomly selected but were not observed (n=52, primarily no-shows) were similar in sex (67% women, c2 =0.119, P=.73 ) but were younger than those patients who were observed (mean age=32.1 years, t=3.79, P=.001).

On average, 2.7 problems were raised during an encounter Table 2. Forty-four percent of all problems were classified as acute, 30% chronic, 14% prevention, 4% administrative, 2% psychosocial, and 6% were classified as other. Of the observed encounters, 73% had more than one problem addressed. The physician raised 36% of these additional problems, and patients raised 58%. The problems raised by physicians were most frequently pertaining to chronic illness, prevention, and follow-up issues. The problems raised by patients were most likely to be acute illness problems. Additional problems were least likely to arise when the first problem addressed was an acute problem (61%) compared with visits during which the first problem addressed was chronic or prevention focused, where 88% and 87%, respectively, included additional problems during the visit (c2=21.2, P <.001).

On average, 8 (SD=4.5) physician actions were observed per encounter Table 2. Physicians performed an average of 3.3 (SD=1.2) actions per problem. The most common physician actions were questioning (77%), physical examination (49%), prescription writing (32%), providing advice (31%), and reassurance (25%). Of the 452 additional problems raised, only 3% of problems were ignored, and 6% were deferred to another visit.

The association of the number of problems addressed with the duration of the visit was assessed by analysis of variance and a test for linear trend. As shown in Figure 1, the duration of the visit increased approximately 2.5 minutes for each additional problem addressed (P <.001 for linear trend). The visit duration within each of the number of problem groups varied greatly as indicated by the large range for each group; however, the SD for each of the groups as indicated by the shaded bars are a similar size for each of the groups (Levene’s test of equality of error variance=1.48, P=.195).

The concordance between the number of problems observed and the number of problems on the billing sheet was modest, with a trend toward billing for fewer problems than were observed. As shown in Figure 2, 29% of encounters represented a match between the number of problems observed and the number of problems on the billing sheet. Fifty-eight percent of the encounters had more problems observed than recorded on the billing sheet. A much smaller proportion of encounters recorded more problems on the billing sheet than were observed during the encounter.

Discussion

Our exploratory study suggests that it is common for multiple problems to be addressed during visits to a family physician regardless of the initial reason for the visit. Additional problems are raised by both physicians and patients and are rarely deferred or ignored by the physician. Although the phenomenon of integrating a broad health agenda and addressing multiple problems during a single outpatient visit may be well known by practicing community-based family physicians, it may not be recognized by policymakers or health services researchers whose window into the process of outpatient care is provided by the medical record and billing data.

Addressing the majority of a patient’s health care needs and providing comprehensive care is a core feature of quality primary care.^16-20 Previous work has documented the wide range of diagnoses and clusters of diagnoses that family physicians commonly address during outpatient care.^13,21 However, truly comprehensive care goes beyond providing a broad array of services; it also involves the integration of care in a physician-patient relationship context. Prioritizing, providing, and orchestrating care for acute and undifferentiated illness, chronic disease, preventive services, and mental health care represents a key feature of primary care practice such that the care is greater than the sum of its individual commodities.¹ These data suggest that single visits often address a broad agenda of health care.

Overall, as the number of problems increase so does the length of the visit. Others have found that ordering or performing more tests, providing preventive services, and conducting ambulatory surgical procedures increase the length of the visit.²² It is not surprising that doing more is associated with a longer visit. However, the findings from our study suggest that longer visits and more physician actions are associated with addressing multiple unrelated problems during the patient encounter, which provides a different perspective on the intensity of the physician’s work.^23-26

Factors that affect the duration of the visit are of interest to those who use physician productivity as a measure for making policy and management decisions. Primary care physician productivity is commonly defined as the number of patients seen per hour.^27,28 Such indicators of productivity would rate a physician who saw many patients in a short time productive, while a physician who provided care to fewer patients but addressed multiple problems would be viewed as less productive. This viewpoint overlooks the cost savings that may result from the reduced number of future visits the patient may require to address these problems, the enhanced quality of care that may be attributable to follow-up of previously identified health concerns, and the enhanced patient satisfaction that may result from the physician’s expanded approach. The current measures of productivity are crude and possibly misleading indicators of the work involved with providing comprehensive primary care to patients. Perhaps health service researchers and policymakers should reconsider the definition of productivity in light of the number of problems addressed or the number of physician actions necessary to address the problems during a patient visit.

Our findings also have implications for evaluating the quality of care provided by family physicians. The current narrowly diseased-focused assessments of quality care are limited because they neglect to take into account the wide range of competing multiple illnesses, prevention, and psychosocial and family context issues confronting family physicians. Quality indicators for primary care should also assess the degree to which family physicians are making the right choices about how to prioritize among the multiple problems that could be addressed during an outpatient visit.

In combination with other reports,²⁹ these data should caution the use of billing records as an indicator of the content of the visit. These data indicate that the billing record generally underrepresents the number of problems addressed during the visit. The lack of concordance between what was observed and what was billed may have several explanations. Underrecording on the billing sheet may be due to the lack of an adequate way to code some problems addressed. Some physicians may approach the completion of the billing sheet by documenting just enough to justify the time spent. Also, the mode of recording the billing (forms or computer programs) may limit the number of problems that can be recorded per visit. Nonconcordance may have also occurred if the physician made decisions about management of ongoing illnesses that were not overtly apparent to the observer.

Limitations

The generalizability of our findings is limited by the modest-sized convenience sample of physicians observed. The higher no-show rate by younger patients may have increased the number of problems seen per visit, since older patients tend to have more problems. However, the patient visits included in our study were randomly selected from all adult patient visits during the 6-week data collection period and were similar in sex to the few patients who were not observed and are likely to be reflective of the patients presenting for care. Although not assessed directly, inter-rater reliability among the 7 students was maximized through the use of videotaped patient encounters for practicing completing the data collection form and for calibrating the observers before data collection in the field.

Conclusions

Prioritizing and delivering a diverse array of services within a relationship context is a hallmark of family practice. Our data suggest that addressing multiple problems during a single outpatient visit is one important mechanism family physicians use to provide comprehensive care. The value of addressing multiple problems per visit in terms of patient satisfaction, cost, and quality of care deserves further investigation.

Acknowledgments

We are grateful to Catharine Symmonds, Catherine Bettcher, Elizabeth Welsh, Tracy Lemonovich, Robin Baines, and Sarah Younkin who contributed to the study design and data collection phase and without whose participation our study would not have been possible. William R. Phillips, MD, MPH, and Kurt C. Stange, MD, PhD, provided valuable suggestions on an earlier draft of this paper.

Related Resources

• Center for Research in Family Practice and Primary Care http://mediswww.cwru.edu/dept/CRFPPC.
• American Academy of Family Practice policy studies in family practice and primary care http://www.aafppolicy.org

Primary care disciplines continue to have a central role in the health care of Americans. They provide breadth of care within an ongoing relationship, bridging the boundaries between health and illness and guiding access to more narrowly focused care when needed.¹ The ability to orchestrate a broad health agenda during a visit is central to primary care, but this ability is challenged by competing demands for time.²

Attempts to influence provision of care and treatment decisions by primary care physicians, such as financial incentives, administrative restrictions, and the implementation of evidence-based clinical guidelines add to the demands on physicians’ time and may affect how time is allocated during the day and with each patient. Within this context a primary care physician must prioritize the agenda for each patient visit. This may include providing services beyond the patient’s primary reason for the visit as time permits, such as including preventive services,³ follow-up of acute or chronic illnesses,¹ mental health⁴ or family issues,^5-7 or investigating “by the way” patient comments that may indicate serious medical issues.

The competing demands for time are compounded by patient requests during the visit. Based on an audiotape of 139 patient encounters, Kravitz and colleagues⁸ reported that on average a patient makes 5 requests for physician action or information per visit, and the number of unfulfilled requests was negatively associated with patient satisfaction. Such findings may fuel a sense of pressure to address patient requests. Also, another recent report indicates that the majority of patients do not have the opportunity to express all of their concerns before the physician redirects the interview; once redirected, additional patient concerns are rarely elicited.⁹ Fitting both the physician’s and patient’s agenda into the time allotted for an outpatient visit has important implications for the duration of the visit, physician productivity, and possibly patient outcomes.

Data on the number of problems raised and addressed have been limited by the lack of appropriate collection methods. Primarily audio and video technology have been used for the study of physician-patient communication.^10-12 Direct observation of patient encounters^12,13 and incorporation of ethnographic approaches have more recently been employed to fill a large void in the understanding of the content, context, and complexity of primary care.^13-15 Findings from the Direct Observation of Primary Care study, which employed such methods, indicate that among 4454 patient visits care was provided to a secondary patient during 18% of the visits and preventive services were addressed during 32% of the illness visits.³ Data from that study provide a glimpse into some types of problems addressed in addition to the main reason for the visit; however, data about the number of problems addressed during patient encounters were not specifically collected by the nurse observer.

When additional issues are raised during a patient encounter, little is known about the nature of these problems, how additional problems affect the duration of the visit, and how well additional problems are reflected in the billing record. This led us to conduct an observational study to ask: How many problems are addressed during family practice outpatient visits, and who is raising additional problems? How much work and time is associated with addressing problems raised beyond the initial problem? How well does the billing list represent the number of problems addressed during the outpatient visit? Our study was designed to directly observe and record how many problems were raised and addressed during outpatient visits to family physicians.

Methods

Seven first-year medical students observed patient care provided by their summer fellowship family physician preceptor and other physicians in the preceptor’s practice from June through August 1999. Six of the sites were located in Northeast Ohio, and one was in Tulsa, Oklahoma.

Each student collected data on one randomly selected adult patient encounter for each half day of precepting. At the beginning of each half-day of patient care the student rolled a die to generate a random number to select a patient from the patient schedule. To ensure random selection of encounters within each half-day session, on alternating days the random number was counted from the beginning or the end of the half-day schedule. If the selected patient was aged younger than 18 years, the patient or physician preferred the encounter not be observed, or the patient did not show up for the scheduled appointment, the next scheduled appointment was selected as a replacement. Patient age and sex were collected for those who were no-shows or chose not to be observed, so they could be compared with those patients who were observed. Each student was to collect data on approximately 50 patient encounters during the 6-week summer fellowship. The physicians were blinded to the study purpose and were not told which patient encounter would be included in the study.

A problem was operationalized as an issue requiring physician action in the form of a decision, diagnosis, treatment, or monitoring. Each item was listed as it was raised, and the type of problem, who raised it, and what physician actions were involved to address it were coded. Each problem was coded as 1 of 14 categories: acute, acute follow-up, chronic, chronic follow-up, prevention, prevention follow-up, psychosocial, psychosocial follow-up, work-related administrative, health care system-related administrative, other family member’s problem, pregnancy, emergent, and other. The person who raised the problem was coded as 1 of 3 options: the physician, the patient or another person in the room. Multiple physician actions could be coded for how the problem was addressed. The 19 physician action categories included: question, reassurance, examination, procedure, referral, return visit, advice, review tests, order laboratory testing, prescription, provide written material, imaging, admits uncertainty, counseling, return to work/time off work letter, defer, complementary/alternative medicine, ignored or lost, and other.

Patient characteristics, the duration of the visit, and the billing diagnoses for each visit were also recorded on the data collection form. Videotaped encounters were used to pilot test the data collection form, to allow the observers to practice using the form in real time, and to calibrate the observers before data collection in the field.

We used descriptive statistics to address most research questions. Student t tests and chi-square tests were used to compare age and sex differences between participants and nonparticipants. We tested the association of the number of problems with the duration of the visit with analysis of variance and a test for linear trend. A difference score of the number of problems observed and the number of problems recorded on the billing sheet for the encounter was computed and summarized graphically.

Results

We collected usable data on 266 encounters representing 37 physicians. Patient and visit characteristics are displayed in Table 1. The patients had an average age of 48 years, and 69% were women. They were predominately white. A large proportion was observed visiting their regular primary care physician (83%), and 85% were established patients of the practice. Most of the observed patients had some kind of commercial health care insurance, 19% had Medicare, and a small proportion had Medicaid or no insurance. The visit duration ranged from 2 to 65 minutes; the median was 15 minutes with a mean of 19.3 (standard deviation [SD]=12.7). The first problem raised was most commonly an acute problem (49%); prevention and chronic illness were the first problem raised during 21% and 19% of encounters, respectively. Patients who were randomly selected but were not observed (n=52, primarily no-shows) were similar in sex (67% women, c2 =0.119, P=.73 ) but were younger than those patients who were observed (mean age=32.1 years, t=3.79, P=.001).

On average, 2.7 problems were raised during an encounter Table 2. Forty-four percent of all problems were classified as acute, 30% chronic, 14% prevention, 4% administrative, 2% psychosocial, and 6% were classified as other. Of the observed encounters, 73% had more than one problem addressed. The physician raised 36% of these additional problems, and patients raised 58%. The problems raised by physicians were most frequently pertaining to chronic illness, prevention, and follow-up issues. The problems raised by patients were most likely to be acute illness problems. Additional problems were least likely to arise when the first problem addressed was an acute problem (61%) compared with visits during which the first problem addressed was chronic or prevention focused, where 88% and 87%, respectively, included additional problems during the visit (c2=21.2, P <.001).

On average, 8 (SD=4.5) physician actions were observed per encounter Table 2. Physicians performed an average of 3.3 (SD=1.2) actions per problem. The most common physician actions were questioning (77%), physical examination (49%), prescription writing (32%), providing advice (31%), and reassurance (25%). Of the 452 additional problems raised, only 3% of problems were ignored, and 6% were deferred to another visit.

The association of the number of problems addressed with the duration of the visit was assessed by analysis of variance and a test for linear trend. As shown in Figure 1, the duration of the visit increased approximately 2.5 minutes for each additional problem addressed (P <.001 for linear trend). The visit duration within each of the number of problem groups varied greatly as indicated by the large range for each group; however, the SD for each of the groups as indicated by the shaded bars are a similar size for each of the groups (Levene’s test of equality of error variance=1.48, P=.195).

The concordance between the number of problems observed and the number of problems on the billing sheet was modest, with a trend toward billing for fewer problems than were observed. As shown in Figure 2, 29% of encounters represented a match between the number of problems observed and the number of problems on the billing sheet. Fifty-eight percent of the encounters had more problems observed than recorded on the billing sheet. A much smaller proportion of encounters recorded more problems on the billing sheet than were observed during the encounter.

Discussion

Our exploratory study suggests that it is common for multiple problems to be addressed during visits to a family physician regardless of the initial reason for the visit. Additional problems are raised by both physicians and patients and are rarely deferred or ignored by the physician. Although the phenomenon of integrating a broad health agenda and addressing multiple problems during a single outpatient visit may be well known by practicing community-based family physicians, it may not be recognized by policymakers or health services researchers whose window into the process of outpatient care is provided by the medical record and billing data.

Addressing the majority of a patient’s health care needs and providing comprehensive care is a core feature of quality primary care.^16-20 Previous work has documented the wide range of diagnoses and clusters of diagnoses that family physicians commonly address during outpatient care.^13,21 However, truly comprehensive care goes beyond providing a broad array of services; it also involves the integration of care in a physician-patient relationship context. Prioritizing, providing, and orchestrating care for acute and undifferentiated illness, chronic disease, preventive services, and mental health care represents a key feature of primary care practice such that the care is greater than the sum of its individual commodities.¹ These data suggest that single visits often address a broad agenda of health care.

Overall, as the number of problems increase so does the length of the visit. Others have found that ordering or performing more tests, providing preventive services, and conducting ambulatory surgical procedures increase the length of the visit.²² It is not surprising that doing more is associated with a longer visit. However, the findings from our study suggest that longer visits and more physician actions are associated with addressing multiple unrelated problems during the patient encounter, which provides a different perspective on the intensity of the physician’s work.^23-26

Factors that affect the duration of the visit are of interest to those who use physician productivity as a measure for making policy and management decisions. Primary care physician productivity is commonly defined as the number of patients seen per hour.^27,28 Such indicators of productivity would rate a physician who saw many patients in a short time productive, while a physician who provided care to fewer patients but addressed multiple problems would be viewed as less productive. This viewpoint overlooks the cost savings that may result from the reduced number of future visits the patient may require to address these problems, the enhanced quality of care that may be attributable to follow-up of previously identified health concerns, and the enhanced patient satisfaction that may result from the physician’s expanded approach. The current measures of productivity are crude and possibly misleading indicators of the work involved with providing comprehensive primary care to patients. Perhaps health service researchers and policymakers should reconsider the definition of productivity in light of the number of problems addressed or the number of physician actions necessary to address the problems during a patient visit.

Our findings also have implications for evaluating the quality of care provided by family physicians. The current narrowly diseased-focused assessments of quality care are limited because they neglect to take into account the wide range of competing multiple illnesses, prevention, and psychosocial and family context issues confronting family physicians. Quality indicators for primary care should also assess the degree to which family physicians are making the right choices about how to prioritize among the multiple problems that could be addressed during an outpatient visit.

In combination with other reports,²⁹ these data should caution the use of billing records as an indicator of the content of the visit. These data indicate that the billing record generally underrepresents the number of problems addressed during the visit. The lack of concordance between what was observed and what was billed may have several explanations. Underrecording on the billing sheet may be due to the lack of an adequate way to code some problems addressed. Some physicians may approach the completion of the billing sheet by documenting just enough to justify the time spent. Also, the mode of recording the billing (forms or computer programs) may limit the number of problems that can be recorded per visit. Nonconcordance may have also occurred if the physician made decisions about management of ongoing illnesses that were not overtly apparent to the observer.

Limitations

The generalizability of our findings is limited by the modest-sized convenience sample of physicians observed. The higher no-show rate by younger patients may have increased the number of problems seen per visit, since older patients tend to have more problems. However, the patient visits included in our study were randomly selected from all adult patient visits during the 6-week data collection period and were similar in sex to the few patients who were not observed and are likely to be reflective of the patients presenting for care. Although not assessed directly, inter-rater reliability among the 7 students was maximized through the use of videotaped patient encounters for practicing completing the data collection form and for calibrating the observers before data collection in the field.

Conclusions

Prioritizing and delivering a diverse array of services within a relationship context is a hallmark of family practice. Our data suggest that addressing multiple problems during a single outpatient visit is one important mechanism family physicians use to provide comprehensive care. The value of addressing multiple problems per visit in terms of patient satisfaction, cost, and quality of care deserves further investigation.

Acknowledgments

We are grateful to Catharine Symmonds, Catherine Bettcher, Elizabeth Welsh, Tracy Lemonovich, Robin Baines, and Sarah Younkin who contributed to the study design and data collection phase and without whose participation our study would not have been possible. William R. Phillips, MD, MPH, and Kurt C. Stange, MD, PhD, provided valuable suggestions on an earlier draft of this paper.

Related Resources

• Center for Research in Family Practice and Primary Care http://mediswww.cwru.edu/dept/CRFPPC.
• American Academy of Family Practice policy studies in family practice and primary care http://www.aafppolicy.org

Like other adult learners, physicians will seek and retain new knowledge only when motivated to do so (ie, when they have the need to know). As a result, efforts to increase clinicians’ use of the best information at the point of care must focus on providing them with well-validated evidence showing a direct and relevant benefit to their patients (eg, Patient-Oriented Evidence that Matters [POEMs] reviews¹).

Previously,² we described our efforts to identify the relatively few research findings in the medical literature that provide both relevant and valid new information for practicing clinicians. Of 8085 articles published in 85 medical journals over a 6-month period, only 2.6% (211) met these criteria.

These 211 research articles were summarized in issues of the newsletter Evidence-Based Practice and are incorporated into InfoRetriever, an electronic database using POEMs to improve information access at the point of care.³ Other such services, such as Journal Watch and Best Evidence⁴ provide similar reviews of the recent medical literature. However, Journal Watch has no published criteria explaining how articles are chosen for inclusion⁵ or how the validity of the information is determined. Best Evidence focuses primarily on the validity of research, and the criteria for relevance are not clearly defined.⁶ This valuing of rigor over relevance may lead to providing information to clinicians that they do not really need or omitting important information that they do need. In this exploratory study we aimed to find out how much overlap in content exists between Evidence-Based Practice and Best Evidence.

Methods

To evaluate the differences between Best Evidence and Evidence-Based Practice, we compared the articles in ACP Journal Club and the discontinued Evidence-Based Medicine, which are now combined into Best Evidence, with those summarized in Evidence-Based Practice. We chose for comparison the 5 issues of Evidence-Based Practice published between January and May of 1998. Since the time to publication of the ACP Journal Club and Evidence-Based Medicine is longer than that for Evidence-Based Practice, we used for comparison 6 bimonthly issues of both ACP Journal Club and Evidence-Based Medicine, starting with the November-December 1997 issues and ending with the November-December 1998 issues.

Results

Over a 5-month period, 85 POEMs were published in Evidence-Based Practice. There was little overlap between the 3 publications. Only 11 (12.9%) of these POEMs were also published in either ACP Journal Club or Evidence-Based Medicine. To compare in the other direction, 3 bimonthly issues of Evidence-Based Medicine and ACP Journal Club were chosen and compared with all issues of Evidence-Based Practice. The results are summarized in the Table. A total of 109 synopses were published in the 2 Best Evidence publications during this time. Most of these synopses (n=82, 75.2%) were not considered POEMs and were not published in Evidence-Based Practice. Of the 49 distinct articles (33 articles were reviewed in both publications) found in these publications but not selected for Evidence-Based Practice, 22 (45%) studied interventions or diseases not relevant to family practice, 15 (31%) would not induce a change in practice, 8 (16%) were in journals not covered by Evidence-Based Practice (only 1 of these articles was a POEM), 3 (6%) evaluated disease-oriented outcomes, and 1 article (2%) was a POEM that had been earmarked for inclusion in Evidence Based Practice but was lost in transmission.

Discussion

This small informal study shows the marked difference between Evidence-Based Practice and the content of Best Evidence. Readers of only Best Evidence would miss a significant amount of high-quality information directly applicable to primary care practice. Although Evidence-Based Practice and Best Evidence use essentially the same validity criteria^6,7 to screen preliminary research results, the key difference between them resides in the relevance of the information each source chooses to present. Evidence-Based Practice focuses on patient-oriented evidence that matters, which is information that must pass 3 relevance criteria: (1) the affected outcome must be one that patients care about (ie, not disease-oriented outcomes); (2) the proposed intervention must be feasible and deal with a common problem; and (3) the results being presented should require a change in practice on the part of most clinicians.

The concepts embodied in evidence-based medicine have been described as the long-awaited bridge between research and clinical practice. Although the techniques of evidence-based medicine have greatly enhanced and simplified the evaluation of the validity of clinical research, they are not practical to meet the day-to-day needs of busy, real-life clinicians. This method is problem driven; the search for information begins with the generation of a specific patient-based question. However, primary care clinicians are usually in a more general keeping-up mode, where foraging for information is just as important as hunting for answers to patients’ specific questions.⁸

The traditional evidence-based medicine approach, although attractive to academics, has not been widely embraced by clinicians because it focuses on identifying and validating information communicated by the written word, making it unrealistic and too time consuming for most clinicians. This approach of rigor over relevance is rooted deep in the foundation of pedagogy,⁹ but is less valid when applied to adult learners.

Two specific tools are needed to help physicians efficiently identify information that is highly relevant and valid. Clinicians need a first-alert method—a POEM bulletin board—for relevant new information as it becomes available. Resources—newsletters, Web sites, continuing education, and others—used by clinicians to update their knowledge should carefully filter out preliminary or unverified information so that this keeping-up process is efficient.

Clinicians also need a way of rapidly retrieving the information to which they have been alerted but that has not yet been cemented into their minds. Computer-based resources (especially handheld portable devices) are available that can provide information in less than 30 seconds. To be lifelong learners, physicians have to use tools that help them to hunt and forage through the jungle of information.

Irrelevant information, even if highly valid, is not useful in the scope of a busy daily practice. Sifting through valid, but irrelevant, information wastes valuable information-gathering time. To be relevant and complete, a comprehensive foraging source for new information must contain specialty-specific POEMs (a POEM alert system). The one-stop shopping approach of information for all specialties offered by Best Evidence does not meet this need.

Like other adult learners, physicians will seek and retain new knowledge only when motivated to do so (ie, when they have the need to know). As a result, efforts to increase clinicians’ use of the best information at the point of care must focus on providing them with well-validated evidence showing a direct and relevant benefit to their patients (eg, Patient-Oriented Evidence that Matters [POEMs] reviews¹).

Previously,² we described our efforts to identify the relatively few research findings in the medical literature that provide both relevant and valid new information for practicing clinicians. Of 8085 articles published in 85 medical journals over a 6-month period, only 2.6% (211) met these criteria.

These 211 research articles were summarized in issues of the newsletter Evidence-Based Practice and are incorporated into InfoRetriever, an electronic database using POEMs to improve information access at the point of care.³ Other such services, such as Journal Watch and Best Evidence⁴ provide similar reviews of the recent medical literature. However, Journal Watch has no published criteria explaining how articles are chosen for inclusion⁵ or how the validity of the information is determined. Best Evidence focuses primarily on the validity of research, and the criteria for relevance are not clearly defined.⁶ This valuing of rigor over relevance may lead to providing information to clinicians that they do not really need or omitting important information that they do need. In this exploratory study we aimed to find out how much overlap in content exists between Evidence-Based Practice and Best Evidence.

Methods

To evaluate the differences between Best Evidence and Evidence-Based Practice, we compared the articles in ACP Journal Club and the discontinued Evidence-Based Medicine, which are now combined into Best Evidence, with those summarized in Evidence-Based Practice. We chose for comparison the 5 issues of Evidence-Based Practice published between January and May of 1998. Since the time to publication of the ACP Journal Club and Evidence-Based Medicine is longer than that for Evidence-Based Practice, we used for comparison 6 bimonthly issues of both ACP Journal Club and Evidence-Based Medicine, starting with the November-December 1997 issues and ending with the November-December 1998 issues.

Results

Over a 5-month period, 85 POEMs were published in Evidence-Based Practice. There was little overlap between the 3 publications. Only 11 (12.9%) of these POEMs were also published in either ACP Journal Club or Evidence-Based Medicine. To compare in the other direction, 3 bimonthly issues of Evidence-Based Medicine and ACP Journal Club were chosen and compared with all issues of Evidence-Based Practice. The results are summarized in the Table. A total of 109 synopses were published in the 2 Best Evidence publications during this time. Most of these synopses (n=82, 75.2%) were not considered POEMs and were not published in Evidence-Based Practice. Of the 49 distinct articles (33 articles were reviewed in both publications) found in these publications but not selected for Evidence-Based Practice, 22 (45%) studied interventions or diseases not relevant to family practice, 15 (31%) would not induce a change in practice, 8 (16%) were in journals not covered by Evidence-Based Practice (only 1 of these articles was a POEM), 3 (6%) evaluated disease-oriented outcomes, and 1 article (2%) was a POEM that had been earmarked for inclusion in Evidence Based Practice but was lost in transmission.

Discussion

This small informal study shows the marked difference between Evidence-Based Practice and the content of Best Evidence. Readers of only Best Evidence would miss a significant amount of high-quality information directly applicable to primary care practice. Although Evidence-Based Practice and Best Evidence use essentially the same validity criteria^6,7 to screen preliminary research results, the key difference between them resides in the relevance of the information each source chooses to present. Evidence-Based Practice focuses on patient-oriented evidence that matters, which is information that must pass 3 relevance criteria: (1) the affected outcome must be one that patients care about (ie, not disease-oriented outcomes); (2) the proposed intervention must be feasible and deal with a common problem; and (3) the results being presented should require a change in practice on the part of most clinicians.

The concepts embodied in evidence-based medicine have been described as the long-awaited bridge between research and clinical practice. Although the techniques of evidence-based medicine have greatly enhanced and simplified the evaluation of the validity of clinical research, they are not practical to meet the day-to-day needs of busy, real-life clinicians. This method is problem driven; the search for information begins with the generation of a specific patient-based question. However, primary care clinicians are usually in a more general keeping-up mode, where foraging for information is just as important as hunting for answers to patients’ specific questions.⁸

The traditional evidence-based medicine approach, although attractive to academics, has not been widely embraced by clinicians because it focuses on identifying and validating information communicated by the written word, making it unrealistic and too time consuming for most clinicians. This approach of rigor over relevance is rooted deep in the foundation of pedagogy,⁹ but is less valid when applied to adult learners.

Two specific tools are needed to help physicians efficiently identify information that is highly relevant and valid. Clinicians need a first-alert method—a POEM bulletin board—for relevant new information as it becomes available. Resources—newsletters, Web sites, continuing education, and others—used by clinicians to update their knowledge should carefully filter out preliminary or unverified information so that this keeping-up process is efficient.

Clinicians also need a way of rapidly retrieving the information to which they have been alerted but that has not yet been cemented into their minds. Computer-based resources (especially handheld portable devices) are available that can provide information in less than 30 seconds. To be lifelong learners, physicians have to use tools that help them to hunt and forage through the jungle of information.

Irrelevant information, even if highly valid, is not useful in the scope of a busy daily practice. Sifting through valid, but irrelevant, information wastes valuable information-gathering time. To be relevant and complete, a comprehensive foraging source for new information must contain specialty-specific POEMs (a POEM alert system). The one-stop shopping approach of information for all specialties offered by Best Evidence does not meet this need.

Like other adult learners, physicians will seek and retain new knowledge only when motivated to do so (ie, when they have the need to know). As a result, efforts to increase clinicians’ use of the best information at the point of care must focus on providing them with well-validated evidence showing a direct and relevant benefit to their patients (eg, Patient-Oriented Evidence that Matters [POEMs] reviews¹).

Previously,² we described our efforts to identify the relatively few research findings in the medical literature that provide both relevant and valid new information for practicing clinicians. Of 8085 articles published in 85 medical journals over a 6-month period, only 2.6% (211) met these criteria.

These 211 research articles were summarized in issues of the newsletter Evidence-Based Practice and are incorporated into InfoRetriever, an electronic database using POEMs to improve information access at the point of care.³ Other such services, such as Journal Watch and Best Evidence⁴ provide similar reviews of the recent medical literature. However, Journal Watch has no published criteria explaining how articles are chosen for inclusion⁵ or how the validity of the information is determined. Best Evidence focuses primarily on the validity of research, and the criteria for relevance are not clearly defined.⁶ This valuing of rigor over relevance may lead to providing information to clinicians that they do not really need or omitting important information that they do need. In this exploratory study we aimed to find out how much overlap in content exists between Evidence-Based Practice and Best Evidence.

Methods

To evaluate the differences between Best Evidence and Evidence-Based Practice, we compared the articles in ACP Journal Club and the discontinued Evidence-Based Medicine, which are now combined into Best Evidence, with those summarized in Evidence-Based Practice. We chose for comparison the 5 issues of Evidence-Based Practice published between January and May of 1998. Since the time to publication of the ACP Journal Club and Evidence-Based Medicine is longer than that for Evidence-Based Practice, we used for comparison 6 bimonthly issues of both ACP Journal Club and Evidence-Based Medicine, starting with the November-December 1997 issues and ending with the November-December 1998 issues.

Results

Over a 5-month period, 85 POEMs were published in Evidence-Based Practice. There was little overlap between the 3 publications. Only 11 (12.9%) of these POEMs were also published in either ACP Journal Club or Evidence-Based Medicine. To compare in the other direction, 3 bimonthly issues of Evidence-Based Medicine and ACP Journal Club were chosen and compared with all issues of Evidence-Based Practice. The results are summarized in the Table. A total of 109 synopses were published in the 2 Best Evidence publications during this time. Most of these synopses (n=82, 75.2%) were not considered POEMs and were not published in Evidence-Based Practice. Of the 49 distinct articles (33 articles were reviewed in both publications) found in these publications but not selected for Evidence-Based Practice, 22 (45%) studied interventions or diseases not relevant to family practice, 15 (31%) would not induce a change in practice, 8 (16%) were in journals not covered by Evidence-Based Practice (only 1 of these articles was a POEM), 3 (6%) evaluated disease-oriented outcomes, and 1 article (2%) was a POEM that had been earmarked for inclusion in Evidence Based Practice but was lost in transmission.

Discussion

This small informal study shows the marked difference between Evidence-Based Practice and the content of Best Evidence. Readers of only Best Evidence would miss a significant amount of high-quality information directly applicable to primary care practice. Although Evidence-Based Practice and Best Evidence use essentially the same validity criteria^6,7 to screen preliminary research results, the key difference between them resides in the relevance of the information each source chooses to present. Evidence-Based Practice focuses on patient-oriented evidence that matters, which is information that must pass 3 relevance criteria: (1) the affected outcome must be one that patients care about (ie, not disease-oriented outcomes); (2) the proposed intervention must be feasible and deal with a common problem; and (3) the results being presented should require a change in practice on the part of most clinicians.

The concepts embodied in evidence-based medicine have been described as the long-awaited bridge between research and clinical practice. Although the techniques of evidence-based medicine have greatly enhanced and simplified the evaluation of the validity of clinical research, they are not practical to meet the day-to-day needs of busy, real-life clinicians. This method is problem driven; the search for information begins with the generation of a specific patient-based question. However, primary care clinicians are usually in a more general keeping-up mode, where foraging for information is just as important as hunting for answers to patients’ specific questions.⁸

The traditional evidence-based medicine approach, although attractive to academics, has not been widely embraced by clinicians because it focuses on identifying and validating information communicated by the written word, making it unrealistic and too time consuming for most clinicians. This approach of rigor over relevance is rooted deep in the foundation of pedagogy,⁹ but is less valid when applied to adult learners.

Two specific tools are needed to help physicians efficiently identify information that is highly relevant and valid. Clinicians need a first-alert method—a POEM bulletin board—for relevant new information as it becomes available. Resources—newsletters, Web sites, continuing education, and others—used by clinicians to update their knowledge should carefully filter out preliminary or unverified information so that this keeping-up process is efficient.

Clinicians also need a way of rapidly retrieving the information to which they have been alerted but that has not yet been cemented into their minds. Computer-based resources (especially handheld portable devices) are available that can provide information in less than 30 seconds. To be lifelong learners, physicians have to use tools that help them to hunt and forage through the jungle of information.

Irrelevant information, even if highly valid, is not useful in the scope of a busy daily practice. Sifting through valid, but irrelevant, information wastes valuable information-gathering time. To be relevant and complete, a comprehensive foraging source for new information must contain specialty-specific POEMs (a POEM alert system). The one-stop shopping approach of information for all specialties offered by Best Evidence does not meet this need.

There is a strong impetus to better understand the contemporary experience of dying and to improve end-of-life care in the United States.¹ Effective communication between patients and their care providers has been promoted as a key element within this experience and as a critical factor for maintaining quality of life during end-of-life care.² Multiple factors can impede or facilitate communication in this arena: attitudes toward death and dying,³ provider and patient anxiety over diagnostic and prognostic discussions,⁴ and cultural, socioeconomic, and educational influences.⁵ These factors highlight the nature and context of the provider-patient relationship, which plays a central role in end-of-life medical care for primary care physicians.⁶ Although much emphasis has been given to improving providers’communication skills⁷ and to facilitating decision making,^8-10 there is a paucity of research on the actual dynamics of the provider-patient encounter in end-of-life care.

One theoretical model of physician attitudes and roles in their encounters with dying patients incorporated 3 dimensions: direct involvement with the patient, the physician’s own needs and development, and cooperation with other caregivers.¹¹ Subsequent qualitative research with family physicians supported these dimensions and emphasized the primacy of the patient relationship and the area of physician personal domain.^12,13 However, the applicability of these findings and the model they provide to other specialties and disciplines that are involved in end-of-life care remains unclear. Studies of critical care nurses also identified patient concerns and their own personal concerns as key areas and raised the issue of frustration in their limited role in the management of patients at the end of life.¹⁴ In a study of palliative care workers, most perceived themselves as open and sensitive to their patients, although many felt poorly supported by other staff members.¹⁵

A greater and more complete understanding of the provider’s perspective of the provider-patient encounters in end-of-life care could enhance communication and decision making. This area of inquiry is especially salient in light of the recent Commonwealth-Cummings project on the quality of care at the end of life, which has identified medical provider interventions as one of 4 critical components in a multifaceted framework for a good death.¹ In addition, facilitating the communication and decision-making processes that accompany care at the end of life has been suggested as a key role for family physicians and other generalist providers.¹⁶ We conducted a study using qualitative research methods to explore and describe the nature of provider-patient encounters at the end of life. Our specific aim with this study was to describe factors that both medical and nursing providers identify as influential in their positive and negative encounters with dying patients.

Methods

Design

Because of the exploratory nature of our study, we selected a qualitative research method (semi structured interviews) to gain a richer and more complete description of the multifaceted phenomena that occur between medical and nursing providers and their dying patients. Semi structured interviews are dialogues that are guided and open ended enough to produce transcripts of the discussion as primary data.¹⁷ Editing analysis, in which meaningful units or segments of text are identified into categories or codes that can be used to construct and interpret common themes and patterns, was chosen as a way to generate new understanding in the area of provider perceptions in end-of-life care encounters.¹⁷

Sampling

We used maximum variation sampling to obtain the broadest range of provider perspectives on end-of-life encounters.¹⁸ This approach allowed us to obtain a wide range of information and perspectives on provider perceptions of their encounters with dying patients. We therefore selected a diverse sample of physicians and nurses based on the following criteria:(1) providers in specialties that had the highest probability of encountering dying patients in their practice and (2) providers who had experience in multiple clinical settings, such as academic health centers, private practice, and community settings.

Participants

The Human Subjects Committee of the University of Kansas Medical Center approved our study. We interviewed 6 physicians and 6 nurses from a large Midwestern metropolitan area. The physicians represented the following specialties: geriatric medicine, cardiology, pulmonary medicine/critical care, internal medicine/hospice, oncology, and family practice. The nurse participants were selected from the following specialties: neurology, nephrology/dialysis unit, oncology, hospice, inpatient palliative care, and cardiology/critical care.

The average age of the participants in our sample was 44 years (range=35-54 years);2 were men, and 10 were women. The physician participants had been in practice an average of 10 years (range=5-18 years),while nurse participants had an average of 13 years’(range=5-20 years) experience.

Data Collection

We identified 12 potential participants who met the study criteria and were either known to us professionally or were suggested to us by other colleagues. These participants were contacted and invited to participate. There were no refusals. Semi structured interviews were conducted between September 1998 and April 1999 in either the participant ’s office or the investigator’s office and lasted from 1 hour to 1.5 hours. We conducted individual interviews with participants from our respective disciplines (S. A.F, nurses; T. P.D.,physicians).

Participants were asked to reflect on 2 contrasting end-of-life patient encounters that they had personally experienced in their practices. The encounters were to involve multiple contacts with a single patient who had since died. The first reflection focused on encounters that resulted in a very positive and rewarding outcome from the participant’s perspective. The second reflection involved encounters that were viewed by the participant as personally difficult and troubling. The interview began with the positive reflection, and the participants were allowed to speak freely with few interruptions. An interview guide (Appendix) provided the template for follow-up questions oriented around treatment decisions, the communication process, and participant attitudes, values, and beliefs regarding end-of-life care. All interview sessions were audiotaped and professionally transcribed.

Data Analysis

The interview transcripts were checked for accuracy and verified against the original audiotapes by the investigator who conducted the interview, and the transcripts were then formatted and entered into a qualitative software package (QSR NU*DIST).¹⁹ All data (text, codes, categories, and notes) were entered, retrieved, and analyzed using the computerized software. Independent meaningful units or segments of text that were relevant to the study purpose were identified, labeled, and organized simultaneously by the investigator/analyst in a process called coding.²⁰ After the initial coding was completed, we reviewed the data to evaluate emerging patterns and themes. This iterative process was used to search for systematic relationships and for contrasts and irregularities among the codes and categories, and it continued until consensus or agreement by the analysts was reached. Credibility was maintained by peer debriefing, a search for rival explanations in the data, and by member checking.^18,21 The interdisciplinary nature of the investigators and the iterative coding process enhanced peer debriefing and rival explanation searching, which occurred repeatedly. Member checking was conducted with 4 participants who agreed with the study findings.

Results

Participant interviews were grouped into 6 general categories:(1) provider characteristics,(2) patient characteristics,(3) interpersonal relationships,(4) prognostic certainty/clarity,(5) decision making, and (6) effective communication. Each of these categories is described below with verbatim quotes from the participants.

Provider Characteristics

Participants identified specific provider attitudes, knowledge, and skills as important components of positive encounters. Provider attitudes such as openness to discussions about death and dying, a willingness to be vulnerable, and death acceptance were associated with positive encounters. A heuristic perspective—one that incorporated a receptiveness to learning from patients, past experiences, and even painful mistakes—was inherent in these encounters.

A cardiologist shared his willingness to learn from an elderly patient who challenged his traditional treatment paradigm by refusing a recommendation for bypass surgery:

Maybe my bias was to do what I had been trained to do, which was to send her for bypass, put her through some misery, and maybe not have changed the outcome that much, you know maybe added a year or so, put her through quite a lot of trauma. But she removed my personal bias.

Vulnerability was typified by one participant who related:

This one was hard for me personally…. It was painful for me, because I really liked him, but rewarding on the other hand because I was able to really support him in his autonomous decision to do what he wanted to do.

Death attitudes, whether avoidant or accepting, were also key elements in this category. A nurse participant suggested that death avoidant attitudes were major factors in negative encounters:

But this particular physician can sidestep some of these tough conversations. And that doesn’t make him a bad physician, that’s just his style. He doesn’t do well with those conversations. He’s actually quite a good oncologist, just bad with end-of-life conversations.

A sufficient knowledge base and skill level in caring for dying patients also were important characteristics. Most participants stated their lack of formal education in palliative care as an impediment to providing good care in these settings. As one physician candidly shared, “We learn end-of-life care by the seat of our pants, by the mistakes that we make.”

Patient Characteristics

Patient characteristics such as attitudes, values, and knowledge were important components of participants’perceptions of positive and negative encounters. Within the context of positive encounters, patients were described as proactive, information-seeking, educated, having a clear vision, and focused on reality. Patients in these settings were perceived as facing reality from the onset of diagnosis and being capable of finding something positive in the face of death.

He [the patient] realized that not everything in life is fixable. He said, “This is no tragedy [death]; this is what happens in life.”But as treatments didn’t do any good anymore, he made a decision. He said, “I’ve got to get out and enjoy life.”

In contrast, patients in the negative encounters were perceived as fearful, demanding, untrusting, continually searching for ways to fight the disease, unwilling to give up, and living in chaotic family systems. One patient who had battled cancer for 6 years and spent the last weeks of his life in the intensive care unit was described by a nurse participant:

He [patient] did not want to give up, wanted to be positive, did not want to hear a negative thought, a negative piece of information, and had a physician who kind of went along with that …[which] created a big problem because here you have a man who’s dying basically, and they [family] don’t want to hear it. So that’s probably the problem in a nutshell.

Interpersonal Relationships

Interpersonal relationships were networks of family members, friends, and care providers that either facilitated or impeded the encounters. Participants identified multiple elements in positive encounters: the provider’s sense of connection and a congruent belief system with the patient, patients with supportive relationships with family members and friends, and patients with an existential or spiritual support system. The provider’s sense of connection was captured by an oncologist who related: “We connected on a personality level. I was a shepherd of his, so to speak…personally. I connected with him probably more than I do with most people.”

These positive relationships were also strengthened by common belief systems that were inclusive of patients’ existential or spiritual beliefs. One participant stated, “We could talk about spiritual issues, and maybe that was the thing that bonded us.”Participants also identified coherent and consistent patient belief systems in their positive encounters, “He talked openly about dying and that he wasn’t afraid. He felt that spiritually he was ready.” Also, in their depictions of positive encounters participants cited families and friends who provided emotional support yet allowed for the patient’s wishes to be honored.

Negative encounters were highlighted by strained uncomfortable provider-patient relationships characterized by a lack of trust. Participants identified patients and family members in conflict, hindering supportive relationships. These strained relationships were marked by a lack of acceptance of diagnostic and prognostic information by both patients and families. This lack of acceptance would be manifested in anger and hostility toward providers as one participant related:

He would burst into tears and say, ‘you can’t just let her die. You have to do everything.’ So there was never a no code on this woman even though she was clearly end stage.

Participants cited the lack of an existential orientation or spiritual reference in many patients with whom they had a negative encounter. An oncology nurse typified this perspective:

Some people are clearly afraid to die. I have seen the people who don’t seem to have much of a faith base, not religiously speaking but spiritually. I mean, if their heart and soul and mind aren’t all connected, and they don’t have a sense of purpose, of a beginning and an end, those people are very frightened to die. And some have a religious base, and some don’t. But it’s that sense of the beginning and the end. We all are born, and we’re all going to die. And people who don’t have that sense don’t do well when it comes time to die. Emotionally it’s very difficult for them.

Prognostic Certainty/Clarity

Participants identified the precision of the patient’s diagnosis and prognostic accuracy, as well as the concomitant therapeutic plan, as a facilitating factor in positive encounters. A nursing participant who works with amyotrophic lateral sclerosis (ALS) patients was representative of this orientation:

I guess we start out with the diagnosis basically [following a lengthy diagnostic work-up] and talk to them about what the diagnosis is, that it is a terminal disease.

Another participant shared that prognostic criteria provided clarity and assisted in anticipatory care planning. A pulmonologist/critical care physician shared:

[We] watch the pulmonary function decline, and they turn dyspneic enough—it’s clear, you know—we’ve got great criteria for respiratory failure, and those criteria are very black and white.

In positive encounters, both patients and physicians were able to discern with greater clarity the timing in transitioning care from cure to comfort. For some this was based on the clarity of prognosis, while others used prognosis and an analysis of the burdens and benefits of continued treatment within their conceptualization of quality of life. A family physician related the following:

I was on call, and there was a lady who was from out of town visiting a family, who had a pretty massive stroke…the family thought long and hard about “do we put a feeding tube in her”and the swallowing study showed she could swallow. But she wasn’t able to tell us what she wanted. And she wasn’t making a lot of progress…. They talked with her pastor, who had been her pastor for approximately 40 years…. They decided not to feed her or give her any intravenous fluids. They thought she had a good life, she’s had a lot of faith, she’s told them she’s ready to go and that type of stuff.

Prognostic clarity greatly enhanced positive encounters but was difficult to achieve even for experienced physicians, one of whom stated that, “you can’t outline, you know, the myriad different things that can happen.”However, participants perceived that for family members, clarity of ten was difficult to achieve, especially when interacting with multiple physicians. Patients in intensive care unit settings with multiple specialists and revolving call coverage made prognostic clarity difficult to achieve for some family members. Family members were depicted as misinterpreting different versions of the same story or being provided conflicting information:

He [the patient] had multiple physicians who were trading off and on day to day so there wasn’t the continuity…and so the family heard many different versions of what his prognosis was.

Decision Making

Participants described the degree of unanimity or being on the same page with patients and family members with regard to outcome expectations and treatment decisions as an essential element, one that differentiated positive and negative encounters. Being on the same page was facilitated by shared provider/patient characteristics, such as death attitudes and enriched interpersonal relationships, and by prognostic clarity.

Negative encounters were marked by conflict due to a discongruous assessment or care plan among providers or between family members and providers. One nursing participant described conflict with a physician son who limited the amount of pain medication administered to his dying mother. This unilateral decision to reduce pain medication left the participant feeling conflicted regarding the quality of the patient’s dying:

She had the abdominal surgery and her belly full of cancer, so she had lots of reason for pain, although his [son’s] perception was she had no reason for pain…. The whole thing just, just didn’t feel good to me.

Decision making was an ongoing process that included seeking, sharing, hearing, and processing information within one’s value system. Mismatched expectations for treatment outcomes were inherent in negative encounters and were promoted by a patient or family member’s inability to assimilate and process information accurately. One physician participant shared his perception of unrealistic expectations of family members due to technology:

Perhaps that [high-tech interventions in the intensive care unit] gave them a false sense of security that appropriate therapy had been given, and therefore there was a chance of survival. I did give them statistics saying that, you know, the mortality for this condition is 80%.I think patients whenever they hear that always think they’re the 20% that’s going to survive. They grasp onto the 20% rather than hone in on the adverse.

Effective Communication

Effective communication described the sharing of information in a direct and honest manner by an interdisciplinary team, as well as the exploration of goals, values, and feelings at multiple points in time. It was the primary category cited by all participants. Effective communication can be manifested as an evolutionary process of discussing treatment options at multiple points in time and sometimes over years. Time was a major component of effective communication, time to have complex discussions as well as selecting the optimal time to approach patients and families with difficult issues. One geriatrician noted the link between communication and time: “Communication is difficult and hard, especially if it is a sudden bolt out of the blue. There isn’t time for information to percolate. Families need time.”

Participants related that timing issues were influenced by provider and patient characteristics and by disease prognosis and clarity. of ten discussions were triggered by acute medical events or by ineffective therapeutic interventions. In positive encounters, frequent discussions occurred within the context of an established provider-patient relationship and without a sense of urgency. Communication in negative encounters was characterized by an atmosphere of fear and anger, a lack of trust, and a crisis orientation:

I just think there was no time to get used to the idea that we have a malignancy that is not going to be cured, and if you had the opportunity to build a professional relationship that can be a therapeutic relationship, you can work through some of those issues. But in her case there was absolutely no time to do it. She was extremely frightened at the beginning, and so there was no time to build; there was no trust.

The timing to approach end-of-life care discussions was difficult to judge, and participants had conflicting opinions about whether discussions should occur when patients were still healthy or when there was evidence of a decline. The variability and uncertainty of chronic disease trajectories despite functional status confounded the timing of discussions. An internist who serves as a hospice medical director reflected on a discussion she recently had with a patient:

The good news is you’ve been pretty functional, you feel pretty good. The bad news is you could die any-time. And you know, who is the decision maker ifyou can’t speak for yourself, what are your goals, and how aggressive should we be.

In most positive encounters the participants described spending time with patients and families, a task that was difficult and influenced by the realities of clinical practice in settings that were “overbooked in every slot,”as one participant explained.

A cohesive interdisciplinary team enhanced effective communication. Participants viewed social workers and chaplains as invaluable in assisting patients and families to assimilate information and to facilitate an acceptance of death. A common value system and knowledge base regarding treatment plans among care team members also marked positive encounters:

And some people get it right away and are very realistic. But for most people it takes several conversations over time, and it takes the whole team being on the same page basically. We all have to be saying the same thing.

In contrast, participants identified problems that occurred when interdisciplinary members functioned in independent roles, lacked a team concept and a shared understanding of treatment goals, and did not communicate among themselves.

Discussion

The purpose of our study was to explore the complex phenomena that occur between nursing and physician providers and their dying patients and to describe those factors that providers identify as important in those encounters. The results of our study have both theoretical and practical applications for providers of end-of-life care. Although our original aim was to identify factors, our results suggested that interrelationships existed among the categories. A descriptive conceptual framework was therefore built to graphically depict these relationships (Figure). A conceptual framework explains key categories either graphically or narratively and the presumed relationships among them.²² We constructed an initial framework and several iterations until consensus was reached.

The framework outlines a dynamic reorientation process of both patient and provider norms, values, and behaviors from a curative biomedical perspective to a palliative course that is centered around assisting patients in achieving a quality, comfortable death. This process has been characterized as socialization to dying.²³ In medical settings socialization is inclusive of the content and characteristics of learning social and cultural cues and adaptations, as well as the manner and process by which ideas and ideals are communicated and reinforced.²⁴

Our framework suggests that 4 content domains (provider and patient characteristics, interpersonal relationships, prognostic certainty, and clarity) and 2 process domains (effective communication, and decision making and planning) are key components for understanding the socialization process that providers and patients undergo in end-of-life care. Attitudinal variables such as death acceptance and openness to discussions about death and dying were the major common elements of both provider and patient characteristics in our study. Recent attitudinal research on end-of-life care has focused on the volatile issue of physician-assisted suicide and euthanasia.^25,26 The balance of these studies suggest that demographic and social variables such as sex, education level, and religiosity are tied to both physician and patient attitudes toward death and dying. Religiosity and sex—in addition to mental health status and general health status—are key variables in understanding death attitudes in elderly populations.^27,28

Structuration theory (the construction of meaning through social interaction) provides a useful way to view these findings and the larger current ambivalence and confusion in the United States of how to best understand and situate death.²⁹ This orientation maintains that social determinants such as education, religion, and culture are major elements that facilitate the interpretation and understanding of death and dying.³⁰ If death attitudes serve as proxies for an understanding of death and dying, our results and our framework are congruent with this perspective and have practical implications as well. Socially constructed death attitudes greatly contribute and may be predictive of the end-of-life care experience for both patients and providers. For example, patients or providers with death attitudes characterized by a fear of death or death anxiety may rarely consider—much less begin—the difficult transition from serious illness to dying. As a result, the quality-of-life for these dying patients would be greatly diminished and minimally affected by interventions that are not cognizant of this process. By providing care that is longitudinal, comprehensive, and patient-centered,³¹ family physicians and other primary care physicians are in a unique and advantageous position to assess the impact of social and cultural influences in their dying patients and to incorporate these determinants in their care plans.

Participants in our study cited the quality and character of interpersonal relationships in demarcating positive and negative encounters. Previous qualitative work with patients has also validated the importance of strengthening relationships with loved ones as a domain of quality end-of-life care.³² Our findings suggest that these relationships may be inclusive of provider-patient relationships, as well as an identified spiritual component. According to our study participants, the prognostic clarity of the disease facilitated positive encounters. There is widespread interest in developing and refining prognostic criteria in diseases involving chronic organ disease, although the clinical prediction of nonmalignant disease remains largely ineffective.³³ From a practical viewpoint, our study suggests that primary care providers may consider redirecting their attention away from exacting a disease prognosis toward identifying and enhancing supportive relationships for the patient and developing treatment plans based on personal values and quality of life.

Effective communication is the keystone of the framework and is intertwined with the additional process domain of decision making and planning. Participants amplified several tasks and characteristics of effective communication as central to guiding the patient through the critical transition of curative orientation—in the face of a life-limiting illness—to a dying one. The central place of effective communication in our framework highlights an additional role for family physicians as they care for patients at the end of life. Although disseminating information and empowering patients and family members have been promoted as key functions for providers,³⁴ participants in our study suggest that this process is more inclusive than these tasks. The continuous assessment and identification of patient goals, values, and feelings at multiple time points are functions that are of ten relegated to nonmedical or non-nursing (ie, social work, pastoral care) providers, yet participants in our study cited these responsibilities as vital.

Limitations

There are several limitations to our study. As an exploratory study, the sample size is small and the conceptual framework generally should be considered preliminary and open to modification. Qualitative studies are not intended to be statistically representative of any population but to provide an in-depth examination of complex phenomena. The frequency and validation of participant experiences were not determined. However, the strength of this investigation lies in the in-depth examination and the emerging conceptual framework. Although our categories may be self-evident as factors potentially influencing the outcome of provider-patient encounters in end-of-life care (eg, patient and provider death attitudes),our framework depicts them as a dynamic, complex interaction of modifiable and potentially nonmodifiable factors. Recent initiatives on provider training to improve the quality of care at the end of life (eg, the American Medical Association’s Education for Physicians on End-of-life Care project) are extremely important yet may simplify this complex interaction and the difficulty (or in some cases the impossibility) of assisting a patient’s socialization to dying. Finally, our theoretical framework is based on the provider’s perspective and does not address the patient, family, or caregiver view.

Conclusions

Our conceptual framework, characterized by a socialization process from a curative orientation to a dying orientation that is centered around the process of effective communication, is more inclusive and dynamic than previously described. Future research should test and refine the applicability of this framework and the interventions that facilitate socialization to dying.

Acknowledgments

Our work was supported by the Robert Wood Johnson Foundation Generalist Faculty Scholars Program (T. P.D),the John A. Hartford Foundation (S. A.H.,T. P.D.),and the School of Nursing office of Grants and Research and the Center on Aging at the University of Kansas Medical Center (S. A.H.,T. P.D.).We thank Ann Kuckelman Cobb, RN, PhD, and Anne Walling, MD, for their review of our manuscript.

There is a strong impetus to better understand the contemporary experience of dying and to improve end-of-life care in the United States.¹ Effective communication between patients and their care providers has been promoted as a key element within this experience and as a critical factor for maintaining quality of life during end-of-life care.² Multiple factors can impede or facilitate communication in this arena: attitudes toward death and dying,³ provider and patient anxiety over diagnostic and prognostic discussions,⁴ and cultural, socioeconomic, and educational influences.⁵ These factors highlight the nature and context of the provider-patient relationship, which plays a central role in end-of-life medical care for primary care physicians.⁶ Although much emphasis has been given to improving providers’communication skills⁷ and to facilitating decision making,^8-10 there is a paucity of research on the actual dynamics of the provider-patient encounter in end-of-life care.

One theoretical model of physician attitudes and roles in their encounters with dying patients incorporated 3 dimensions: direct involvement with the patient, the physician’s own needs and development, and cooperation with other caregivers.¹¹ Subsequent qualitative research with family physicians supported these dimensions and emphasized the primacy of the patient relationship and the area of physician personal domain.^12,13 However, the applicability of these findings and the model they provide to other specialties and disciplines that are involved in end-of-life care remains unclear. Studies of critical care nurses also identified patient concerns and their own personal concerns as key areas and raised the issue of frustration in their limited role in the management of patients at the end of life.¹⁴ In a study of palliative care workers, most perceived themselves as open and sensitive to their patients, although many felt poorly supported by other staff members.¹⁵

A greater and more complete understanding of the provider’s perspective of the provider-patient encounters in end-of-life care could enhance communication and decision making. This area of inquiry is especially salient in light of the recent Commonwealth-Cummings project on the quality of care at the end of life, which has identified medical provider interventions as one of 4 critical components in a multifaceted framework for a good death.¹ In addition, facilitating the communication and decision-making processes that accompany care at the end of life has been suggested as a key role for family physicians and other generalist providers.¹⁶ We conducted a study using qualitative research methods to explore and describe the nature of provider-patient encounters at the end of life. Our specific aim with this study was to describe factors that both medical and nursing providers identify as influential in their positive and negative encounters with dying patients.

Methods

Design

Because of the exploratory nature of our study, we selected a qualitative research method (semi structured interviews) to gain a richer and more complete description of the multifaceted phenomena that occur between medical and nursing providers and their dying patients. Semi structured interviews are dialogues that are guided and open ended enough to produce transcripts of the discussion as primary data.¹⁷ Editing analysis, in which meaningful units or segments of text are identified into categories or codes that can be used to construct and interpret common themes and patterns, was chosen as a way to generate new understanding in the area of provider perceptions in end-of-life care encounters.¹⁷

Sampling

We used maximum variation sampling to obtain the broadest range of provider perspectives on end-of-life encounters.¹⁸ This approach allowed us to obtain a wide range of information and perspectives on provider perceptions of their encounters with dying patients. We therefore selected a diverse sample of physicians and nurses based on the following criteria:(1) providers in specialties that had the highest probability of encountering dying patients in their practice and (2) providers who had experience in multiple clinical settings, such as academic health centers, private practice, and community settings.

Participants

The Human Subjects Committee of the University of Kansas Medical Center approved our study. We interviewed 6 physicians and 6 nurses from a large Midwestern metropolitan area. The physicians represented the following specialties: geriatric medicine, cardiology, pulmonary medicine/critical care, internal medicine/hospice, oncology, and family practice. The nurse participants were selected from the following specialties: neurology, nephrology/dialysis unit, oncology, hospice, inpatient palliative care, and cardiology/critical care.

The average age of the participants in our sample was 44 years (range=35-54 years);2 were men, and 10 were women. The physician participants had been in practice an average of 10 years (range=5-18 years),while nurse participants had an average of 13 years’(range=5-20 years) experience.

Data Collection

We identified 12 potential participants who met the study criteria and were either known to us professionally or were suggested to us by other colleagues. These participants were contacted and invited to participate. There were no refusals. Semi structured interviews were conducted between September 1998 and April 1999 in either the participant ’s office or the investigator’s office and lasted from 1 hour to 1.5 hours. We conducted individual interviews with participants from our respective disciplines (S. A.F, nurses; T. P.D.,physicians).

Participants were asked to reflect on 2 contrasting end-of-life patient encounters that they had personally experienced in their practices. The encounters were to involve multiple contacts with a single patient who had since died. The first reflection focused on encounters that resulted in a very positive and rewarding outcome from the participant’s perspective. The second reflection involved encounters that were viewed by the participant as personally difficult and troubling. The interview began with the positive reflection, and the participants were allowed to speak freely with few interruptions. An interview guide (Appendix) provided the template for follow-up questions oriented around treatment decisions, the communication process, and participant attitudes, values, and beliefs regarding end-of-life care. All interview sessions were audiotaped and professionally transcribed.

Data Analysis

The interview transcripts were checked for accuracy and verified against the original audiotapes by the investigator who conducted the interview, and the transcripts were then formatted and entered into a qualitative software package (QSR NU*DIST).¹⁹ All data (text, codes, categories, and notes) were entered, retrieved, and analyzed using the computerized software. Independent meaningful units or segments of text that were relevant to the study purpose were identified, labeled, and organized simultaneously by the investigator/analyst in a process called coding.²⁰ After the initial coding was completed, we reviewed the data to evaluate emerging patterns and themes. This iterative process was used to search for systematic relationships and for contrasts and irregularities among the codes and categories, and it continued until consensus or agreement by the analysts was reached. Credibility was maintained by peer debriefing, a search for rival explanations in the data, and by member checking.^18,21 The interdisciplinary nature of the investigators and the iterative coding process enhanced peer debriefing and rival explanation searching, which occurred repeatedly. Member checking was conducted with 4 participants who agreed with the study findings.

Results

Participant interviews were grouped into 6 general categories:(1) provider characteristics,(2) patient characteristics,(3) interpersonal relationships,(4) prognostic certainty/clarity,(5) decision making, and (6) effective communication. Each of these categories is described below with verbatim quotes from the participants.

Provider Characteristics

Participants identified specific provider attitudes, knowledge, and skills as important components of positive encounters. Provider attitudes such as openness to discussions about death and dying, a willingness to be vulnerable, and death acceptance were associated with positive encounters. A heuristic perspective—one that incorporated a receptiveness to learning from patients, past experiences, and even painful mistakes—was inherent in these encounters.

A cardiologist shared his willingness to learn from an elderly patient who challenged his traditional treatment paradigm by refusing a recommendation for bypass surgery:

Maybe my bias was to do what I had been trained to do, which was to send her for bypass, put her through some misery, and maybe not have changed the outcome that much, you know maybe added a year or so, put her through quite a lot of trauma. But she removed my personal bias.

Vulnerability was typified by one participant who related:

This one was hard for me personally…. It was painful for me, because I really liked him, but rewarding on the other hand because I was able to really support him in his autonomous decision to do what he wanted to do.

Death attitudes, whether avoidant or accepting, were also key elements in this category. A nurse participant suggested that death avoidant attitudes were major factors in negative encounters:

But this particular physician can sidestep some of these tough conversations. And that doesn’t make him a bad physician, that’s just his style. He doesn’t do well with those conversations. He’s actually quite a good oncologist, just bad with end-of-life conversations.

A sufficient knowledge base and skill level in caring for dying patients also were important characteristics. Most participants stated their lack of formal education in palliative care as an impediment to providing good care in these settings. As one physician candidly shared, “We learn end-of-life care by the seat of our pants, by the mistakes that we make.”

Patient Characteristics

Patient characteristics such as attitudes, values, and knowledge were important components of participants’perceptions of positive and negative encounters. Within the context of positive encounters, patients were described as proactive, information-seeking, educated, having a clear vision, and focused on reality. Patients in these settings were perceived as facing reality from the onset of diagnosis and being capable of finding something positive in the face of death.

He [the patient] realized that not everything in life is fixable. He said, “This is no tragedy [death]; this is what happens in life.”But as treatments didn’t do any good anymore, he made a decision. He said, “I’ve got to get out and enjoy life.”

In contrast, patients in the negative encounters were perceived as fearful, demanding, untrusting, continually searching for ways to fight the disease, unwilling to give up, and living in chaotic family systems. One patient who had battled cancer for 6 years and spent the last weeks of his life in the intensive care unit was described by a nurse participant:

He [patient] did not want to give up, wanted to be positive, did not want to hear a negative thought, a negative piece of information, and had a physician who kind of went along with that …[which] created a big problem because here you have a man who’s dying basically, and they [family] don’t want to hear it. So that’s probably the problem in a nutshell.

Interpersonal Relationships

Interpersonal relationships were networks of family members, friends, and care providers that either facilitated or impeded the encounters. Participants identified multiple elements in positive encounters: the provider’s sense of connection and a congruent belief system with the patient, patients with supportive relationships with family members and friends, and patients with an existential or spiritual support system. The provider’s sense of connection was captured by an oncologist who related: “We connected on a personality level. I was a shepherd of his, so to speak…personally. I connected with him probably more than I do with most people.”

These positive relationships were also strengthened by common belief systems that were inclusive of patients’ existential or spiritual beliefs. One participant stated, “We could talk about spiritual issues, and maybe that was the thing that bonded us.”Participants also identified coherent and consistent patient belief systems in their positive encounters, “He talked openly about dying and that he wasn’t afraid. He felt that spiritually he was ready.” Also, in their depictions of positive encounters participants cited families and friends who provided emotional support yet allowed for the patient’s wishes to be honored.

Negative encounters were highlighted by strained uncomfortable provider-patient relationships characterized by a lack of trust. Participants identified patients and family members in conflict, hindering supportive relationships. These strained relationships were marked by a lack of acceptance of diagnostic and prognostic information by both patients and families. This lack of acceptance would be manifested in anger and hostility toward providers as one participant related:

He would burst into tears and say, ‘you can’t just let her die. You have to do everything.’ So there was never a no code on this woman even though she was clearly end stage.

Participants cited the lack of an existential orientation or spiritual reference in many patients with whom they had a negative encounter. An oncology nurse typified this perspective:

Some people are clearly afraid to die. I have seen the people who don’t seem to have much of a faith base, not religiously speaking but spiritually. I mean, if their heart and soul and mind aren’t all connected, and they don’t have a sense of purpose, of a beginning and an end, those people are very frightened to die. And some have a religious base, and some don’t. But it’s that sense of the beginning and the end. We all are born, and we’re all going to die. And people who don’t have that sense don’t do well when it comes time to die. Emotionally it’s very difficult for them.

Prognostic Certainty/Clarity

Participants identified the precision of the patient’s diagnosis and prognostic accuracy, as well as the concomitant therapeutic plan, as a facilitating factor in positive encounters. A nursing participant who works with amyotrophic lateral sclerosis (ALS) patients was representative of this orientation:

I guess we start out with the diagnosis basically [following a lengthy diagnostic work-up] and talk to them about what the diagnosis is, that it is a terminal disease.

Another participant shared that prognostic criteria provided clarity and assisted in anticipatory care planning. A pulmonologist/critical care physician shared:

[We] watch the pulmonary function decline, and they turn dyspneic enough—it’s clear, you know—we’ve got great criteria for respiratory failure, and those criteria are very black and white.

In positive encounters, both patients and physicians were able to discern with greater clarity the timing in transitioning care from cure to comfort. For some this was based on the clarity of prognosis, while others used prognosis and an analysis of the burdens and benefits of continued treatment within their conceptualization of quality of life. A family physician related the following:

I was on call, and there was a lady who was from out of town visiting a family, who had a pretty massive stroke…the family thought long and hard about “do we put a feeding tube in her”and the swallowing study showed she could swallow. But she wasn’t able to tell us what she wanted. And she wasn’t making a lot of progress…. They talked with her pastor, who had been her pastor for approximately 40 years…. They decided not to feed her or give her any intravenous fluids. They thought she had a good life, she’s had a lot of faith, she’s told them she’s ready to go and that type of stuff.

Prognostic clarity greatly enhanced positive encounters but was difficult to achieve even for experienced physicians, one of whom stated that, “you can’t outline, you know, the myriad different things that can happen.”However, participants perceived that for family members, clarity of ten was difficult to achieve, especially when interacting with multiple physicians. Patients in intensive care unit settings with multiple specialists and revolving call coverage made prognostic clarity difficult to achieve for some family members. Family members were depicted as misinterpreting different versions of the same story or being provided conflicting information:

He [the patient] had multiple physicians who were trading off and on day to day so there wasn’t the continuity…and so the family heard many different versions of what his prognosis was.

Decision Making

Participants described the degree of unanimity or being on the same page with patients and family members with regard to outcome expectations and treatment decisions as an essential element, one that differentiated positive and negative encounters. Being on the same page was facilitated by shared provider/patient characteristics, such as death attitudes and enriched interpersonal relationships, and by prognostic clarity.

Negative encounters were marked by conflict due to a discongruous assessment or care plan among providers or between family members and providers. One nursing participant described conflict with a physician son who limited the amount of pain medication administered to his dying mother. This unilateral decision to reduce pain medication left the participant feeling conflicted regarding the quality of the patient’s dying:

She had the abdominal surgery and her belly full of cancer, so she had lots of reason for pain, although his [son’s] perception was she had no reason for pain…. The whole thing just, just didn’t feel good to me.

Decision making was an ongoing process that included seeking, sharing, hearing, and processing information within one’s value system. Mismatched expectations for treatment outcomes were inherent in negative encounters and were promoted by a patient or family member’s inability to assimilate and process information accurately. One physician participant shared his perception of unrealistic expectations of family members due to technology:

Perhaps that [high-tech interventions in the intensive care unit] gave them a false sense of security that appropriate therapy had been given, and therefore there was a chance of survival. I did give them statistics saying that, you know, the mortality for this condition is 80%.I think patients whenever they hear that always think they’re the 20% that’s going to survive. They grasp onto the 20% rather than hone in on the adverse.

Effective Communication

Effective communication described the sharing of information in a direct and honest manner by an interdisciplinary team, as well as the exploration of goals, values, and feelings at multiple points in time. It was the primary category cited by all participants. Effective communication can be manifested as an evolutionary process of discussing treatment options at multiple points in time and sometimes over years. Time was a major component of effective communication, time to have complex discussions as well as selecting the optimal time to approach patients and families with difficult issues. One geriatrician noted the link between communication and time: “Communication is difficult and hard, especially if it is a sudden bolt out of the blue. There isn’t time for information to percolate. Families need time.”

Participants related that timing issues were influenced by provider and patient characteristics and by disease prognosis and clarity. of ten discussions were triggered by acute medical events or by ineffective therapeutic interventions. In positive encounters, frequent discussions occurred within the context of an established provider-patient relationship and without a sense of urgency. Communication in negative encounters was characterized by an atmosphere of fear and anger, a lack of trust, and a crisis orientation:

I just think there was no time to get used to the idea that we have a malignancy that is not going to be cured, and if you had the opportunity to build a professional relationship that can be a therapeutic relationship, you can work through some of those issues. But in her case there was absolutely no time to do it. She was extremely frightened at the beginning, and so there was no time to build; there was no trust.

The timing to approach end-of-life care discussions was difficult to judge, and participants had conflicting opinions about whether discussions should occur when patients were still healthy or when there was evidence of a decline. The variability and uncertainty of chronic disease trajectories despite functional status confounded the timing of discussions. An internist who serves as a hospice medical director reflected on a discussion she recently had with a patient:

The good news is you’ve been pretty functional, you feel pretty good. The bad news is you could die any-time. And you know, who is the decision maker ifyou can’t speak for yourself, what are your goals, and how aggressive should we be.

In most positive encounters the participants described spending time with patients and families, a task that was difficult and influenced by the realities of clinical practice in settings that were “overbooked in every slot,”as one participant explained.

A cohesive interdisciplinary team enhanced effective communication. Participants viewed social workers and chaplains as invaluable in assisting patients and families to assimilate information and to facilitate an acceptance of death. A common value system and knowledge base regarding treatment plans among care team members also marked positive encounters:

And some people get it right away and are very realistic. But for most people it takes several conversations over time, and it takes the whole team being on the same page basically. We all have to be saying the same thing.

In contrast, participants identified problems that occurred when interdisciplinary members functioned in independent roles, lacked a team concept and a shared understanding of treatment goals, and did not communicate among themselves.

Discussion

The purpose of our study was to explore the complex phenomena that occur between nursing and physician providers and their dying patients and to describe those factors that providers identify as important in those encounters. The results of our study have both theoretical and practical applications for providers of end-of-life care. Although our original aim was to identify factors, our results suggested that interrelationships existed among the categories. A descriptive conceptual framework was therefore built to graphically depict these relationships (Figure). A conceptual framework explains key categories either graphically or narratively and the presumed relationships among them.²² We constructed an initial framework and several iterations until consensus was reached.

The framework outlines a dynamic reorientation process of both patient and provider norms, values, and behaviors from a curative biomedical perspective to a palliative course that is centered around assisting patients in achieving a quality, comfortable death. This process has been characterized as socialization to dying.²³ In medical settings socialization is inclusive of the content and characteristics of learning social and cultural cues and adaptations, as well as the manner and process by which ideas and ideals are communicated and reinforced.²⁴

Our framework suggests that 4 content domains (provider and patient characteristics, interpersonal relationships, prognostic certainty, and clarity) and 2 process domains (effective communication, and decision making and planning) are key components for understanding the socialization process that providers and patients undergo in end-of-life care. Attitudinal variables such as death acceptance and openness to discussions about death and dying were the major common elements of both provider and patient characteristics in our study. Recent attitudinal research on end-of-life care has focused on the volatile issue of physician-assisted suicide and euthanasia.^25,26 The balance of these studies suggest that demographic and social variables such as sex, education level, and religiosity are tied to both physician and patient attitudes toward death and dying. Religiosity and sex—in addition to mental health status and general health status—are key variables in understanding death attitudes in elderly populations.^27,28

Structuration theory (the construction of meaning through social interaction) provides a useful way to view these findings and the larger current ambivalence and confusion in the United States of how to best understand and situate death.²⁹ This orientation maintains that social determinants such as education, religion, and culture are major elements that facilitate the interpretation and understanding of death and dying.³⁰ If death attitudes serve as proxies for an understanding of death and dying, our results and our framework are congruent with this perspective and have practical implications as well. Socially constructed death attitudes greatly contribute and may be predictive of the end-of-life care experience for both patients and providers. For example, patients or providers with death attitudes characterized by a fear of death or death anxiety may rarely consider—much less begin—the difficult transition from serious illness to dying. As a result, the quality-of-life for these dying patients would be greatly diminished and minimally affected by interventions that are not cognizant of this process. By providing care that is longitudinal, comprehensive, and patient-centered,³¹ family physicians and other primary care physicians are in a unique and advantageous position to assess the impact of social and cultural influences in their dying patients and to incorporate these determinants in their care plans.

Participants in our study cited the quality and character of interpersonal relationships in demarcating positive and negative encounters. Previous qualitative work with patients has also validated the importance of strengthening relationships with loved ones as a domain of quality end-of-life care.³² Our findings suggest that these relationships may be inclusive of provider-patient relationships, as well as an identified spiritual component. According to our study participants, the prognostic clarity of the disease facilitated positive encounters. There is widespread interest in developing and refining prognostic criteria in diseases involving chronic organ disease, although the clinical prediction of nonmalignant disease remains largely ineffective.³³ From a practical viewpoint, our study suggests that primary care providers may consider redirecting their attention away from exacting a disease prognosis toward identifying and enhancing supportive relationships for the patient and developing treatment plans based on personal values and quality of life.

Effective communication is the keystone of the framework and is intertwined with the additional process domain of decision making and planning. Participants amplified several tasks and characteristics of effective communication as central to guiding the patient through the critical transition of curative orientation—in the face of a life-limiting illness—to a dying one. The central place of effective communication in our framework highlights an additional role for family physicians as they care for patients at the end of life. Although disseminating information and empowering patients and family members have been promoted as key functions for providers,³⁴ participants in our study suggest that this process is more inclusive than these tasks. The continuous assessment and identification of patient goals, values, and feelings at multiple time points are functions that are of ten relegated to nonmedical or non-nursing (ie, social work, pastoral care) providers, yet participants in our study cited these responsibilities as vital.

Limitations

There are several limitations to our study. As an exploratory study, the sample size is small and the conceptual framework generally should be considered preliminary and open to modification. Qualitative studies are not intended to be statistically representative of any population but to provide an in-depth examination of complex phenomena. The frequency and validation of participant experiences were not determined. However, the strength of this investigation lies in the in-depth examination and the emerging conceptual framework. Although our categories may be self-evident as factors potentially influencing the outcome of provider-patient encounters in end-of-life care (eg, patient and provider death attitudes),our framework depicts them as a dynamic, complex interaction of modifiable and potentially nonmodifiable factors. Recent initiatives on provider training to improve the quality of care at the end of life (eg, the American Medical Association’s Education for Physicians on End-of-life Care project) are extremely important yet may simplify this complex interaction and the difficulty (or in some cases the impossibility) of assisting a patient’s socialization to dying. Finally, our theoretical framework is based on the provider’s perspective and does not address the patient, family, or caregiver view.

Conclusions

Our conceptual framework, characterized by a socialization process from a curative orientation to a dying orientation that is centered around the process of effective communication, is more inclusive and dynamic than previously described. Future research should test and refine the applicability of this framework and the interventions that facilitate socialization to dying.

Acknowledgments

Our work was supported by the Robert Wood Johnson Foundation Generalist Faculty Scholars Program (T. P.D),the John A. Hartford Foundation (S. A.H.,T. P.D.),and the School of Nursing office of Grants and Research and the Center on Aging at the University of Kansas Medical Center (S. A.H.,T. P.D.).We thank Ann Kuckelman Cobb, RN, PhD, and Anne Walling, MD, for their review of our manuscript.

There is a strong impetus to better understand the contemporary experience of dying and to improve end-of-life care in the United States.¹ Effective communication between patients and their care providers has been promoted as a key element within this experience and as a critical factor for maintaining quality of life during end-of-life care.² Multiple factors can impede or facilitate communication in this arena: attitudes toward death and dying,³ provider and patient anxiety over diagnostic and prognostic discussions,⁴ and cultural, socioeconomic, and educational influences.⁵ These factors highlight the nature and context of the provider-patient relationship, which plays a central role in end-of-life medical care for primary care physicians.⁶ Although much emphasis has been given to improving providers’communication skills⁷ and to facilitating decision making,^8-10 there is a paucity of research on the actual dynamics of the provider-patient encounter in end-of-life care.

One theoretical model of physician attitudes and roles in their encounters with dying patients incorporated 3 dimensions: direct involvement with the patient, the physician’s own needs and development, and cooperation with other caregivers.¹¹ Subsequent qualitative research with family physicians supported these dimensions and emphasized the primacy of the patient relationship and the area of physician personal domain.^12,13 However, the applicability of these findings and the model they provide to other specialties and disciplines that are involved in end-of-life care remains unclear. Studies of critical care nurses also identified patient concerns and their own personal concerns as key areas and raised the issue of frustration in their limited role in the management of patients at the end of life.¹⁴ In a study of palliative care workers, most perceived themselves as open and sensitive to their patients, although many felt poorly supported by other staff members.¹⁵

A greater and more complete understanding of the provider’s perspective of the provider-patient encounters in end-of-life care could enhance communication and decision making. This area of inquiry is especially salient in light of the recent Commonwealth-Cummings project on the quality of care at the end of life, which has identified medical provider interventions as one of 4 critical components in a multifaceted framework for a good death.¹ In addition, facilitating the communication and decision-making processes that accompany care at the end of life has been suggested as a key role for family physicians and other generalist providers.¹⁶ We conducted a study using qualitative research methods to explore and describe the nature of provider-patient encounters at the end of life. Our specific aim with this study was to describe factors that both medical and nursing providers identify as influential in their positive and negative encounters with dying patients.

Methods

Design

Because of the exploratory nature of our study, we selected a qualitative research method (semi structured interviews) to gain a richer and more complete description of the multifaceted phenomena that occur between medical and nursing providers and their dying patients. Semi structured interviews are dialogues that are guided and open ended enough to produce transcripts of the discussion as primary data.¹⁷ Editing analysis, in which meaningful units or segments of text are identified into categories or codes that can be used to construct and interpret common themes and patterns, was chosen as a way to generate new understanding in the area of provider perceptions in end-of-life care encounters.¹⁷

Sampling

We used maximum variation sampling to obtain the broadest range of provider perspectives on end-of-life encounters.¹⁸ This approach allowed us to obtain a wide range of information and perspectives on provider perceptions of their encounters with dying patients. We therefore selected a diverse sample of physicians and nurses based on the following criteria:(1) providers in specialties that had the highest probability of encountering dying patients in their practice and (2) providers who had experience in multiple clinical settings, such as academic health centers, private practice, and community settings.

Participants

The Human Subjects Committee of the University of Kansas Medical Center approved our study. We interviewed 6 physicians and 6 nurses from a large Midwestern metropolitan area. The physicians represented the following specialties: geriatric medicine, cardiology, pulmonary medicine/critical care, internal medicine/hospice, oncology, and family practice. The nurse participants were selected from the following specialties: neurology, nephrology/dialysis unit, oncology, hospice, inpatient palliative care, and cardiology/critical care.

The average age of the participants in our sample was 44 years (range=35-54 years);2 were men, and 10 were women. The physician participants had been in practice an average of 10 years (range=5-18 years),while nurse participants had an average of 13 years’(range=5-20 years) experience.

Data Collection

We identified 12 potential participants who met the study criteria and were either known to us professionally or were suggested to us by other colleagues. These participants were contacted and invited to participate. There were no refusals. Semi structured interviews were conducted between September 1998 and April 1999 in either the participant ’s office or the investigator’s office and lasted from 1 hour to 1.5 hours. We conducted individual interviews with participants from our respective disciplines (S. A.F, nurses; T. P.D.,physicians).

Participants were asked to reflect on 2 contrasting end-of-life patient encounters that they had personally experienced in their practices. The encounters were to involve multiple contacts with a single patient who had since died. The first reflection focused on encounters that resulted in a very positive and rewarding outcome from the participant’s perspective. The second reflection involved encounters that were viewed by the participant as personally difficult and troubling. The interview began with the positive reflection, and the participants were allowed to speak freely with few interruptions. An interview guide (Appendix) provided the template for follow-up questions oriented around treatment decisions, the communication process, and participant attitudes, values, and beliefs regarding end-of-life care. All interview sessions were audiotaped and professionally transcribed.

Data Analysis

The interview transcripts were checked for accuracy and verified against the original audiotapes by the investigator who conducted the interview, and the transcripts were then formatted and entered into a qualitative software package (QSR NU*DIST).¹⁹ All data (text, codes, categories, and notes) were entered, retrieved, and analyzed using the computerized software. Independent meaningful units or segments of text that were relevant to the study purpose were identified, labeled, and organized simultaneously by the investigator/analyst in a process called coding.²⁰ After the initial coding was completed, we reviewed the data to evaluate emerging patterns and themes. This iterative process was used to search for systematic relationships and for contrasts and irregularities among the codes and categories, and it continued until consensus or agreement by the analysts was reached. Credibility was maintained by peer debriefing, a search for rival explanations in the data, and by member checking.^18,21 The interdisciplinary nature of the investigators and the iterative coding process enhanced peer debriefing and rival explanation searching, which occurred repeatedly. Member checking was conducted with 4 participants who agreed with the study findings.

Results

Participant interviews were grouped into 6 general categories:(1) provider characteristics,(2) patient characteristics,(3) interpersonal relationships,(4) prognostic certainty/clarity,(5) decision making, and (6) effective communication. Each of these categories is described below with verbatim quotes from the participants.

Provider Characteristics

Participants identified specific provider attitudes, knowledge, and skills as important components of positive encounters. Provider attitudes such as openness to discussions about death and dying, a willingness to be vulnerable, and death acceptance were associated with positive encounters. A heuristic perspective—one that incorporated a receptiveness to learning from patients, past experiences, and even painful mistakes—was inherent in these encounters.

A cardiologist shared his willingness to learn from an elderly patient who challenged his traditional treatment paradigm by refusing a recommendation for bypass surgery:

Maybe my bias was to do what I had been trained to do, which was to send her for bypass, put her through some misery, and maybe not have changed the outcome that much, you know maybe added a year or so, put her through quite a lot of trauma. But she removed my personal bias.

Vulnerability was typified by one participant who related:

This one was hard for me personally…. It was painful for me, because I really liked him, but rewarding on the other hand because I was able to really support him in his autonomous decision to do what he wanted to do.

Death attitudes, whether avoidant or accepting, were also key elements in this category. A nurse participant suggested that death avoidant attitudes were major factors in negative encounters:

But this particular physician can sidestep some of these tough conversations. And that doesn’t make him a bad physician, that’s just his style. He doesn’t do well with those conversations. He’s actually quite a good oncologist, just bad with end-of-life conversations.

A sufficient knowledge base and skill level in caring for dying patients also were important characteristics. Most participants stated their lack of formal education in palliative care as an impediment to providing good care in these settings. As one physician candidly shared, “We learn end-of-life care by the seat of our pants, by the mistakes that we make.”

Patient Characteristics

Patient characteristics such as attitudes, values, and knowledge were important components of participants’perceptions of positive and negative encounters. Within the context of positive encounters, patients were described as proactive, information-seeking, educated, having a clear vision, and focused on reality. Patients in these settings were perceived as facing reality from the onset of diagnosis and being capable of finding something positive in the face of death.

He [the patient] realized that not everything in life is fixable. He said, “This is no tragedy [death]; this is what happens in life.”But as treatments didn’t do any good anymore, he made a decision. He said, “I’ve got to get out and enjoy life.”

In contrast, patients in the negative encounters were perceived as fearful, demanding, untrusting, continually searching for ways to fight the disease, unwilling to give up, and living in chaotic family systems. One patient who had battled cancer for 6 years and spent the last weeks of his life in the intensive care unit was described by a nurse participant:

He [patient] did not want to give up, wanted to be positive, did not want to hear a negative thought, a negative piece of information, and had a physician who kind of went along with that …[which] created a big problem because here you have a man who’s dying basically, and they [family] don’t want to hear it. So that’s probably the problem in a nutshell.

Interpersonal Relationships

Interpersonal relationships were networks of family members, friends, and care providers that either facilitated or impeded the encounters. Participants identified multiple elements in positive encounters: the provider’s sense of connection and a congruent belief system with the patient, patients with supportive relationships with family members and friends, and patients with an existential or spiritual support system. The provider’s sense of connection was captured by an oncologist who related: “We connected on a personality level. I was a shepherd of his, so to speak…personally. I connected with him probably more than I do with most people.”

These positive relationships were also strengthened by common belief systems that were inclusive of patients’ existential or spiritual beliefs. One participant stated, “We could talk about spiritual issues, and maybe that was the thing that bonded us.”Participants also identified coherent and consistent patient belief systems in their positive encounters, “He talked openly about dying and that he wasn’t afraid. He felt that spiritually he was ready.” Also, in their depictions of positive encounters participants cited families and friends who provided emotional support yet allowed for the patient’s wishes to be honored.

Negative encounters were highlighted by strained uncomfortable provider-patient relationships characterized by a lack of trust. Participants identified patients and family members in conflict, hindering supportive relationships. These strained relationships were marked by a lack of acceptance of diagnostic and prognostic information by both patients and families. This lack of acceptance would be manifested in anger and hostility toward providers as one participant related:

He would burst into tears and say, ‘you can’t just let her die. You have to do everything.’ So there was never a no code on this woman even though she was clearly end stage.

Participants cited the lack of an existential orientation or spiritual reference in many patients with whom they had a negative encounter. An oncology nurse typified this perspective:

Some people are clearly afraid to die. I have seen the people who don’t seem to have much of a faith base, not religiously speaking but spiritually. I mean, if their heart and soul and mind aren’t all connected, and they don’t have a sense of purpose, of a beginning and an end, those people are very frightened to die. And some have a religious base, and some don’t. But it’s that sense of the beginning and the end. We all are born, and we’re all going to die. And people who don’t have that sense don’t do well when it comes time to die. Emotionally it’s very difficult for them.

Prognostic Certainty/Clarity

Participants identified the precision of the patient’s diagnosis and prognostic accuracy, as well as the concomitant therapeutic plan, as a facilitating factor in positive encounters. A nursing participant who works with amyotrophic lateral sclerosis (ALS) patients was representative of this orientation:

I guess we start out with the diagnosis basically [following a lengthy diagnostic work-up] and talk to them about what the diagnosis is, that it is a terminal disease.

Another participant shared that prognostic criteria provided clarity and assisted in anticipatory care planning. A pulmonologist/critical care physician shared:

[We] watch the pulmonary function decline, and they turn dyspneic enough—it’s clear, you know—we’ve got great criteria for respiratory failure, and those criteria are very black and white.

In positive encounters, both patients and physicians were able to discern with greater clarity the timing in transitioning care from cure to comfort. For some this was based on the clarity of prognosis, while others used prognosis and an analysis of the burdens and benefits of continued treatment within their conceptualization of quality of life. A family physician related the following:

I was on call, and there was a lady who was from out of town visiting a family, who had a pretty massive stroke…the family thought long and hard about “do we put a feeding tube in her”and the swallowing study showed she could swallow. But she wasn’t able to tell us what she wanted. And she wasn’t making a lot of progress…. They talked with her pastor, who had been her pastor for approximately 40 years…. They decided not to feed her or give her any intravenous fluids. They thought she had a good life, she’s had a lot of faith, she’s told them she’s ready to go and that type of stuff.

Prognostic clarity greatly enhanced positive encounters but was difficult to achieve even for experienced physicians, one of whom stated that, “you can’t outline, you know, the myriad different things that can happen.”However, participants perceived that for family members, clarity of ten was difficult to achieve, especially when interacting with multiple physicians. Patients in intensive care unit settings with multiple specialists and revolving call coverage made prognostic clarity difficult to achieve for some family members. Family members were depicted as misinterpreting different versions of the same story or being provided conflicting information:

He [the patient] had multiple physicians who were trading off and on day to day so there wasn’t the continuity…and so the family heard many different versions of what his prognosis was.

Decision Making

Participants described the degree of unanimity or being on the same page with patients and family members with regard to outcome expectations and treatment decisions as an essential element, one that differentiated positive and negative encounters. Being on the same page was facilitated by shared provider/patient characteristics, such as death attitudes and enriched interpersonal relationships, and by prognostic clarity.

Negative encounters were marked by conflict due to a discongruous assessment or care plan among providers or between family members and providers. One nursing participant described conflict with a physician son who limited the amount of pain medication administered to his dying mother. This unilateral decision to reduce pain medication left the participant feeling conflicted regarding the quality of the patient’s dying:

She had the abdominal surgery and her belly full of cancer, so she had lots of reason for pain, although his [son’s] perception was she had no reason for pain…. The whole thing just, just didn’t feel good to me.

Decision making was an ongoing process that included seeking, sharing, hearing, and processing information within one’s value system. Mismatched expectations for treatment outcomes were inherent in negative encounters and were promoted by a patient or family member’s inability to assimilate and process information accurately. One physician participant shared his perception of unrealistic expectations of family members due to technology:

Perhaps that [high-tech interventions in the intensive care unit] gave them a false sense of security that appropriate therapy had been given, and therefore there was a chance of survival. I did give them statistics saying that, you know, the mortality for this condition is 80%.I think patients whenever they hear that always think they’re the 20% that’s going to survive. They grasp onto the 20% rather than hone in on the adverse.

Effective Communication

Effective communication described the sharing of information in a direct and honest manner by an interdisciplinary team, as well as the exploration of goals, values, and feelings at multiple points in time. It was the primary category cited by all participants. Effective communication can be manifested as an evolutionary process of discussing treatment options at multiple points in time and sometimes over years. Time was a major component of effective communication, time to have complex discussions as well as selecting the optimal time to approach patients and families with difficult issues. One geriatrician noted the link between communication and time: “Communication is difficult and hard, especially if it is a sudden bolt out of the blue. There isn’t time for information to percolate. Families need time.”

Participants related that timing issues were influenced by provider and patient characteristics and by disease prognosis and clarity. of ten discussions were triggered by acute medical events or by ineffective therapeutic interventions. In positive encounters, frequent discussions occurred within the context of an established provider-patient relationship and without a sense of urgency. Communication in negative encounters was characterized by an atmosphere of fear and anger, a lack of trust, and a crisis orientation:

I just think there was no time to get used to the idea that we have a malignancy that is not going to be cured, and if you had the opportunity to build a professional relationship that can be a therapeutic relationship, you can work through some of those issues. But in her case there was absolutely no time to do it. She was extremely frightened at the beginning, and so there was no time to build; there was no trust.

The timing to approach end-of-life care discussions was difficult to judge, and participants had conflicting opinions about whether discussions should occur when patients were still healthy or when there was evidence of a decline. The variability and uncertainty of chronic disease trajectories despite functional status confounded the timing of discussions. An internist who serves as a hospice medical director reflected on a discussion she recently had with a patient:

The good news is you’ve been pretty functional, you feel pretty good. The bad news is you could die any-time. And you know, who is the decision maker ifyou can’t speak for yourself, what are your goals, and how aggressive should we be.

In most positive encounters the participants described spending time with patients and families, a task that was difficult and influenced by the realities of clinical practice in settings that were “overbooked in every slot,”as one participant explained.

A cohesive interdisciplinary team enhanced effective communication. Participants viewed social workers and chaplains as invaluable in assisting patients and families to assimilate information and to facilitate an acceptance of death. A common value system and knowledge base regarding treatment plans among care team members also marked positive encounters:

And some people get it right away and are very realistic. But for most people it takes several conversations over time, and it takes the whole team being on the same page basically. We all have to be saying the same thing.

In contrast, participants identified problems that occurred when interdisciplinary members functioned in independent roles, lacked a team concept and a shared understanding of treatment goals, and did not communicate among themselves.

Discussion

The purpose of our study was to explore the complex phenomena that occur between nursing and physician providers and their dying patients and to describe those factors that providers identify as important in those encounters. The results of our study have both theoretical and practical applications for providers of end-of-life care. Although our original aim was to identify factors, our results suggested that interrelationships existed among the categories. A descriptive conceptual framework was therefore built to graphically depict these relationships (Figure). A conceptual framework explains key categories either graphically or narratively and the presumed relationships among them.²² We constructed an initial framework and several iterations until consensus was reached.

The framework outlines a dynamic reorientation process of both patient and provider norms, values, and behaviors from a curative biomedical perspective to a palliative course that is centered around assisting patients in achieving a quality, comfortable death. This process has been characterized as socialization to dying.²³ In medical settings socialization is inclusive of the content and characteristics of learning social and cultural cues and adaptations, as well as the manner and process by which ideas and ideals are communicated and reinforced.²⁴

Our framework suggests that 4 content domains (provider and patient characteristics, interpersonal relationships, prognostic certainty, and clarity) and 2 process domains (effective communication, and decision making and planning) are key components for understanding the socialization process that providers and patients undergo in end-of-life care. Attitudinal variables such as death acceptance and openness to discussions about death and dying were the major common elements of both provider and patient characteristics in our study. Recent attitudinal research on end-of-life care has focused on the volatile issue of physician-assisted suicide and euthanasia.^25,26 The balance of these studies suggest that demographic and social variables such as sex, education level, and religiosity are tied to both physician and patient attitudes toward death and dying. Religiosity and sex—in addition to mental health status and general health status—are key variables in understanding death attitudes in elderly populations.^27,28

Structuration theory (the construction of meaning through social interaction) provides a useful way to view these findings and the larger current ambivalence and confusion in the United States of how to best understand and situate death.²⁹ This orientation maintains that social determinants such as education, religion, and culture are major elements that facilitate the interpretation and understanding of death and dying.³⁰ If death attitudes serve as proxies for an understanding of death and dying, our results and our framework are congruent with this perspective and have practical implications as well. Socially constructed death attitudes greatly contribute and may be predictive of the end-of-life care experience for both patients and providers. For example, patients or providers with death attitudes characterized by a fear of death or death anxiety may rarely consider—much less begin—the difficult transition from serious illness to dying. As a result, the quality-of-life for these dying patients would be greatly diminished and minimally affected by interventions that are not cognizant of this process. By providing care that is longitudinal, comprehensive, and patient-centered,³¹ family physicians and other primary care physicians are in a unique and advantageous position to assess the impact of social and cultural influences in their dying patients and to incorporate these determinants in their care plans.

Participants in our study cited the quality and character of interpersonal relationships in demarcating positive and negative encounters. Previous qualitative work with patients has also validated the importance of strengthening relationships with loved ones as a domain of quality end-of-life care.³² Our findings suggest that these relationships may be inclusive of provider-patient relationships, as well as an identified spiritual component. According to our study participants, the prognostic clarity of the disease facilitated positive encounters. There is widespread interest in developing and refining prognostic criteria in diseases involving chronic organ disease, although the clinical prediction of nonmalignant disease remains largely ineffective.³³ From a practical viewpoint, our study suggests that primary care providers may consider redirecting their attention away from exacting a disease prognosis toward identifying and enhancing supportive relationships for the patient and developing treatment plans based on personal values and quality of life.

Effective communication is the keystone of the framework and is intertwined with the additional process domain of decision making and planning. Participants amplified several tasks and characteristics of effective communication as central to guiding the patient through the critical transition of curative orientation—in the face of a life-limiting illness—to a dying one. The central place of effective communication in our framework highlights an additional role for family physicians as they care for patients at the end of life. Although disseminating information and empowering patients and family members have been promoted as key functions for providers,³⁴ participants in our study suggest that this process is more inclusive than these tasks. The continuous assessment and identification of patient goals, values, and feelings at multiple time points are functions that are of ten relegated to nonmedical or non-nursing (ie, social work, pastoral care) providers, yet participants in our study cited these responsibilities as vital.

Limitations

There are several limitations to our study. As an exploratory study, the sample size is small and the conceptual framework generally should be considered preliminary and open to modification. Qualitative studies are not intended to be statistically representative of any population but to provide an in-depth examination of complex phenomena. The frequency and validation of participant experiences were not determined. However, the strength of this investigation lies in the in-depth examination and the emerging conceptual framework. Although our categories may be self-evident as factors potentially influencing the outcome of provider-patient encounters in end-of-life care (eg, patient and provider death attitudes),our framework depicts them as a dynamic, complex interaction of modifiable and potentially nonmodifiable factors. Recent initiatives on provider training to improve the quality of care at the end of life (eg, the American Medical Association’s Education for Physicians on End-of-life Care project) are extremely important yet may simplify this complex interaction and the difficulty (or in some cases the impossibility) of assisting a patient’s socialization to dying. Finally, our theoretical framework is based on the provider’s perspective and does not address the patient, family, or caregiver view.

Conclusions

Our conceptual framework, characterized by a socialization process from a curative orientation to a dying orientation that is centered around the process of effective communication, is more inclusive and dynamic than previously described. Future research should test and refine the applicability of this framework and the interventions that facilitate socialization to dying.

Acknowledgments

Our work was supported by the Robert Wood Johnson Foundation Generalist Faculty Scholars Program (T. P.D),the John A. Hartford Foundation (S. A.H.,T. P.D.),and the School of Nursing office of Grants and Research and the Center on Aging at the University of Kansas Medical Center (S. A.H.,T. P.D.).We thank Ann Kuckelman Cobb, RN, PhD, and Anne Walling, MD, for their review of our manuscript.

Agrowing body of literature at both individual and ecologic levels has demonstrated the association of primary care and health outcomes.^1-11 Franks and Fiscella,¹² using nationally representative survey data, showed that adult respondents who reported a primary care physician rather than a specialist as their regular source of care had lower subsequent mortality and lower annual health care costs after controlling for differences in demographic characteristics, health insurance status, health perceptions, reported diagnoses, and smoking status. Both Shi^4,6 and Farmer and collegues¹³ found better health outcomes in states with higher primary care physician-population ratios after controlling for sociodemographic measures (% elderly, % urban, % minority, education, income, unemployment, pollution) and lifestyle factors (seatbelt usage, obesity, and smoking). Recent studies further showed that primary care may mitigate the adverse effects of income inequality on health.^14-16 Taken individually, each of the main features of primary care (person-focused care over time, accessible care, comprehensive in the sense of meeting all common health needs, and coordination when people have to receive services elsewhere) are known to improve both the effectiveness as well as the efficiency of care.^1,7,17-24

The mounting evidence associating primary care with improved health outcome has led to a rapid increase in interest in assessing primary care achievement by consumers and patients.^{18,19,21,25-28} Despite its importance, there currently is no way to assess the extent to which people receive adequate primary care; receiving care from a physician or physician designated as a primary care physician is at best only a proxy for actual adequacy of provision of primary care services. As a result, there are efforts to develop instruments that directly assess the adequacy of primary care.^20,29,30

The Primary Care Assessment Tool (PCAT) instruments developed by The Johns Hopkins Primary Care Policy Center for Underserved Populations were designed to measure the extent and quality of primary care services at a provider setting designated by consumers as their main source of general care and consistent with a focus on attributes of primary care that have been demonstrated to produce better outcomes of care at lower costs.²² The PCATfamily of instruments includes the Child Consumer/Client Survey, the Adult Consumer/Client Survey, and the Facility/Provider Survey. All surveys are based on self-report by patients or providers. The Consumer/Client Survey (both adult and child editions) is designed to collect information from consumers or family caretakers regarding their experience using health care resources. It may be used to survey target populations as defined by geography (community surveys), health plans, sites of care, payment mechanisms, or specific health care needs. The survey, which takes approximately 40 minutes to complete, can be administered through either telephone or face-to-face interviews, or by mail. Ahigh school reading level is required to self-administer the questionnaire. The Facility/Provider Survey is designed to collect information about specific operational characteristics and practices related to providing primary care from the viewpoint of practitioners, clinics, group practices, and institutions. This survey can also be implemented either by mail or by face-to-face or telephone interviews. It is parallel in content to the consumer/client survey. All 3 instruments are available for general use on request.

We report on the validation of the Consumer/Client Primary Care Assessment Tool Adult Edition (PCAT-AE). Its companion instrument for children (PCAT-CE) was previously validated.³⁰ Specifically, we assessed the congruence between the theoretically derived measures and the empiric results in terms of the underlying structure of the principal primary care domains within a diverse sample of populations including health maintenance organization (HMO) members and community health center (CHC) users. The validation process also served to reduce the number of items needed to assess the adequacy of primary care.

Methods

Subjects

The study participants were members of 2 health plans in 2 counties of South Carolina. Both counties are part of Columbia, the states capital and third largest city. One of the health plans (referred to as HMO) is licensed as an independent practice association (IPA) HMO model, in which primary care physicians act as gatekeepers and health care managers. Referral to specialists must be made through primary care physicians, and specialists must be affiliated with the HMO. The primary market has been large group employers, including employees of the state agencies and national and regional companies. Members of this plan are primarily from middle-income households. The other health plan (referred to as CHC) is a coalition of 12 Columbia-based health and social services provider organizations, including the county hospital, health department, department of social services, community health centers, and other social service agencies that provide services to lower income persons, such as Medicaid recipients and low-income households. These 2 plans were selected because they represent typical South Carolina managed care organizations and health plans for low-income individuals, respectively. Samples drawn from these 2 plans allowed us to test the reliability of PCATwith a diverse sample of populations, including both middle-income and low-income individuals using regular physician offices and community health centers, respectively.

Estimation of the sample size for this study involved several steps. First, an estimate of the likely proportions or means and standard deviations for each primary care measure was derived from a previous study.²⁵ When data were not available, a conservative estimate (eg, a larger standard deviation or proportion closer to 50/50) was made. Second, the estimates of the proportions, means, and standard deviations for the dependent variables were entered into the standard sample size formula for a two-group, cross-sectional sample. Using a 95% confidence interval, the largest sample size required was 300 per group. The CHC group was oversampled because of additional planned within-group analyses (not the focus of this paper). Finally, the desired sample size was adjusted for anticipated survey nonresponse (anticipated to be higher for a mail survey than a face-to-face interview).

For the HMO group, a mail survey was used since it was deemed most efficient. In 2 previous longitudinal studies of the same HMO, we used mail survey and telephone interviews alternately with a cohort of HMO members and obtained comparable results.^31,32 For this study, we sent a letter with a PCAT-AE questionnaire to 1000 randomly selected adult members to invite them to participate in the project. Because of known frequent changes in addresses, we recruited the non-HMO plan individuals and conducted in-person interviews at all the community health center sites where members came to the clinics for non-urgent visits. Patients were systematically approached while waiting for their scheduled appointment (ie, every nth patient based on expected visits for a particular site) and recruited for the study during a period of 4 weeks for each site.

Measures

Identification of Primary Care Source.Three questions were developed to identify an individuals usual source of care and the strength of that affiliation: (1) Is there a doctor or place that you usually go if you are sick or need advice about your health? (usual source), (2) Is there a doctor or place that knows you best as a person? (knows best), and (3) Is there a doctor or place that is most responsible for your health care? (most responsible). Aperson was considered to have a usual source of care if he or she answered positively to any 1 of the 3 questions (95% for the HMO plan and 90% for the low-income plan). Anegative answer to all 3 questions rendered the individual as not having a usual source of care.

An algorithm based on response to these 3 questions identified the strength of affiliation with the primary care source. If all 3 physicians/places were the same, this was considered evidence of a strong affiliation. If the response to the usual source question was the same as for either of the other 2 questions then that site was used although the affiliation was considered less strong. If the response for a usual source question was different from the other 2 responses but the other 2 responses were the same, then the site where both were the same was used (weak affiliation). If all 3 responses were different (weakest affiliation), then the site identified for usual source was used. All subsequent questions asked about this specific person or place. For those with no identifiable source of primary care, subsequent questions were asked about the last place that was visited.

Domains of Primary Care.The PCAT-AE was modeled on the previously validated PCAT-CE and is consistent with the 1978 Institute of Medicine (IOM) definition of primary care as accessibility, comprehensiveness, coordination, continuity, and accountability³³ and with the 1996 IOM report definition of primary care as the provision of integrated, accessible health care services by clinicians who are accountable for addressing a large majority of personal health care needs, developing a sustained partnership with patients, and practicing in the context of family and the community.³⁴ When combined into scales, the PCATsurvey items dealing with primary care quality were designed to measure each of the core domains of primary care; that is, first contact, longitudinality, comprehensiveness, and coordination (definitions of the primary care domains are provided in the Appendix).

Nine experts were asked to rate the appropriateness and representativeness of the primary care domain items. These experts consisted of 3 policymakers in federal agencies, 2 directors of community pediatrics at major medical centers, a health research director at a major HMO, 2 family medicine professors, and a general internal medicine physician. Acard sorting technique was used to determine the degree of congruence between each item and the domain it was designed to measure. Each survey question with its response categories and descriptions of each of the primary care domains was printed on separate index cards and mailed to the experts who assigned each question to one of the defined domains and suggested revisions and/or addition of other items. The percent agreement among the experts was used to determine the degree of congruence on the placement of each item in a particular domain. In addition, students in a graduate course on primary care independently assigned each item to a domain as well as to its appropriate subdomain.

In addition to the 4 core primary care domains, 3 other related domains (family centeredness, community orientation, and cultural competence) were included; these domains were considered derivative in that their achievement would be related to the achievement of the major domains.¹ However, they were separately specified as ancillary domains because of widespread appreciation of their likely importance.

Thus, the PCAT-AE consists of 7 domains represented by 9 scales. Each of the 4 core domains of primary care is represented by 2 components, 1 representing a characteristic of the facility of providers service organization and 1 representing a behavior of the provider or consumer.¹ One of these 8 potential components (longitudinality strength of affiliation) is represented by an index rather than a scale and is scored from the responses to the 3 questions noted under the heading Identification of the Primary Care Source. One subdomain, the facility characteristics related to the achievement of coordination, is obtainable only from the facility or provider, since consumers would not be expected to know the nature of information systems that facilitate coordination of care. Thus, the PCATinstrument has 6 scales representing the 4 primary care domains: first contactaccessibility, first contactutilization (first contact domain), longitudinalityinterpersonal relationships or ongoing care (longitudinality domain), coordination of services (coordination domain), comprehensiveness services available, comprehensivenessservices received (comprehensiveness domain) and the 3 ancillary domains of family centeredness, community orientation, and cultural competence.

For first contactaccessibility 12 questions were developed to measure access to the source of care. For first contactutilization 3 questions addressed the extent to which the source of care is first used for various types of problems. Twenty questions addressed the nature and strength of the person-focused relationship with the source of care over time (longitudinality). Eight questions were used to address the coordination of services between a primary care provider and specialty care. The comprehensivenessservices available domain included 24 items of important primary care services. An additional 13 questions were used to measure comprehensivenessservices received. Two items were used to measure family-centeredness, 5 community orientation, and 3 cultural competence. Copies of both the original questionnaire and the revised condensed version are available on request.

For consistency in response and scoring, all items representing the primary care domains were represented by a 4-point Likert-type scale (1=definitely not; 2=probably not; 3=probably; and 4=definitely). The sum score for each domain was derived by adding (after reverse-coding where appropriate) the values for all the items under each domain. An additional Dont Know/Cannot Remember option was also provided for each item. At least 3 methods could be used to code this category. The missing value method treats this item as missing for those who answer Dont Know/Cant Remember. The median value method assigns a value of 2.5 for those who answer Dont Know/Cant Remember. The imputation method imputes the response based on the mean of the results from other items within the domain when at least 50% of the items have been answered. Since the internal consistency reliability (a) is the highest based on the imputation method, this method is adopted in coding the Dont Know/Cant Remember category. However, the other 2 methods also produced high internal consistency reliability (results available on request).

Analysis

The purpose of the validation was to assess the congruence between the theoretically derived measures and the empiric results in terms of the underlying structure of the principal primary care domains. Although conceptual framework was relied on in the construction of primary care measures, empiric validation was used to reduce the number of items so that the questionnaire became more concise.

The validation of PCAT-AE with the South Carolina sample involved several steps. First, principal component factor analysis was used to explore the structure of the PCAT-AE items and examine its construct validity by determining if the items fell into the hypothesized scales (factors; definitions of measurement-related concepts used in this paper can be found in the Appendix). Factor analysis was also used for item selection and placement into scales based on the pattern of the factor loadings.³⁵ Four criteria were used in deleting items and the determination of the final factors.^36-37 Afactor loading greater than 0.35 was considered meaningful and used as a criterion for retaining items. In addition, each retained factor should have at least 3 items with loadings greater than 0.35. All retained items should share the same conceptual meaning or construct. Also, all retained items should not have secondary loadings greater than 0.35.

Second, internal consistency reliability of the primary care scales was assessed by Cronbachs coefficient alpha (a)³⁸ and item-total correlation for items in each domain. Cronbachs coefficient alpha is based on the covariance among individual items in a scale and the number of items. It ranges from 0, indicating total lack of consistency, to 1, indicating complete internal consistency reliability. The item-total correlation is the correlation between an individual item and the sum of the remaining items that constitute the scale. If an item-total correlation is small, the item is not considered to be measuring the same construct that is measured by the other items in the scale. All the retained items exceeded the minimum acceptable item-total correlation of 0.30.³⁸

Third, the Likert scaling assumptions were tested for the final items related to the primary care scales. Likerts method of summated rating scales is based on the assumption that item responses in each scale can be summed without standardization or weighting.³⁹ The underlying assumptions that must be met include: (1) item-convergent validity (tested by item-scale correlations); (2) item-discriminant validity (tested using the scaling success rate, ie, correlation of each item with other items within the same scale is greater than with items from different scales); (3) equal item variance (tested by examining item means and standard deviations and the equivalence of the intraclass correlation and Scotts homogeneity ratio for each scale); (4) equal item-scale correlation (tested by examining the range of item-scale correlations); and (5) score reliability (tested by Cronbachs coefficient a.

Fourth, descriptive statistics were performed for the revised primary care scales, including mean, standard deviation, range, percentile, skewness, kurtosis, and interscale correlation. Since respondents who never saw a specialist did not answer the coordination questions, analyses were performed both with and without those questions, including the coordination domain.

Results

Subjects

For the HMO group, a total of 350 individuals responded after 3 mailings. Excluding the nonresponses due to wrong addresses and changed plans (n=340), the effective response rate was 53 percent (350/660). The respondents and nonrespondents were not significantly different in age, sex, race, and zip codes of mailing addresses. For the CHC group, a total of 1000 individuals were systematically selected and approached. Among them, 265 refused to be interviewed, 195 were not able to complete the interview prior to their appointment, and 540 completed the interview. Taking only refusal into account, the response rate was 67% (540/540+265). Men were more likely to refuse the interview than women. There were no significant differences in age and race between respondents and nonrespondents. All interviews were conducted by graduate public health students trained in interactive sessions and were completed in 1999.

The sample included 823 adults with an identified usual source of care. Among them, most (69% of HMO and 60% of CHC respondents) indicated a strong affiliation with their usual source of care (ie, all 3 doctors/places were the same). Very few (0.6% of HMO and 1.2% of CHC respondents) indicated the weakest affiliation with their usual source of care (ie, all 3 responses were different). Just over half of respondents (56%) were non-white (primarily black). Over half (55%) had an annual household income under $25,000. Most respondents (76%) had health insurance coverage all year and had been seeing their regular source of care for more than 1 year (82%). Sixty-three percent had seen their regular source of care for more than 2 years. The majority chose their own usual source of care (78%) and did not have trouble paying for their health care (74%). More than half of the respondents made at least 1 visit to a specialist (56%). This relatively high rate may be due to a somewhat elderly sample; more than 20% of the respondents were older than 65 years.

Table 1 compares the HMO sample with the CHC sample on sociodemographic and health care utilization measures. The HMO sample included predominantly white (81.6%) and higher income subjects (86.8% with annual household income of $25,000 or more). In contrast, the CHC sample included predominantly non-white (83.2%) and lower income subjects (85.9% with an annual household income less than $25,000). Compared with the CHC respondents, HMO subjects had been seeing their regular source of care for a longer time, were more likely to choose their own doctors and visit a specialist, and less likely to have trouble paying for their health care.

Factor Analysis and Construct Validity

In the initial exploratory factor analysis, all 92 applicable questionnaire items measuring the subdomains and domains of primary carefirst contact, longitudinality, comprehensiveness, coordination, family centeredness, community orientation, and cultural competencewere included. Based on the results of the initial factor analysis, 4 criteria were applied to reach the final solution (Table 2; initial factor analyses not shown but available upon request).

Seven common factors were extracted, corresponding to the hypothesized primary care scales: first contactaccessibility, first contactutilization, longitudinalityinterpersonal relationships, comprehensivenessservices available, comprehensivenessservices received, coordination, and community orientation (Table 2). Those extracted factors explained 88.1% of the common variance. Eigenvalues ranged from 16.17 to 1.16. All principal primary care domains were extracted as hypothesized. Only 1 of the 3 derivative features, community orientation, was separately identifiable.

Derivation and Reliability of the Primary Care Scales

Table 3 presents the results of the reliability analyses for both the original items and the final items (based on factor analysis). Item descriptive results (means and standard deviations) are also presented. Scale reliability measures include item-total correlation and alpha coefficient reliability. The distribution of the items varied significantly from a mean of 1.85 (ask about gun safety) to 3.73 (Provider answers questions in ways you understand) on the 4-point Likert-type scale. The distribution tends to skew toward more favorable answers (above 2.5). Apart from the gun safety item, only 2 items fell below a mean of 2 (1.94 for Provider knows neighborhood problems, 1.90 for Provider makes home visits). The first contactutilization and longitudiinalityinterpersonal relationships scales achieved the highest mean scores, whereas scales with lower means were community orientation, first contact-accessibility, and comprehensiveness-services received.

Eighteen of the 92 initial items were deleted on the basis of the criteria imposed for factor analyses. No items were deleted for first contact-utilization, coordination of services, comprehensiveness-services received, and community orientation scales. All items were deleted for family centeredness as were two thirds of the items for first contact-accessibility. Two items (out of 22) were deleted for longitudinality-interpersonal relationships and 3 (out of 24) for comprehensivenessservices available. Items from cultural competence were combined into first contact-accessibility. The revised scales demonstrate internal consistency reliability that was higher than or equal to the original scales, despite the reduction in number of items. Item-total correlations were also high and ranged from 0.34 (If sick, seen same day if office is open) to 0.91 (How to prevent hot water burns and How to prevent falls).

Testing the Likert Scaling Assumptions

Table 4 presents a summary of the results of the tests of Likert scaling assumptions using the revised items. All item-scale correlations well exceeded the accepted minimum (0.30) with the majority greater than 0.50 (Assumption 1). All 7 multi-item scales achieved 100% scaling success, indicating that all items in these scales correlated substantially higher with items in their hypothesized scale than with items in other scales (Assumption 2). Item means within each revised scale generally differed by less than six tenths of a point (except for first contact-accessibility) and item standard deviations within each scale by less than four tenths of a point (Assumption 3). Formal evidence of equal item variance was supported by the equivalence of the intraclass correlation and Scotts homogeneity ratio for each scale. Equal-item scale correlation (Assumption 4) was also observed through the range of item-scale correlations. As shown in column 1 (range of item-scale correlations), the range is relatively narrow (from .17 for coordination of services to .38 for comprehensiveness-services received). Finally, score reliability (Assumption 5) showed that except for first contact-utilization (only 3 items), all alpha levels exceeded .70 and were sufficiently high. Five of the 7 scales had alpha levels above .85.

Descriptive Feature of PCAT-AE

Table 5 displays estimates of central tendency and dispersion of scale score distributions for the 7 primary care scales in this South Carolina sample. Except for community orientation, all primary care scales were negatively skewed, indicating distributions with more positive ratings of primary care. The community orientation scale was positively skewed, indicating distributions with more negative ratings on the community orientation aspect of primary care. The full range of possible scores was observed for all scales except ongoing care.

The percentage of respondents scoring at the floor (the lowest score) or ceiling (the highest score) was acceptably low for all scales except first contactutilization, where 50% of the respondents scored the maximum score.

Table 6 compares the alpha coefficient and interfactor correlation for each primary care scale. The alpha coefficient of each scale substantially exceeded its correlation with all other primary care scales. None of the inter-factor correlations were excessively high, demonstrating that each primary care scale has significant unique contribution. All significant correlations were positive, indicating the complementary nature of primary care domains. Relatively high and positive interfactor correlations were observed between comprehensivenessservices received and comprehensiveness-services available (0.44), with the former and longitudinalityinterpersonal relationships (0.43), with the latter and coordination (0.38), and with comprehensivenessservices received and community orientation (0.37).

Discussion

Using patient-provided survey information collected within 2 health plans in South Carolina, we assessed the validity and reliability of the PCAT-AE. The results indicate that the hypothesized scales for primary care (first contactaccessibility, first con-tactutilization, longitudinalityinterpersonal relationships, comprehensivenessservices available, comprehensivenessservices received, and coordination) have substantial reliability and validity, consistent with the findings from the testing of the PCAT-CE.³⁰ The 2 versions of the instrument differ only in the comprehensiveness domains, as comprehensiveness implies that all common needs are met, and health needs in childhood are different from those in adults. In contrast, challenges to accessibility, to the nature of interpersonal relationships, and to coordination and community orientation are similar for both children and adults and thus can be assessed by the same items. Only 1 ancillary feature of primary care, community orientation, was retained as a separate dimension after factor analyses. The extracted factors explained 88.1 percent of the total variance in the item scores.

All of the 5 assumptions, including item-conver-gent validity, item-discriminant validity, equal item variance, equal item-scale correlation, and score reliability, were met. These results suggest that these items may be used to represent the primary care scales, and the scoring of these items may be summed without standardization or weighting, as with Likerts method of summated rating scales.³⁹

The resulting instrument has 74 items. Although the retained items adequately addressed first contactutilization, longitudinalityinterpersonal relationships, comprehensivenessservices available, comprehensivenessservices received, and coordination, and are consistent with the framework, those representing first contactaccessibility fell short. Only 4 of the 12 items measuring accessibility were retained. When more detail on accessibility is required, items that were deleted because they had lower item-total correlation may be added back in. Users should also review the comprehensiveness items to ascertain their relevance in the setting in which they are to be used. Items may be deleted if they are inappropriate in the context in which they are used; for example, in health systems that do not offer on-site testing for human immunodeficiency virus (HIV), because HIV is uncommon. Since continuity of care is an important component of primary care quality, a minimum number of visits or minimum duration with a regular source of care should be part of the assessment tool.

Separate factor analyses were performed with the 2 health plans. The results were largely comparable in terms of the factors that emerged as significant, indicating the generalizability of the tool to both vulnerable and middle-income populations. The only major differences are that the CHC subpopulation analysis yielded an additional significant factor, cultural competence, which the HMO subpopulation and the total population analyses failed to identify. In contrast, the HMO subpopulation analysis yielded an additional significant factor, family centeredness, which the CHC subpopulation and the total population analyses failed to identify. Thus, when using PCATon vulnerable populations (especially racial and ethnic minorities), questions measuring cultural competence might be retained. Family centeredness seemed to emerge as a distinct concept, primarily in the middle-income population.

There are a number of uses for a valid and reliable instrument such as the PCAT-AE. First, understanding primary care as a multidimensional concept is consistent with the IOMs conceptualization of primary care and more precisely captures the quality of primary care than unidimensional proxies, such as a clinicians medical specialty. With the 6 scales representing 4 core domains, the index representing strength of affiliation with a primary care provider, a scale for community orientation and the optional scales for family centeredness and cultural competence, all the important features of primary care are addressed. Second, PCAT-AE can be used as a quality measurement tool that assesses the adequacy of primary care experience rendered under different health care systems or settings, and for patients with different sociodemographic attributes. Third, PCAT-AE can also serve as a quality control tool that compares the quality of primary care given by providers of different types. The instrument can be used with other outcomes to assess the effect of policy interventions and systems changes on the delivery of critical aspects of primary care.

Limitations

Interpretation of our results should take into account some limitations. First, because our study was restricted to 1 locale, the generalizability of the PCAT-AE to other sites and states is not assured. Additional testing and validation is necessary to corroborate the current results. Second, the 74-item questionnaire remains lengthy and could have contributed to relatively high nonresponse and incompletion rates. Future validation work will concentrate on further reduction of the items to the very essential in order to reduce response burden. Regarding the ceiling effect of first contactutilization, future tests will be conducted in other settings with less of a managed care focus, as there well may be quite different distributions of responses in other settings. Third, outcomes of primary care are not the focus of the assessment tool. However, numerous studies have linked primary care to better health outcomes. Subsequent research may help explain which attributes are most conducive to better outcomes so that limited resources can be used to focus on them or a combination of them. Fourth, the measurement of primary care achievement is entirely based on respondents self-report. While self-report may be the best way to ascertain peoples experiences, it is subject to recall and response bias. Moreover, some aspects of technical quality cannot be assessed by patientsor consumers reports.

Despite these limitations, PCAT-AE is a valuable tool for capturing the principal domains of primary care. The next phase of our work seeks to assess the predictive validity of PCAT-AE, by examining the extent to which the principal attributes of primary care can be linked to the achievement of favorable health outcomes, their ability to manage their illnesses, and their satisfaction with the care received. Such work would advance our understanding of the relationship between how primary care is delivered and the health outcomes that result.

Related technical terms

Primary Care Attributes

First contactcare implies accessibility to and use of services for each new problem or new episode of a problem for which people seek health care.

Longitudinalitypresupposes the existence of a regular source of care and its use over time.

Comprehensivenessimplies that primary care facilities must be able to arrange for all types of health care services, including referrals to secondary services for consultation, tertiary services for specific conditions, and essential supporting services, such as home care and other community services.

Coordinationof care requires some form of continuity, either by practitioners, medical records, or both, as well as recognition of problems that are addressed elsewhere and the integration of their care into the total care of patients.

Family centerednessrefers to recognition of family factors related to the genesis and management of illness.

Community orientationrefers to the providers knowledge of community needs and involvement in the community.

Cultural competencerefers to the providers adaptation to facilitate relationships with populations having special cultural characteristics.

Measurement Concepts

Measurement validityrefers to the extent that important dimensions of a concept and their categories have been taken into account and appropriately operationalized.

Measurement reliabilityrefers to the extent that consistent results are obtained when a particular measure is applied to similar elements.

Construct validityis present when the measure captures the major dimensions of the concept under study.

Content validityrefers to the representativeness of the response categories used to represent each of the dimensions of a concept.

Concurrent validitymay be tested by comparing results of one measurement with those of a similar measurement administered to the same population and at approximately the same time. If both measurements yield similar results, then concurrent validity can be established.

Predictive validity exists when the results obtained from the measurement succeed in predicting the expected later-occurring event or circumstance.

Test-retest reliabilityinvolves administering the same measurement to the same individuals at 2 different times. If the correlation between the same measures is high, then the measurement is believed to be reliable.

Split-half reliabilityinvolves preparing 2 sets of measurement of the same concept, applying them to research subjects at one setting, and comparing the correlation between the 2 sets of measurement. To the extent the correlation is high, then the measurement is reliable.

Interrater reliabilityinvolves using different people to conduct the same procedure, whether it be interview, observation, coding, rating, and the like, and comparing the results of their work. To the extent that the results are highly similar, interrater reliability is established.

Item-convergent validityrefers to the substantial correlation between each item and its hypothesized scale.

Item-discriminant validityrefers to items within a scale that correlate more substantially with their hypothesized scale than with any other scale.

Equal item variancerefers to items within a scale that have approximately equal means and variances.

Equal item-scale correlationrefers to items in a scale that contribute approximately the same proportion of information about the underlying concept.

Score reliabilityrefers to scores of scales that are reproducible and reliable.

Skewnessrefers to distribution of observations that is not symmetric, ie, when more observations are found at one end of the distribution than the other.

Kurtosisrefers to the extent observations cluster around a central point more than in normal distribution.

Agrowing body of literature at both individual and ecologic levels has demonstrated the association of primary care and health outcomes.^1-11 Franks and Fiscella,¹² using nationally representative survey data, showed that adult respondents who reported a primary care physician rather than a specialist as their regular source of care had lower subsequent mortality and lower annual health care costs after controlling for differences in demographic characteristics, health insurance status, health perceptions, reported diagnoses, and smoking status. Both Shi^4,6 and Farmer and collegues¹³ found better health outcomes in states with higher primary care physician-population ratios after controlling for sociodemographic measures (% elderly, % urban, % minority, education, income, unemployment, pollution) and lifestyle factors (seatbelt usage, obesity, and smoking). Recent studies further showed that primary care may mitigate the adverse effects of income inequality on health.^14-16 Taken individually, each of the main features of primary care (person-focused care over time, accessible care, comprehensive in the sense of meeting all common health needs, and coordination when people have to receive services elsewhere) are known to improve both the effectiveness as well as the efficiency of care.^1,7,17-24

The mounting evidence associating primary care with improved health outcome has led to a rapid increase in interest in assessing primary care achievement by consumers and patients.^{18,19,21,25-28} Despite its importance, there currently is no way to assess the extent to which people receive adequate primary care; receiving care from a physician or physician designated as a primary care physician is at best only a proxy for actual adequacy of provision of primary care services. As a result, there are efforts to develop instruments that directly assess the adequacy of primary care.^20,29,30

The Primary Care Assessment Tool (PCAT) instruments developed by The Johns Hopkins Primary Care Policy Center for Underserved Populations were designed to measure the extent and quality of primary care services at a provider setting designated by consumers as their main source of general care and consistent with a focus on attributes of primary care that have been demonstrated to produce better outcomes of care at lower costs.²² The PCATfamily of instruments includes the Child Consumer/Client Survey, the Adult Consumer/Client Survey, and the Facility/Provider Survey. All surveys are based on self-report by patients or providers. The Consumer/Client Survey (both adult and child editions) is designed to collect information from consumers or family caretakers regarding their experience using health care resources. It may be used to survey target populations as defined by geography (community surveys), health plans, sites of care, payment mechanisms, or specific health care needs. The survey, which takes approximately 40 minutes to complete, can be administered through either telephone or face-to-face interviews, or by mail. Ahigh school reading level is required to self-administer the questionnaire. The Facility/Provider Survey is designed to collect information about specific operational characteristics and practices related to providing primary care from the viewpoint of practitioners, clinics, group practices, and institutions. This survey can also be implemented either by mail or by face-to-face or telephone interviews. It is parallel in content to the consumer/client survey. All 3 instruments are available for general use on request.

We report on the validation of the Consumer/Client Primary Care Assessment Tool Adult Edition (PCAT-AE). Its companion instrument for children (PCAT-CE) was previously validated.³⁰ Specifically, we assessed the congruence between the theoretically derived measures and the empiric results in terms of the underlying structure of the principal primary care domains within a diverse sample of populations including health maintenance organization (HMO) members and community health center (CHC) users. The validation process also served to reduce the number of items needed to assess the adequacy of primary care.

Methods

Subjects

The study participants were members of 2 health plans in 2 counties of South Carolina. Both counties are part of Columbia, the states capital and third largest city. One of the health plans (referred to as HMO) is licensed as an independent practice association (IPA) HMO model, in which primary care physicians act as gatekeepers and health care managers. Referral to specialists must be made through primary care physicians, and specialists must be affiliated with the HMO. The primary market has been large group employers, including employees of the state agencies and national and regional companies. Members of this plan are primarily from middle-income households. The other health plan (referred to as CHC) is a coalition of 12 Columbia-based health and social services provider organizations, including the county hospital, health department, department of social services, community health centers, and other social service agencies that provide services to lower income persons, such as Medicaid recipients and low-income households. These 2 plans were selected because they represent typical South Carolina managed care organizations and health plans for low-income individuals, respectively. Samples drawn from these 2 plans allowed us to test the reliability of PCATwith a diverse sample of populations, including both middle-income and low-income individuals using regular physician offices and community health centers, respectively.

Estimation of the sample size for this study involved several steps. First, an estimate of the likely proportions or means and standard deviations for each primary care measure was derived from a previous study.²⁵ When data were not available, a conservative estimate (eg, a larger standard deviation or proportion closer to 50/50) was made. Second, the estimates of the proportions, means, and standard deviations for the dependent variables were entered into the standard sample size formula for a two-group, cross-sectional sample. Using a 95% confidence interval, the largest sample size required was 300 per group. The CHC group was oversampled because of additional planned within-group analyses (not the focus of this paper). Finally, the desired sample size was adjusted for anticipated survey nonresponse (anticipated to be higher for a mail survey than a face-to-face interview).

For the HMO group, a mail survey was used since it was deemed most efficient. In 2 previous longitudinal studies of the same HMO, we used mail survey and telephone interviews alternately with a cohort of HMO members and obtained comparable results.^31,32 For this study, we sent a letter with a PCAT-AE questionnaire to 1000 randomly selected adult members to invite them to participate in the project. Because of known frequent changes in addresses, we recruited the non-HMO plan individuals and conducted in-person interviews at all the community health center sites where members came to the clinics for non-urgent visits. Patients were systematically approached while waiting for their scheduled appointment (ie, every nth patient based on expected visits for a particular site) and recruited for the study during a period of 4 weeks for each site.

Measures

Identification of Primary Care Source.Three questions were developed to identify an individuals usual source of care and the strength of that affiliation: (1) Is there a doctor or place that you usually go if you are sick or need advice about your health? (usual source), (2) Is there a doctor or place that knows you best as a person? (knows best), and (3) Is there a doctor or place that is most responsible for your health care? (most responsible). Aperson was considered to have a usual source of care if he or she answered positively to any 1 of the 3 questions (95% for the HMO plan and 90% for the low-income plan). Anegative answer to all 3 questions rendered the individual as not having a usual source of care.

An algorithm based on response to these 3 questions identified the strength of affiliation with the primary care source. If all 3 physicians/places were the same, this was considered evidence of a strong affiliation. If the response to the usual source question was the same as for either of the other 2 questions then that site was used although the affiliation was considered less strong. If the response for a usual source question was different from the other 2 responses but the other 2 responses were the same, then the site where both were the same was used (weak affiliation). If all 3 responses were different (weakest affiliation), then the site identified for usual source was used. All subsequent questions asked about this specific person or place. For those with no identifiable source of primary care, subsequent questions were asked about the last place that was visited.

Domains of Primary Care.The PCAT-AE was modeled on the previously validated PCAT-CE and is consistent with the 1978 Institute of Medicine (IOM) definition of primary care as accessibility, comprehensiveness, coordination, continuity, and accountability³³ and with the 1996 IOM report definition of primary care as the provision of integrated, accessible health care services by clinicians who are accountable for addressing a large majority of personal health care needs, developing a sustained partnership with patients, and practicing in the context of family and the community.³⁴ When combined into scales, the PCATsurvey items dealing with primary care quality were designed to measure each of the core domains of primary care; that is, first contact, longitudinality, comprehensiveness, and coordination (definitions of the primary care domains are provided in the Appendix).

Nine experts were asked to rate the appropriateness and representativeness of the primary care domain items. These experts consisted of 3 policymakers in federal agencies, 2 directors of community pediatrics at major medical centers, a health research director at a major HMO, 2 family medicine professors, and a general internal medicine physician. Acard sorting technique was used to determine the degree of congruence between each item and the domain it was designed to measure. Each survey question with its response categories and descriptions of each of the primary care domains was printed on separate index cards and mailed to the experts who assigned each question to one of the defined domains and suggested revisions and/or addition of other items. The percent agreement among the experts was used to determine the degree of congruence on the placement of each item in a particular domain. In addition, students in a graduate course on primary care independently assigned each item to a domain as well as to its appropriate subdomain.

In addition to the 4 core primary care domains, 3 other related domains (family centeredness, community orientation, and cultural competence) were included; these domains were considered derivative in that their achievement would be related to the achievement of the major domains.¹ However, they were separately specified as ancillary domains because of widespread appreciation of their likely importance.

Thus, the PCAT-AE consists of 7 domains represented by 9 scales. Each of the 4 core domains of primary care is represented by 2 components, 1 representing a characteristic of the facility of providers service organization and 1 representing a behavior of the provider or consumer.¹ One of these 8 potential components (longitudinality strength of affiliation) is represented by an index rather than a scale and is scored from the responses to the 3 questions noted under the heading Identification of the Primary Care Source. One subdomain, the facility characteristics related to the achievement of coordination, is obtainable only from the facility or provider, since consumers would not be expected to know the nature of information systems that facilitate coordination of care. Thus, the PCATinstrument has 6 scales representing the 4 primary care domains: first contactaccessibility, first contactutilization (first contact domain), longitudinalityinterpersonal relationships or ongoing care (longitudinality domain), coordination of services (coordination domain), comprehensiveness services available, comprehensivenessservices received (comprehensiveness domain) and the 3 ancillary domains of family centeredness, community orientation, and cultural competence.

For first contactaccessibility 12 questions were developed to measure access to the source of care. For first contactutilization 3 questions addressed the extent to which the source of care is first used for various types of problems. Twenty questions addressed the nature and strength of the person-focused relationship with the source of care over time (longitudinality). Eight questions were used to address the coordination of services between a primary care provider and specialty care. The comprehensivenessservices available domain included 24 items of important primary care services. An additional 13 questions were used to measure comprehensivenessservices received. Two items were used to measure family-centeredness, 5 community orientation, and 3 cultural competence. Copies of both the original questionnaire and the revised condensed version are available on request.

For consistency in response and scoring, all items representing the primary care domains were represented by a 4-point Likert-type scale (1=definitely not; 2=probably not; 3=probably; and 4=definitely). The sum score for each domain was derived by adding (after reverse-coding where appropriate) the values for all the items under each domain. An additional Dont Know/Cannot Remember option was also provided for each item. At least 3 methods could be used to code this category. The missing value method treats this item as missing for those who answer Dont Know/Cant Remember. The median value method assigns a value of 2.5 for those who answer Dont Know/Cant Remember. The imputation method imputes the response based on the mean of the results from other items within the domain when at least 50% of the items have been answered. Since the internal consistency reliability (a) is the highest based on the imputation method, this method is adopted in coding the Dont Know/Cant Remember category. However, the other 2 methods also produced high internal consistency reliability (results available on request).

Analysis

The purpose of the validation was to assess the congruence between the theoretically derived measures and the empiric results in terms of the underlying structure of the principal primary care domains. Although conceptual framework was relied on in the construction of primary care measures, empiric validation was used to reduce the number of items so that the questionnaire became more concise.

The validation of PCAT-AE with the South Carolina sample involved several steps. First, principal component factor analysis was used to explore the structure of the PCAT-AE items and examine its construct validity by determining if the items fell into the hypothesized scales (factors; definitions of measurement-related concepts used in this paper can be found in the Appendix). Factor analysis was also used for item selection and placement into scales based on the pattern of the factor loadings.³⁵ Four criteria were used in deleting items and the determination of the final factors.^36-37 Afactor loading greater than 0.35 was considered meaningful and used as a criterion for retaining items. In addition, each retained factor should have at least 3 items with loadings greater than 0.35. All retained items should share the same conceptual meaning or construct. Also, all retained items should not have secondary loadings greater than 0.35.

Second, internal consistency reliability of the primary care scales was assessed by Cronbachs coefficient alpha (a)³⁸ and item-total correlation for items in each domain. Cronbachs coefficient alpha is based on the covariance among individual items in a scale and the number of items. It ranges from 0, indicating total lack of consistency, to 1, indicating complete internal consistency reliability. The item-total correlation is the correlation between an individual item and the sum of the remaining items that constitute the scale. If an item-total correlation is small, the item is not considered to be measuring the same construct that is measured by the other items in the scale. All the retained items exceeded the minimum acceptable item-total correlation of 0.30.³⁸

Third, the Likert scaling assumptions were tested for the final items related to the primary care scales. Likerts method of summated rating scales is based on the assumption that item responses in each scale can be summed without standardization or weighting.³⁹ The underlying assumptions that must be met include: (1) item-convergent validity (tested by item-scale correlations); (2) item-discriminant validity (tested using the scaling success rate, ie, correlation of each item with other items within the same scale is greater than with items from different scales); (3) equal item variance (tested by examining item means and standard deviations and the equivalence of the intraclass correlation and Scotts homogeneity ratio for each scale); (4) equal item-scale correlation (tested by examining the range of item-scale correlations); and (5) score reliability (tested by Cronbachs coefficient a.

Fourth, descriptive statistics were performed for the revised primary care scales, including mean, standard deviation, range, percentile, skewness, kurtosis, and interscale correlation. Since respondents who never saw a specialist did not answer the coordination questions, analyses were performed both with and without those questions, including the coordination domain.

Results

Subjects

For the HMO group, a total of 350 individuals responded after 3 mailings. Excluding the nonresponses due to wrong addresses and changed plans (n=340), the effective response rate was 53 percent (350/660). The respondents and nonrespondents were not significantly different in age, sex, race, and zip codes of mailing addresses. For the CHC group, a total of 1000 individuals were systematically selected and approached. Among them, 265 refused to be interviewed, 195 were not able to complete the interview prior to their appointment, and 540 completed the interview. Taking only refusal into account, the response rate was 67% (540/540+265). Men were more likely to refuse the interview than women. There were no significant differences in age and race between respondents and nonrespondents. All interviews were conducted by graduate public health students trained in interactive sessions and were completed in 1999.

The sample included 823 adults with an identified usual source of care. Among them, most (69% of HMO and 60% of CHC respondents) indicated a strong affiliation with their usual source of care (ie, all 3 doctors/places were the same). Very few (0.6% of HMO and 1.2% of CHC respondents) indicated the weakest affiliation with their usual source of care (ie, all 3 responses were different). Just over half of respondents (56%) were non-white (primarily black). Over half (55%) had an annual household income under $25,000. Most respondents (76%) had health insurance coverage all year and had been seeing their regular source of care for more than 1 year (82%). Sixty-three percent had seen their regular source of care for more than 2 years. The majority chose their own usual source of care (78%) and did not have trouble paying for their health care (74%). More than half of the respondents made at least 1 visit to a specialist (56%). This relatively high rate may be due to a somewhat elderly sample; more than 20% of the respondents were older than 65 years.

Table 1 compares the HMO sample with the CHC sample on sociodemographic and health care utilization measures. The HMO sample included predominantly white (81.6%) and higher income subjects (86.8% with annual household income of $25,000 or more). In contrast, the CHC sample included predominantly non-white (83.2%) and lower income subjects (85.9% with an annual household income less than $25,000). Compared with the CHC respondents, HMO subjects had been seeing their regular source of care for a longer time, were more likely to choose their own doctors and visit a specialist, and less likely to have trouble paying for their health care.

Factor Analysis and Construct Validity

In the initial exploratory factor analysis, all 92 applicable questionnaire items measuring the subdomains and domains of primary carefirst contact, longitudinality, comprehensiveness, coordination, family centeredness, community orientation, and cultural competencewere included. Based on the results of the initial factor analysis, 4 criteria were applied to reach the final solution (Table 2; initial factor analyses not shown but available upon request).

Seven common factors were extracted, corresponding to the hypothesized primary care scales: first contactaccessibility, first contactutilization, longitudinalityinterpersonal relationships, comprehensivenessservices available, comprehensivenessservices received, coordination, and community orientation (Table 2). Those extracted factors explained 88.1% of the common variance. Eigenvalues ranged from 16.17 to 1.16. All principal primary care domains were extracted as hypothesized. Only 1 of the 3 derivative features, community orientation, was separately identifiable.

Derivation and Reliability of the Primary Care Scales

Table 3 presents the results of the reliability analyses for both the original items and the final items (based on factor analysis). Item descriptive results (means and standard deviations) are also presented. Scale reliability measures include item-total correlation and alpha coefficient reliability. The distribution of the items varied significantly from a mean of 1.85 (ask about gun safety) to 3.73 (Provider answers questions in ways you understand) on the 4-point Likert-type scale. The distribution tends to skew toward more favorable answers (above 2.5). Apart from the gun safety item, only 2 items fell below a mean of 2 (1.94 for Provider knows neighborhood problems, 1.90 for Provider makes home visits). The first contactutilization and longitudiinalityinterpersonal relationships scales achieved the highest mean scores, whereas scales with lower means were community orientation, first contact-accessibility, and comprehensiveness-services received.

Eighteen of the 92 initial items were deleted on the basis of the criteria imposed for factor analyses. No items were deleted for first contact-utilization, coordination of services, comprehensiveness-services received, and community orientation scales. All items were deleted for family centeredness as were two thirds of the items for first contact-accessibility. Two items (out of 22) were deleted for longitudinality-interpersonal relationships and 3 (out of 24) for comprehensivenessservices available. Items from cultural competence were combined into first contact-accessibility. The revised scales demonstrate internal consistency reliability that was higher than or equal to the original scales, despite the reduction in number of items. Item-total correlations were also high and ranged from 0.34 (If sick, seen same day if office is open) to 0.91 (How to prevent hot water burns and How to prevent falls).

Testing the Likert Scaling Assumptions

Table 4 presents a summary of the results of the tests of Likert scaling assumptions using the revised items. All item-scale correlations well exceeded the accepted minimum (0.30) with the majority greater than 0.50 (Assumption 1). All 7 multi-item scales achieved 100% scaling success, indicating that all items in these scales correlated substantially higher with items in their hypothesized scale than with items in other scales (Assumption 2). Item means within each revised scale generally differed by less than six tenths of a point (except for first contact-accessibility) and item standard deviations within each scale by less than four tenths of a point (Assumption 3). Formal evidence of equal item variance was supported by the equivalence of the intraclass correlation and Scotts homogeneity ratio for each scale. Equal-item scale correlation (Assumption 4) was also observed through the range of item-scale correlations. As shown in column 1 (range of item-scale correlations), the range is relatively narrow (from .17 for coordination of services to .38 for comprehensiveness-services received). Finally, score reliability (Assumption 5) showed that except for first contact-utilization (only 3 items), all alpha levels exceeded .70 and were sufficiently high. Five of the 7 scales had alpha levels above .85.

Descriptive Feature of PCAT-AE

Table 5 displays estimates of central tendency and dispersion of scale score distributions for the 7 primary care scales in this South Carolina sample. Except for community orientation, all primary care scales were negatively skewed, indicating distributions with more positive ratings of primary care. The community orientation scale was positively skewed, indicating distributions with more negative ratings on the community orientation aspect of primary care. The full range of possible scores was observed for all scales except ongoing care.

The percentage of respondents scoring at the floor (the lowest score) or ceiling (the highest score) was acceptably low for all scales except first contactutilization, where 50% of the respondents scored the maximum score.

Table 6 compares the alpha coefficient and interfactor correlation for each primary care scale. The alpha coefficient of each scale substantially exceeded its correlation with all other primary care scales. None of the inter-factor correlations were excessively high, demonstrating that each primary care scale has significant unique contribution. All significant correlations were positive, indicating the complementary nature of primary care domains. Relatively high and positive interfactor correlations were observed between comprehensivenessservices received and comprehensiveness-services available (0.44), with the former and longitudinalityinterpersonal relationships (0.43), with the latter and coordination (0.38), and with comprehensivenessservices received and community orientation (0.37).

Discussion

Using patient-provided survey information collected within 2 health plans in South Carolina, we assessed the validity and reliability of the PCAT-AE. The results indicate that the hypothesized scales for primary care (first contactaccessibility, first con-tactutilization, longitudinalityinterpersonal relationships, comprehensivenessservices available, comprehensivenessservices received, and coordination) have substantial reliability and validity, consistent with the findings from the testing of the PCAT-CE.³⁰ The 2 versions of the instrument differ only in the comprehensiveness domains, as comprehensiveness implies that all common needs are met, and health needs in childhood are different from those in adults. In contrast, challenges to accessibility, to the nature of interpersonal relationships, and to coordination and community orientation are similar for both children and adults and thus can be assessed by the same items. Only 1 ancillary feature of primary care, community orientation, was retained as a separate dimension after factor analyses. The extracted factors explained 88.1 percent of the total variance in the item scores.

All of the 5 assumptions, including item-conver-gent validity, item-discriminant validity, equal item variance, equal item-scale correlation, and score reliability, were met. These results suggest that these items may be used to represent the primary care scales, and the scoring of these items may be summed without standardization or weighting, as with Likerts method of summated rating scales.³⁹

The resulting instrument has 74 items. Although the retained items adequately addressed first contactutilization, longitudinalityinterpersonal relationships, comprehensivenessservices available, comprehensivenessservices received, and coordination, and are consistent with the framework, those representing first contactaccessibility fell short. Only 4 of the 12 items measuring accessibility were retained. When more detail on accessibility is required, items that were deleted because they had lower item-total correlation may be added back in. Users should also review the comprehensiveness items to ascertain their relevance in the setting in which they are to be used. Items may be deleted if they are inappropriate in the context in which they are used; for example, in health systems that do not offer on-site testing for human immunodeficiency virus (HIV), because HIV is uncommon. Since continuity of care is an important component of primary care quality, a minimum number of visits or minimum duration with a regular source of care should be part of the assessment tool.

Separate factor analyses were performed with the 2 health plans. The results were largely comparable in terms of the factors that emerged as significant, indicating the generalizability of the tool to both vulnerable and middle-income populations. The only major differences are that the CHC subpopulation analysis yielded an additional significant factor, cultural competence, which the HMO subpopulation and the total population analyses failed to identify. In contrast, the HMO subpopulation analysis yielded an additional significant factor, family centeredness, which the CHC subpopulation and the total population analyses failed to identify. Thus, when using PCATon vulnerable populations (especially racial and ethnic minorities), questions measuring cultural competence might be retained. Family centeredness seemed to emerge as a distinct concept, primarily in the middle-income population.

There are a number of uses for a valid and reliable instrument such as the PCAT-AE. First, understanding primary care as a multidimensional concept is consistent with the IOMs conceptualization of primary care and more precisely captures the quality of primary care than unidimensional proxies, such as a clinicians medical specialty. With the 6 scales representing 4 core domains, the index representing strength of affiliation with a primary care provider, a scale for community orientation and the optional scales for family centeredness and cultural competence, all the important features of primary care are addressed. Second, PCAT-AE can be used as a quality measurement tool that assesses the adequacy of primary care experience rendered under different health care systems or settings, and for patients with different sociodemographic attributes. Third, PCAT-AE can also serve as a quality control tool that compares the quality of primary care given by providers of different types. The instrument can be used with other outcomes to assess the effect of policy interventions and systems changes on the delivery of critical aspects of primary care.

Limitations

Interpretation of our results should take into account some limitations. First, because our study was restricted to 1 locale, the generalizability of the PCAT-AE to other sites and states is not assured. Additional testing and validation is necessary to corroborate the current results. Second, the 74-item questionnaire remains lengthy and could have contributed to relatively high nonresponse and incompletion rates. Future validation work will concentrate on further reduction of the items to the very essential in order to reduce response burden. Regarding the ceiling effect of first contactutilization, future tests will be conducted in other settings with less of a managed care focus, as there well may be quite different distributions of responses in other settings. Third, outcomes of primary care are not the focus of the assessment tool. However, numerous studies have linked primary care to better health outcomes. Subsequent research may help explain which attributes are most conducive to better outcomes so that limited resources can be used to focus on them or a combination of them. Fourth, the measurement of primary care achievement is entirely based on respondents self-report. While self-report may be the best way to ascertain peoples experiences, it is subject to recall and response bias. Moreover, some aspects of technical quality cannot be assessed by patientsor consumers reports.

Despite these limitations, PCAT-AE is a valuable tool for capturing the principal domains of primary care. The next phase of our work seeks to assess the predictive validity of PCAT-AE, by examining the extent to which the principal attributes of primary care can be linked to the achievement of favorable health outcomes, their ability to manage their illnesses, and their satisfaction with the care received. Such work would advance our understanding of the relationship between how primary care is delivered and the health outcomes that result.

Related technical terms

Primary Care Attributes

First contactcare implies accessibility to and use of services for each new problem or new episode of a problem for which people seek health care.

Longitudinalitypresupposes the existence of a regular source of care and its use over time.

Comprehensivenessimplies that primary care facilities must be able to arrange for all types of health care services, including referrals to secondary services for consultation, tertiary services for specific conditions, and essential supporting services, such as home care and other community services.

Coordinationof care requires some form of continuity, either by practitioners, medical records, or both, as well as recognition of problems that are addressed elsewhere and the integration of their care into the total care of patients.

Family centerednessrefers to recognition of family factors related to the genesis and management of illness.

Community orientationrefers to the providers knowledge of community needs and involvement in the community.

Cultural competencerefers to the providers adaptation to facilitate relationships with populations having special cultural characteristics.

Measurement Concepts

Measurement validityrefers to the extent that important dimensions of a concept and their categories have been taken into account and appropriately operationalized.

Measurement reliabilityrefers to the extent that consistent results are obtained when a particular measure is applied to similar elements.

Construct validityis present when the measure captures the major dimensions of the concept under study.

Content validityrefers to the representativeness of the response categories used to represent each of the dimensions of a concept.

Concurrent validitymay be tested by comparing results of one measurement with those of a similar measurement administered to the same population and at approximately the same time. If both measurements yield similar results, then concurrent validity can be established.

Predictive validity exists when the results obtained from the measurement succeed in predicting the expected later-occurring event or circumstance.

Test-retest reliabilityinvolves administering the same measurement to the same individuals at 2 different times. If the correlation between the same measures is high, then the measurement is believed to be reliable.

Split-half reliabilityinvolves preparing 2 sets of measurement of the same concept, applying them to research subjects at one setting, and comparing the correlation between the 2 sets of measurement. To the extent the correlation is high, then the measurement is reliable.

Interrater reliabilityinvolves using different people to conduct the same procedure, whether it be interview, observation, coding, rating, and the like, and comparing the results of their work. To the extent that the results are highly similar, interrater reliability is established.

Item-convergent validityrefers to the substantial correlation between each item and its hypothesized scale.

Item-discriminant validityrefers to items within a scale that correlate more substantially with their hypothesized scale than with any other scale.

Equal item variancerefers to items within a scale that have approximately equal means and variances.

Equal item-scale correlationrefers to items in a scale that contribute approximately the same proportion of information about the underlying concept.

Score reliabilityrefers to scores of scales that are reproducible and reliable.

Skewnessrefers to distribution of observations that is not symmetric, ie, when more observations are found at one end of the distribution than the other.

Kurtosisrefers to the extent observations cluster around a central point more than in normal distribution.

Agrowing body of literature at both individual and ecologic levels has demonstrated the association of primary care and health outcomes.^1-11 Franks and Fiscella,¹² using nationally representative survey data, showed that adult respondents who reported a primary care physician rather than a specialist as their regular source of care had lower subsequent mortality and lower annual health care costs after controlling for differences in demographic characteristics, health insurance status, health perceptions, reported diagnoses, and smoking status. Both Shi^4,6 and Farmer and collegues¹³ found better health outcomes in states with higher primary care physician-population ratios after controlling for sociodemographic measures (% elderly, % urban, % minority, education, income, unemployment, pollution) and lifestyle factors (seatbelt usage, obesity, and smoking). Recent studies further showed that primary care may mitigate the adverse effects of income inequality on health.^14-16 Taken individually, each of the main features of primary care (person-focused care over time, accessible care, comprehensive in the sense of meeting all common health needs, and coordination when people have to receive services elsewhere) are known to improve both the effectiveness as well as the efficiency of care.^1,7,17-24

The mounting evidence associating primary care with improved health outcome has led to a rapid increase in interest in assessing primary care achievement by consumers and patients.^{18,19,21,25-28} Despite its importance, there currently is no way to assess the extent to which people receive adequate primary care; receiving care from a physician or physician designated as a primary care physician is at best only a proxy for actual adequacy of provision of primary care services. As a result, there are efforts to develop instruments that directly assess the adequacy of primary care.^20,29,30

The Primary Care Assessment Tool (PCAT) instruments developed by The Johns Hopkins Primary Care Policy Center for Underserved Populations were designed to measure the extent and quality of primary care services at a provider setting designated by consumers as their main source of general care and consistent with a focus on attributes of primary care that have been demonstrated to produce better outcomes of care at lower costs.²² The PCATfamily of instruments includes the Child Consumer/Client Survey, the Adult Consumer/Client Survey, and the Facility/Provider Survey. All surveys are based on self-report by patients or providers. The Consumer/Client Survey (both adult and child editions) is designed to collect information from consumers or family caretakers regarding their experience using health care resources. It may be used to survey target populations as defined by geography (community surveys), health plans, sites of care, payment mechanisms, or specific health care needs. The survey, which takes approximately 40 minutes to complete, can be administered through either telephone or face-to-face interviews, or by mail. Ahigh school reading level is required to self-administer the questionnaire. The Facility/Provider Survey is designed to collect information about specific operational characteristics and practices related to providing primary care from the viewpoint of practitioners, clinics, group practices, and institutions. This survey can also be implemented either by mail or by face-to-face or telephone interviews. It is parallel in content to the consumer/client survey. All 3 instruments are available for general use on request.

We report on the validation of the Consumer/Client Primary Care Assessment Tool Adult Edition (PCAT-AE). Its companion instrument for children (PCAT-CE) was previously validated.³⁰ Specifically, we assessed the congruence between the theoretically derived measures and the empiric results in terms of the underlying structure of the principal primary care domains within a diverse sample of populations including health maintenance organization (HMO) members and community health center (CHC) users. The validation process also served to reduce the number of items needed to assess the adequacy of primary care.

Methods

Subjects

The study participants were members of 2 health plans in 2 counties of South Carolina. Both counties are part of Columbia, the states capital and third largest city. One of the health plans (referred to as HMO) is licensed as an independent practice association (IPA) HMO model, in which primary care physicians act as gatekeepers and health care managers. Referral to specialists must be made through primary care physicians, and specialists must be affiliated with the HMO. The primary market has been large group employers, including employees of the state agencies and national and regional companies. Members of this plan are primarily from middle-income households. The other health plan (referred to as CHC) is a coalition of 12 Columbia-based health and social services provider organizations, including the county hospital, health department, department of social services, community health centers, and other social service agencies that provide services to lower income persons, such as Medicaid recipients and low-income households. These 2 plans were selected because they represent typical South Carolina managed care organizations and health plans for low-income individuals, respectively. Samples drawn from these 2 plans allowed us to test the reliability of PCATwith a diverse sample of populations, including both middle-income and low-income individuals using regular physician offices and community health centers, respectively.

Estimation of the sample size for this study involved several steps. First, an estimate of the likely proportions or means and standard deviations for each primary care measure was derived from a previous study.²⁵ When data were not available, a conservative estimate (eg, a larger standard deviation or proportion closer to 50/50) was made. Second, the estimates of the proportions, means, and standard deviations for the dependent variables were entered into the standard sample size formula for a two-group, cross-sectional sample. Using a 95% confidence interval, the largest sample size required was 300 per group. The CHC group was oversampled because of additional planned within-group analyses (not the focus of this paper). Finally, the desired sample size was adjusted for anticipated survey nonresponse (anticipated to be higher for a mail survey than a face-to-face interview).

For the HMO group, a mail survey was used since it was deemed most efficient. In 2 previous longitudinal studies of the same HMO, we used mail survey and telephone interviews alternately with a cohort of HMO members and obtained comparable results.^31,32 For this study, we sent a letter with a PCAT-AE questionnaire to 1000 randomly selected adult members to invite them to participate in the project. Because of known frequent changes in addresses, we recruited the non-HMO plan individuals and conducted in-person interviews at all the community health center sites where members came to the clinics for non-urgent visits. Patients were systematically approached while waiting for their scheduled appointment (ie, every nth patient based on expected visits for a particular site) and recruited for the study during a period of 4 weeks for each site.

Measures

Identification of Primary Care Source.Three questions were developed to identify an individuals usual source of care and the strength of that affiliation: (1) Is there a doctor or place that you usually go if you are sick or need advice about your health? (usual source), (2) Is there a doctor or place that knows you best as a person? (knows best), and (3) Is there a doctor or place that is most responsible for your health care? (most responsible). Aperson was considered to have a usual source of care if he or she answered positively to any 1 of the 3 questions (95% for the HMO plan and 90% for the low-income plan). Anegative answer to all 3 questions rendered the individual as not having a usual source of care.

An algorithm based on response to these 3 questions identified the strength of affiliation with the primary care source. If all 3 physicians/places were the same, this was considered evidence of a strong affiliation. If the response to the usual source question was the same as for either of the other 2 questions then that site was used although the affiliation was considered less strong. If the response for a usual source question was different from the other 2 responses but the other 2 responses were the same, then the site where both were the same was used (weak affiliation). If all 3 responses were different (weakest affiliation), then the site identified for usual source was used. All subsequent questions asked about this specific person or place. For those with no identifiable source of primary care, subsequent questions were asked about the last place that was visited.

Domains of Primary Care.The PCAT-AE was modeled on the previously validated PCAT-CE and is consistent with the 1978 Institute of Medicine (IOM) definition of primary care as accessibility, comprehensiveness, coordination, continuity, and accountability³³ and with the 1996 IOM report definition of primary care as the provision of integrated, accessible health care services by clinicians who are accountable for addressing a large majority of personal health care needs, developing a sustained partnership with patients, and practicing in the context of family and the community.³⁴ When combined into scales, the PCATsurvey items dealing with primary care quality were designed to measure each of the core domains of primary care; that is, first contact, longitudinality, comprehensiveness, and coordination (definitions of the primary care domains are provided in the Appendix).

Nine experts were asked to rate the appropriateness and representativeness of the primary care domain items. These experts consisted of 3 policymakers in federal agencies, 2 directors of community pediatrics at major medical centers, a health research director at a major HMO, 2 family medicine professors, and a general internal medicine physician. Acard sorting technique was used to determine the degree of congruence between each item and the domain it was designed to measure. Each survey question with its response categories and descriptions of each of the primary care domains was printed on separate index cards and mailed to the experts who assigned each question to one of the defined domains and suggested revisions and/or addition of other items. The percent agreement among the experts was used to determine the degree of congruence on the placement of each item in a particular domain. In addition, students in a graduate course on primary care independently assigned each item to a domain as well as to its appropriate subdomain.

In addition to the 4 core primary care domains, 3 other related domains (family centeredness, community orientation, and cultural competence) were included; these domains were considered derivative in that their achievement would be related to the achievement of the major domains.¹ However, they were separately specified as ancillary domains because of widespread appreciation of their likely importance.

Thus, the PCAT-AE consists of 7 domains represented by 9 scales. Each of the 4 core domains of primary care is represented by 2 components, 1 representing a characteristic of the facility of providers service organization and 1 representing a behavior of the provider or consumer.¹ One of these 8 potential components (longitudinality strength of affiliation) is represented by an index rather than a scale and is scored from the responses to the 3 questions noted under the heading Identification of the Primary Care Source. One subdomain, the facility characteristics related to the achievement of coordination, is obtainable only from the facility or provider, since consumers would not be expected to know the nature of information systems that facilitate coordination of care. Thus, the PCATinstrument has 6 scales representing the 4 primary care domains: first contactaccessibility, first contactutilization (first contact domain), longitudinalityinterpersonal relationships or ongoing care (longitudinality domain), coordination of services (coordination domain), comprehensiveness services available, comprehensivenessservices received (comprehensiveness domain) and the 3 ancillary domains of family centeredness, community orientation, and cultural competence.

For first contactaccessibility 12 questions were developed to measure access to the source of care. For first contactutilization 3 questions addressed the extent to which the source of care is first used for various types of problems. Twenty questions addressed the nature and strength of the person-focused relationship with the source of care over time (longitudinality). Eight questions were used to address the coordination of services between a primary care provider and specialty care. The comprehensivenessservices available domain included 24 items of important primary care services. An additional 13 questions were used to measure comprehensivenessservices received. Two items were used to measure family-centeredness, 5 community orientation, and 3 cultural competence. Copies of both the original questionnaire and the revised condensed version are available on request.

For consistency in response and scoring, all items representing the primary care domains were represented by a 4-point Likert-type scale (1=definitely not; 2=probably not; 3=probably; and 4=definitely). The sum score for each domain was derived by adding (after reverse-coding where appropriate) the values for all the items under each domain. An additional Dont Know/Cannot Remember option was also provided for each item. At least 3 methods could be used to code this category. The missing value method treats this item as missing for those who answer Dont Know/Cant Remember. The median value method assigns a value of 2.5 for those who answer Dont Know/Cant Remember. The imputation method imputes the response based on the mean of the results from other items within the domain when at least 50% of the items have been answered. Since the internal consistency reliability (a) is the highest based on the imputation method, this method is adopted in coding the Dont Know/Cant Remember category. However, the other 2 methods also produced high internal consistency reliability (results available on request).

Analysis

The purpose of the validation was to assess the congruence between the theoretically derived measures and the empiric results in terms of the underlying structure of the principal primary care domains. Although conceptual framework was relied on in the construction of primary care measures, empiric validation was used to reduce the number of items so that the questionnaire became more concise.

The validation of PCAT-AE with the South Carolina sample involved several steps. First, principal component factor analysis was used to explore the structure of the PCAT-AE items and examine its construct validity by determining if the items fell into the hypothesized scales (factors; definitions of measurement-related concepts used in this paper can be found in the Appendix). Factor analysis was also used for item selection and placement into scales based on the pattern of the factor loadings.³⁵ Four criteria were used in deleting items and the determination of the final factors.^36-37 Afactor loading greater than 0.35 was considered meaningful and used as a criterion for retaining items. In addition, each retained factor should have at least 3 items with loadings greater than 0.35. All retained items should share the same conceptual meaning or construct. Also, all retained items should not have secondary loadings greater than 0.35.

Second, internal consistency reliability of the primary care scales was assessed by Cronbachs coefficient alpha (a)³⁸ and item-total correlation for items in each domain. Cronbachs coefficient alpha is based on the covariance among individual items in a scale and the number of items. It ranges from 0, indicating total lack of consistency, to 1, indicating complete internal consistency reliability. The item-total correlation is the correlation between an individual item and the sum of the remaining items that constitute the scale. If an item-total correlation is small, the item is not considered to be measuring the same construct that is measured by the other items in the scale. All the retained items exceeded the minimum acceptable item-total correlation of 0.30.³⁸

Third, the Likert scaling assumptions were tested for the final items related to the primary care scales. Likerts method of summated rating scales is based on the assumption that item responses in each scale can be summed without standardization or weighting.³⁹ The underlying assumptions that must be met include: (1) item-convergent validity (tested by item-scale correlations); (2) item-discriminant validity (tested using the scaling success rate, ie, correlation of each item with other items within the same scale is greater than with items from different scales); (3) equal item variance (tested by examining item means and standard deviations and the equivalence of the intraclass correlation and Scotts homogeneity ratio for each scale); (4) equal item-scale correlation (tested by examining the range of item-scale correlations); and (5) score reliability (tested by Cronbachs coefficient a.

Fourth, descriptive statistics were performed for the revised primary care scales, including mean, standard deviation, range, percentile, skewness, kurtosis, and interscale correlation. Since respondents who never saw a specialist did not answer the coordination questions, analyses were performed both with and without those questions, including the coordination domain.

Results

Subjects

For the HMO group, a total of 350 individuals responded after 3 mailings. Excluding the nonresponses due to wrong addresses and changed plans (n=340), the effective response rate was 53 percent (350/660). The respondents and nonrespondents were not significantly different in age, sex, race, and zip codes of mailing addresses. For the CHC group, a total of 1000 individuals were systematically selected and approached. Among them, 265 refused to be interviewed, 195 were not able to complete the interview prior to their appointment, and 540 completed the interview. Taking only refusal into account, the response rate was 67% (540/540+265). Men were more likely to refuse the interview than women. There were no significant differences in age and race between respondents and nonrespondents. All interviews were conducted by graduate public health students trained in interactive sessions and were completed in 1999.

The sample included 823 adults with an identified usual source of care. Among them, most (69% of HMO and 60% of CHC respondents) indicated a strong affiliation with their usual source of care (ie, all 3 doctors/places were the same). Very few (0.6% of HMO and 1.2% of CHC respondents) indicated the weakest affiliation with their usual source of care (ie, all 3 responses were different). Just over half of respondents (56%) were non-white (primarily black). Over half (55%) had an annual household income under $25,000. Most respondents (76%) had health insurance coverage all year and had been seeing their regular source of care for more than 1 year (82%). Sixty-three percent had seen their regular source of care for more than 2 years. The majority chose their own usual source of care (78%) and did not have trouble paying for their health care (74%). More than half of the respondents made at least 1 visit to a specialist (56%). This relatively high rate may be due to a somewhat elderly sample; more than 20% of the respondents were older than 65 years.

Table 1 compares the HMO sample with the CHC sample on sociodemographic and health care utilization measures. The HMO sample included predominantly white (81.6%) and higher income subjects (86.8% with annual household income of $25,000 or more). In contrast, the CHC sample included predominantly non-white (83.2%) and lower income subjects (85.9% with an annual household income less than $25,000). Compared with the CHC respondents, HMO subjects had been seeing their regular source of care for a longer time, were more likely to choose their own doctors and visit a specialist, and less likely to have trouble paying for their health care.

Factor Analysis and Construct Validity

In the initial exploratory factor analysis, all 92 applicable questionnaire items measuring the subdomains and domains of primary carefirst contact, longitudinality, comprehensiveness, coordination, family centeredness, community orientation, and cultural competencewere included. Based on the results of the initial factor analysis, 4 criteria were applied to reach the final solution (Table 2; initial factor analyses not shown but available upon request).

Seven common factors were extracted, corresponding to the hypothesized primary care scales: first contactaccessibility, first contactutilization, longitudinalityinterpersonal relationships, comprehensivenessservices available, comprehensivenessservices received, coordination, and community orientation (Table 2). Those extracted factors explained 88.1% of the common variance. Eigenvalues ranged from 16.17 to 1.16. All principal primary care domains were extracted as hypothesized. Only 1 of the 3 derivative features, community orientation, was separately identifiable.

Derivation and Reliability of the Primary Care Scales

Table 3 presents the results of the reliability analyses for both the original items and the final items (based on factor analysis). Item descriptive results (means and standard deviations) are also presented. Scale reliability measures include item-total correlation and alpha coefficient reliability. The distribution of the items varied significantly from a mean of 1.85 (ask about gun safety) to 3.73 (Provider answers questions in ways you understand) on the 4-point Likert-type scale. The distribution tends to skew toward more favorable answers (above 2.5). Apart from the gun safety item, only 2 items fell below a mean of 2 (1.94 for Provider knows neighborhood problems, 1.90 for Provider makes home visits). The first contactutilization and longitudiinalityinterpersonal relationships scales achieved the highest mean scores, whereas scales with lower means were community orientation, first contact-accessibility, and comprehensiveness-services received.

Eighteen of the 92 initial items were deleted on the basis of the criteria imposed for factor analyses. No items were deleted for first contact-utilization, coordination of services, comprehensiveness-services received, and community orientation scales. All items were deleted for family centeredness as were two thirds of the items for first contact-accessibility. Two items (out of 22) were deleted for longitudinality-interpersonal relationships and 3 (out of 24) for comprehensivenessservices available. Items from cultural competence were combined into first contact-accessibility. The revised scales demonstrate internal consistency reliability that was higher than or equal to the original scales, despite the reduction in number of items. Item-total correlations were also high and ranged from 0.34 (If sick, seen same day if office is open) to 0.91 (How to prevent hot water burns and How to prevent falls).

Testing the Likert Scaling Assumptions

Table 4 presents a summary of the results of the tests of Likert scaling assumptions using the revised items. All item-scale correlations well exceeded the accepted minimum (0.30) with the majority greater than 0.50 (Assumption 1). All 7 multi-item scales achieved 100% scaling success, indicating that all items in these scales correlated substantially higher with items in their hypothesized scale than with items in other scales (Assumption 2). Item means within each revised scale generally differed by less than six tenths of a point (except for first contact-accessibility) and item standard deviations within each scale by less than four tenths of a point (Assumption 3). Formal evidence of equal item variance was supported by the equivalence of the intraclass correlation and Scotts homogeneity ratio for each scale. Equal-item scale correlation (Assumption 4) was also observed through the range of item-scale correlations. As shown in column 1 (range of item-scale correlations), the range is relatively narrow (from .17 for coordination of services to .38 for comprehensiveness-services received). Finally, score reliability (Assumption 5) showed that except for first contact-utilization (only 3 items), all alpha levels exceeded .70 and were sufficiently high. Five of the 7 scales had alpha levels above .85.

Descriptive Feature of PCAT-AE

Table 5 displays estimates of central tendency and dispersion of scale score distributions for the 7 primary care scales in this South Carolina sample. Except for community orientation, all primary care scales were negatively skewed, indicating distributions with more positive ratings of primary care. The community orientation scale was positively skewed, indicating distributions with more negative ratings on the community orientation aspect of primary care. The full range of possible scores was observed for all scales except ongoing care.

The percentage of respondents scoring at the floor (the lowest score) or ceiling (the highest score) was acceptably low for all scales except first contactutilization, where 50% of the respondents scored the maximum score.

Table 6 compares the alpha coefficient and interfactor correlation for each primary care scale. The alpha coefficient of each scale substantially exceeded its correlation with all other primary care scales. None of the inter-factor correlations were excessively high, demonstrating that each primary care scale has significant unique contribution. All significant correlations were positive, indicating the complementary nature of primary care domains. Relatively high and positive interfactor correlations were observed between comprehensivenessservices received and comprehensiveness-services available (0.44), with the former and longitudinalityinterpersonal relationships (0.43), with the latter and coordination (0.38), and with comprehensivenessservices received and community orientation (0.37).

Discussion

Using patient-provided survey information collected within 2 health plans in South Carolina, we assessed the validity and reliability of the PCAT-AE. The results indicate that the hypothesized scales for primary care (first contactaccessibility, first con-tactutilization, longitudinalityinterpersonal relationships, comprehensivenessservices available, comprehensivenessservices received, and coordination) have substantial reliability and validity, consistent with the findings from the testing of the PCAT-CE.³⁰ The 2 versions of the instrument differ only in the comprehensiveness domains, as comprehensiveness implies that all common needs are met, and health needs in childhood are different from those in adults. In contrast, challenges to accessibility, to the nature of interpersonal relationships, and to coordination and community orientation are similar for both children and adults and thus can be assessed by the same items. Only 1 ancillary feature of primary care, community orientation, was retained as a separate dimension after factor analyses. The extracted factors explained 88.1 percent of the total variance in the item scores.

All of the 5 assumptions, including item-conver-gent validity, item-discriminant validity, equal item variance, equal item-scale correlation, and score reliability, were met. These results suggest that these items may be used to represent the primary care scales, and the scoring of these items may be summed without standardization or weighting, as with Likerts method of summated rating scales.³⁹

The resulting instrument has 74 items. Although the retained items adequately addressed first contactutilization, longitudinalityinterpersonal relationships, comprehensivenessservices available, comprehensivenessservices received, and coordination, and are consistent with the framework, those representing first contactaccessibility fell short. Only 4 of the 12 items measuring accessibility were retained. When more detail on accessibility is required, items that were deleted because they had lower item-total correlation may be added back in. Users should also review the comprehensiveness items to ascertain their relevance in the setting in which they are to be used. Items may be deleted if they are inappropriate in the context in which they are used; for example, in health systems that do not offer on-site testing for human immunodeficiency virus (HIV), because HIV is uncommon. Since continuity of care is an important component of primary care quality, a minimum number of visits or minimum duration with a regular source of care should be part of the assessment tool.

Separate factor analyses were performed with the 2 health plans. The results were largely comparable in terms of the factors that emerged as significant, indicating the generalizability of the tool to both vulnerable and middle-income populations. The only major differences are that the CHC subpopulation analysis yielded an additional significant factor, cultural competence, which the HMO subpopulation and the total population analyses failed to identify. In contrast, the HMO subpopulation analysis yielded an additional significant factor, family centeredness, which the CHC subpopulation and the total population analyses failed to identify. Thus, when using PCATon vulnerable populations (especially racial and ethnic minorities), questions measuring cultural competence might be retained. Family centeredness seemed to emerge as a distinct concept, primarily in the middle-income population.

There are a number of uses for a valid and reliable instrument such as the PCAT-AE. First, understanding primary care as a multidimensional concept is consistent with the IOMs conceptualization of primary care and more precisely captures the quality of primary care than unidimensional proxies, such as a clinicians medical specialty. With the 6 scales representing 4 core domains, the index representing strength of affiliation with a primary care provider, a scale for community orientation and the optional scales for family centeredness and cultural competence, all the important features of primary care are addressed. Second, PCAT-AE can be used as a quality measurement tool that assesses the adequacy of primary care experience rendered under different health care systems or settings, and for patients with different sociodemographic attributes. Third, PCAT-AE can also serve as a quality control tool that compares the quality of primary care given by providers of different types. The instrument can be used with other outcomes to assess the effect of policy interventions and systems changes on the delivery of critical aspects of primary care.

Limitations

Interpretation of our results should take into account some limitations. First, because our study was restricted to 1 locale, the generalizability of the PCAT-AE to other sites and states is not assured. Additional testing and validation is necessary to corroborate the current results. Second, the 74-item questionnaire remains lengthy and could have contributed to relatively high nonresponse and incompletion rates. Future validation work will concentrate on further reduction of the items to the very essential in order to reduce response burden. Regarding the ceiling effect of first contactutilization, future tests will be conducted in other settings with less of a managed care focus, as there well may be quite different distributions of responses in other settings. Third, outcomes of primary care are not the focus of the assessment tool. However, numerous studies have linked primary care to better health outcomes. Subsequent research may help explain which attributes are most conducive to better outcomes so that limited resources can be used to focus on them or a combination of them. Fourth, the measurement of primary care achievement is entirely based on respondents self-report. While self-report may be the best way to ascertain peoples experiences, it is subject to recall and response bias. Moreover, some aspects of technical quality cannot be assessed by patientsor consumers reports.

Despite these limitations, PCAT-AE is a valuable tool for capturing the principal domains of primary care. The next phase of our work seeks to assess the predictive validity of PCAT-AE, by examining the extent to which the principal attributes of primary care can be linked to the achievement of favorable health outcomes, their ability to manage their illnesses, and their satisfaction with the care received. Such work would advance our understanding of the relationship between how primary care is delivered and the health outcomes that result.

Related technical terms

Primary Care Attributes

First contactcare implies accessibility to and use of services for each new problem or new episode of a problem for which people seek health care.

Longitudinalitypresupposes the existence of a regular source of care and its use over time.

Comprehensivenessimplies that primary care facilities must be able to arrange for all types of health care services, including referrals to secondary services for consultation, tertiary services for specific conditions, and essential supporting services, such as home care and other community services.

Coordinationof care requires some form of continuity, either by practitioners, medical records, or both, as well as recognition of problems that are addressed elsewhere and the integration of their care into the total care of patients.

Family centerednessrefers to recognition of family factors related to the genesis and management of illness.

Community orientationrefers to the providers knowledge of community needs and involvement in the community.

Cultural competencerefers to the providers adaptation to facilitate relationships with populations having special cultural characteristics.

Measurement Concepts

Measurement validityrefers to the extent that important dimensions of a concept and their categories have been taken into account and appropriately operationalized.

Measurement reliabilityrefers to the extent that consistent results are obtained when a particular measure is applied to similar elements.

Construct validityis present when the measure captures the major dimensions of the concept under study.

Content validityrefers to the representativeness of the response categories used to represent each of the dimensions of a concept.

Concurrent validitymay be tested by comparing results of one measurement with those of a similar measurement administered to the same population and at approximately the same time. If both measurements yield similar results, then concurrent validity can be established.

Predictive validity exists when the results obtained from the measurement succeed in predicting the expected later-occurring event or circumstance.

Test-retest reliabilityinvolves administering the same measurement to the same individuals at 2 different times. If the correlation between the same measures is high, then the measurement is believed to be reliable.

Split-half reliabilityinvolves preparing 2 sets of measurement of the same concept, applying them to research subjects at one setting, and comparing the correlation between the 2 sets of measurement. To the extent the correlation is high, then the measurement is reliable.

Interrater reliabilityinvolves using different people to conduct the same procedure, whether it be interview, observation, coding, rating, and the like, and comparing the results of their work. To the extent that the results are highly similar, interrater reliability is established.

Item-convergent validityrefers to the substantial correlation between each item and its hypothesized scale.

Item-discriminant validityrefers to items within a scale that correlate more substantially with their hypothesized scale than with any other scale.

Equal item variancerefers to items within a scale that have approximately equal means and variances.

Equal item-scale correlationrefers to items in a scale that contribute approximately the same proportion of information about the underlying concept.

Score reliabilityrefers to scores of scales that are reproducible and reliable.

Skewnessrefers to distribution of observations that is not symmetric, ie, when more observations are found at one end of the distribution than the other.

Kurtosisrefers to the extent observations cluster around a central point more than in normal distribution.

The growth of managed health care organizations and their emphasis on the use of primary care providers as gatekeepers has radically changed the value of a specialty designated as a provider of primary care. Classification as a primary care physician has become important to physicians because it provides a patient base and source of revenue, and it is important to patients because it allows direct access to those physicians.¹ For the most part, general internists, family physicians, and pediatricians are designated as primary care specialists. Increasingly, however, there have been efforts at the state and federal levels to designate obstetrician-gynecologists (OBGs) as primary care providers for women. A bill was introduced in 1999 before both houses of the 106th Congress (S. 6 and H.R. 358, The Patients’ Bill of Rights Act of 1999) that would allow women to choose OBGs as their primary care physicians, and it is still awaiting action.

Traditionally, the specialty of obstetrics-gynecology has considered itself expert in the areas of reproductive health and gynecologic diseases.^2-4 Recently, changing practice philosophy has resulted in an increasing emphasis on providing general medical care.^5-7 In 1993, the American College of Obstetricians and Gynecologists (ACOG) formed a Task Force on Primary and Preventive Healthcare that identified 3 levels of care that can be provided by OBGs: traditional specialty care, primary preventive care, and extended primary care.⁸ OBGs providing primary preventive care take a broader role in health maintenance for women, including health screening and disease prevention. Those providing extended primary care offer primary preventive care and treat medical conditions beyond those pertaining to the reproductive system. Although OBGs have been divided over their role as primary care physicians, a primary care requirement in residency training was implemented in 1996.^3,5

The Committee on the Future of Primary Care of the Institute of Medicine (IOM) defined primary care as “the provision of integrated, accessible health care services by clinicians who are accountable for addressing a large majority of personal health care needs.”⁹ Included in the description of integrated is comprehensive, which means providing care for any health problem at any given stage of a patient’s life cycle. Addressing the majority of health care needs refers to the primary care physician receiving all problems that patients bring—unrestricted by organ system—and having the appropriate training to diagnose and manage the majority of them. Caring for a broad spectrum of medical problems encompassing many organ systems is a component of primary care, and this component is what is included in the ACOG definition of extended primary care. Throughout this article the term primary care will be used to represent the broad-spectrum care accepted by the IOM and ACOG as an element of primary care.

Few studies have measured the degree to which OBGs provide primary care.^10-13 Of those that have, many are based on surveys completed by patients or physicians and are focused on women in their reproductive years. Horton and colleagues¹⁰ surveyed a national random sample of 1250 ACOG members in a variety of practices and found that more than 90% of the responding OBGs performed blood pressure screening, breast examinations, mammography, and Papanicolaou tests. Hendrix and coworkers11 reviewed 739 patient encounters from 335 charts of the private practices of faculty in the Department of Obstetrics and Gynecology at Wayne State University and found that of nonobstetrical visits, 80% were for primary gynecologic care and 7% for primary nongynecologic care. Leader and Perales¹² reviewed data from a 1991 economic survey conducted by ACOG of a stratified random sample of 2000 of its members practicing in the United States. Of 1286 respondants, 48% considered themselves primary care providers. A recent study by Jacoby and colleagues¹⁴ used Medicare claims data to examine the scope of care that OBGs provided to their elderly patients. They found that OBGs provided a substantial amount of preventive care but not much nongynecologic care for elderly women.

We extend the work of these investigators by further exploring the breadth of medical conditions for which OBGs provide care to their elderly patients and by examining the degree to which OBGs in both rural and urban areas care for nongynecologic conditions. Our findings can offer some understanding of the potential impact of legislation to designate OBGs as primary care providers for elderly women.

Methods

Data Sources

We used the 1994 Washington State Medicare Part B claims file as the data source for our study. This database, part of the National Claims History File of the Health Care Financing Administration (HCFA), is an administrative data set that captures diagnostic and therapeutic information about services billed to Medicare. The Medicare Part B file contains a series of line items with each representing a discrete billable service for a Medicare beneficiary. These line items included identifiers for the patient receiving the service, the physician providing the service, the diagnosis coded, and the date of the visit. Items submitted by physicians include a unique physician identification number (UPIN) used to designate the specialty of the physician providing these services.

Physician Sample

All physicians practicing in the state of Washington and submitting Medicare claims for patient visits in 1994 were eligible for the study. We focused on the approximately 80% of OBGs who were participating in Medicare Part B. A variety of subspecialties including 5 medical (dermatology, cardiology, gastroenterology, pulmonology, and rheumatology) and 4 surgical (general surgery, orthopedic surgery, otolaryngology, and urology) were selected for comparison. Family practice and internal medicine were included in the descriptive analysis of visit frequency but not in the domain analysis, because all of primary care for elderly women was considered within the domain of these 2 specialties.

Information from the HCFA UPIN National Directory, the American Board of Medical Specialties (ABMS) database, and the American Medical Association (AMA) masterfile were used to link specialty information to the UPINs in the Part B Medicare file. Physicians were designated as a specific specialty type if both the ABMS certification and the primary self-designated specialty captured in the AMA masterfile were the same. In cases where these differed, the physicians were excluded. Including only those physicians who had the same specialty of training as their reported specialty of practice ensured accurate assignment of specialty.

Practice Location

Physicians were designated as practicing in rural or urban areas based on the ZIP codes of their practice addresses. ZIP codes were assigned as rural or urban on the basis of their proximity to a hospital classified as such by the Washington State Office of Rural Health of the Department of Health. The 5 physicians whose practice addresses were unknown or were in both rural and urban areas were excluded from the practice location analysis.

Patient Visits and Sample

We aggregated all outpatient physician services (eg, diagnostic tests and procedures) provided to an individual on a single date by the same provider into medical encounters. We used the current procedural terminology (CPT) and the HCFA common procedure coding system (HCPCS) for these services to determine whether they involved face-to-face contact with a physician. Those encounters that included such contact were considered visits. Within each visit, we chose one face-to-face line item—the index line—to identify the primary diagnosis for that visit. This index line either contained the evaluation and management code or, in cases without such a code, the face-to-face line item with the highest allowable charge. The International Classification of Diseases, Ninth Revision (ICD-9) diagnosis codes for these index lines were used to identify the primary diagnosis for each visit.¹⁵

We included women patients of the study physicians who were aged 65 years and older, enrolled in Medicare Part B, and alive throughout 1994. We excluded the 15% of patients enrolled in a managed care plan during the year to select elderly patients with unrestricted access to physicians in any specialty and because patient visit data were not available for those enrolled in those plans.

Designation of Specialty Domain

To interpret the claims-based diagnoses for each visit, we used diagnosis clusters to collapse the ICD-9 system into 120 groupings of individual diagnostic codes.^16,17 The codes in each group share similar pathophysiologic characteristics that often receive similar management. Two physicians reviewed the diagnosis clusters and nonclustered individual ICD-9 diagnoses to identify conditions traditionally cared for by physicians in each specialty. These diagnoses were designated as “in domain”; all others were designated as “out of domain.” Domain assignments were then reviewed and revised by the Medicare Carrier Advisory Committee, which was composed of specialists including all of those represented in our study.

For example, the diagnosis of peptic ulcer disease is considered out of domain for an OBG but in domain for a gastroenterologist, while cervical dysplasia is considered in domain for an OBG but out of domain for a gastroenterologist. We considered out of domain care as a measure of the breadth of care provided. The more out of domain care an OBG provided, the greater the degree of primary care that he or she was providing.

Analysis

We examined the distribution of out of domain visits for individual OBGs and the frequency of all diagnoses and out of domain diagnoses for OBGs in total. We also described visit volumes and percentage of out of domain visits in both rural and urban areas for OBGs and other physician specialists.

Results

Of the 328 Washington physicians in 1994 who submitted Medicare claims and were designated as OBGs by either the ABMS or the AMA masterfile, 285 were designated OBGs by both the ABMS and the AMA masterfile. These 285 OBGs treated 10,522 Medicare patients for 16,743 visits. These patients made a total of 108,720 visits to all physicians during the year. The patients of OBGs averaged 1.6 visits to them and 10.3 visits to any physician over the year. Patients who visited OBGs received 15.4% of their total health care from them.

The visit rate to OBGs (1.6) was the lowest for all the specialties studied; however, it closely resembled rates from other surgical specialties that ranged from 1.8 to 2.4 visits per patient. Medical specialists averaged 1.9 to 3.7 visits per patient, and traditional generalists ranged from 3.7 to 3.8 visits per patient. Of the specialties studied, the percentage of overall health care for patients of physicians of a certain specialty received by those physicians was the third lowest for OBGs (15.4%). This amount was similar to surgical specialists who ranged from 15.3% to 18.9%. Medical specialists ranged from 17.8% to 27.1%, and the patients who saw a family physician or general internist received more than 40% of their total health care from them during the year.

Of the 16,743 visits made to OBGs, 12.2% had diagnoses that were out of domain for the specialty Table 1. Among surgical specialists, general surgeons had the highest percentage of out of domain visits (21.9%), while others had a much lower proportion of out of domain visits (1.5% to 4.5%). Among medical specialists, pulmonologists had the highest percentage of out of domain visits (29.7%); others ranged from 1.2% to 15.0%.

Almost 12% of OBGs had more than 30% of their visits out of domain Figure 1. The median percentage of out of domain visits for individual OBGs was 6.7%. Table 1 shows that rural OBGs provided less out of domain care (8.2%) than their urban counterparts (12.9%). General surgeons were similar to OBGs in providing less out of domain care in the rural setting than urban setting. All the other specialties had the same or more out of domain care in the rural setting than in the urban setting.

Two of the 15 most frequent diagnosis clusters recorded by OBGs were for out of domain care Table 2. The general medical examination was the fifth most frequently reported diagnosis cluster and the most frequent out of domain diagnosis cluster Table 3. Hypertension was the other out of domain diagnosis cluster in the top 15 and comprised 9.0% of all of the out of domain diagnoses. The other out of domain problems diagnosed by OBGs include a full range of primary care conditions.

Discussion

Our study demonstrates that during 1994, the large majority of OBGs provided a limited amount of nongynecologic care to their elderly patients. This is consistent with the findings of other studies examining the scope of OBGs’ practices and suggests that in general OBGs were not serving as primary care providers to their elderly patients.^{11,13-15,17,18}

The visit rates to OBGs were more similar to surgical subspecialties than to the traditional primary care specialties. The number of visits per Medicare patient to OBGs was 1.6, the lowest of the specialties studied and very different from that of traditional primary care specialists (3.7 to 3.8 per patient). Patients of OBGs received 15.4% of their total health care from OBGs, which was much lower than the amount of overall health care that traditional generalists provided to their patients (42.0%-42.9%). This suggests that OBGs were primarily seeing elderly patients in consultation for gynecologic problems. This is consistent with a number of studies that have demonstrated that elderly women are less likely to receive care from OBGs.^10,12,19-22

Our finding that few OBGs provided broad-spectrum care in 1994 suggests that legislation to increase the use of OBGs as primary care providers could affect the way general medical services are delivered to elderly women. Many of these women may have nongynecologic medical conditions requiring treatment and monitoring. If most OBGs do not routinely provide these services, these women will require referral to medical specialists, which could lead to increasing costs, inconvenience, and fragmentation of care.

We did not expect the findings that rural OBGs provided less nongynecologic care than urban OBGs. Because rural areas are often underserved, we hypothesized that rural OBGs would practice as generalists more often and provide more general medical care. We concluded, therefore, that rural OBGs may be in shorter supply than generalists and that they are filling their practices with visits specific to their specialty. In addition, there may be increased competition among urban OBGs for gynecologic visits, so a larger number of patients are seen for nongynecologic problems.

Limitations

Our study has several limitations. We included only the primary diagnosis for each visit. A patient may have presented with both a gynecologic and non-gynecologic problem, but the OBG may have coded the gynecologic diagnosis as the primary one. In addition, patients presenting for a gynecologic complaint may inquire as an aside about a nongynecologic concern that may have been addressed but not coded. This would underestimate the amount of nongynecologic care provided by OBGs as well as the other studied specialists. However, it is unlikely that every time a woman visits an OBG for an upper respiratory infection, joint pain, or glycohemoglobin monitoring she also has an active gynecologic problem. The methodology we used should identify the portion of OBGs who provided a substantial amount of nongynecologic care.

The scope of OBGs’ practices may have changed since the data were collected. The primary reason for such a change would be the implementation of the primary care requirement during residency. Since that change in training occurred in 1996, however, those residents affected are just now entering the work force. Thus, our study’s data are likely to represent current practice patterns. A follow-up study of the scope of OBGs’ practices in 5 to 10 years will help elucidate the effect of the residency primary care requirement on OBGs’ practices.

The assignment of a diagnosis as in or out of domain is subject to interpretation. ICD-9 codes and diagnosis clusters were reviewed separately by 2 physicians to classify the diagnoses. When there was disagreement, the individuals discussed the diagnosis to reach consensus. If they could not agree, the diagnosis was considered out of domain. For example, the general medical examination was considered out of domain because the exact nature of the visit is unclear; however, the diagnosis may have been used for general gynecologic services provided to women without specific diagnoses, such as annual Papanicolaou tests that would be considered in domain. Since this was a conservative approach, it could overestimate the amount of out of domain care provided.

Only one feature of primary care—the breadth of practice—was addressed in this study. We did not examine a number of other features that also characterize primary care—continuity, coordination, and accessibility—as also described by the Committee on the Future of Primary Care of the IOM.⁹ In addition, we did not address quality of care.

Since we limited our study to Washington Medicare beneficiaries aged 65 years and older, we cannot comment on the degree to which OBGs may be providing general medical care to patients younger than 65 years. Younger patients may have different relationships with their physicians and present with different medical issues of varying complexity. We also excluded the 15% of Medicare elderly in Washington who in 1994 were enrolled in a managed care health plan. These results cannot be extrapolated to this population, because nearly all health maintenance organizations restrict access to specialists.

The Effect of Legislation

Our findings raise the question of whether legislation that designates OBGs as primary care providers for elderly women would result in an increase in the use of OBGs as providers of care for problems outside the reproductive system or primarily increase access to OBGs’ specialty services. Passage of legislation such as the Patients’ Bill of Rights Act of 1999 may facilitate elderly women’s obtaining primary gynecologic care, yet it may not have the same effect on their receipt of general medical care. A bill like this would allow women of all ages to designate OBGs as their primary care providers, thus allowing unrestricted and direct access to their services. Studies investigating the care received by elderly women enrolled in private health care plans in which they are able to select OBGs as primary care physicians could provide additional useful information. If the Patients’ Bill of Rights Act of 1999 or similar legislation is passed, studies will be needed to assess its effect on the overall medical care received by elderly women.

Conclusions

In 1994, few OBGs were providing the level of care that ACOG designates as extended primary care and that the IOM considers primary care. Our study does not reflect the degree to which they might have been providing primary preventive care and does not examine OBGs’ abilities to provide care for nongynecologic problems. The extent to which these findings represent current practice is also unknown. Nonetheless, our findings provide a baseline for the type of care provided by OBGs and suggest that without changes in the scope of OBGs’ practice, legislation resulting in more elderly women using OBGs as their primary providers could result in greater fragmentation and costs for their overall medical care.

Alternatively, some OBGs could embrace the movement within their specialty to emphasize treating patients’ general medical problems. OBGs develop close relationships with patients during their reproductive years, and these patients may benefit from a continuing relationship with these physicians as they age. There have been recent changes in obstetrics-gynecology residency requirements, increasing the time spent training in general medicine to better prepare residents to provide nongynecologic care. Other specialties have developed training tracks to specifically prepare physicians to practice primary care. Perhaps this is the time for residency programs in obstetrics-gynecology to do the same for a subset of their residents specifically interested in providing primary care.

Acknowledgments

Our research was supported by grants from the Robert Wood Johnson Foundation, Princeton, NJ, and the Office of Rural Health Policy and the Agency for Health Care Policy and Research of the US Public Health Service, Washington, DC. The views expressed in this article are those of the authors and do not necessarily represent those of the University of Washington, the Health Care Financing Administration, or the Robert Wood Johnson Foundation. The authors would like to acknowledge Peter Houck, MD, for his contributions to the manuscript and Durlin Hickok, MD, MPH, for reviewing the manuscript.

The growth of managed health care organizations and their emphasis on the use of primary care providers as gatekeepers has radically changed the value of a specialty designated as a provider of primary care. Classification as a primary care physician has become important to physicians because it provides a patient base and source of revenue, and it is important to patients because it allows direct access to those physicians.¹ For the most part, general internists, family physicians, and pediatricians are designated as primary care specialists. Increasingly, however, there have been efforts at the state and federal levels to designate obstetrician-gynecologists (OBGs) as primary care providers for women. A bill was introduced in 1999 before both houses of the 106th Congress (S. 6 and H.R. 358, The Patients’ Bill of Rights Act of 1999) that would allow women to choose OBGs as their primary care physicians, and it is still awaiting action.

Traditionally, the specialty of obstetrics-gynecology has considered itself expert in the areas of reproductive health and gynecologic diseases.^2-4 Recently, changing practice philosophy has resulted in an increasing emphasis on providing general medical care.^5-7 In 1993, the American College of Obstetricians and Gynecologists (ACOG) formed a Task Force on Primary and Preventive Healthcare that identified 3 levels of care that can be provided by OBGs: traditional specialty care, primary preventive care, and extended primary care.⁸ OBGs providing primary preventive care take a broader role in health maintenance for women, including health screening and disease prevention. Those providing extended primary care offer primary preventive care and treat medical conditions beyond those pertaining to the reproductive system. Although OBGs have been divided over their role as primary care physicians, a primary care requirement in residency training was implemented in 1996.^3,5

The Committee on the Future of Primary Care of the Institute of Medicine (IOM) defined primary care as “the provision of integrated, accessible health care services by clinicians who are accountable for addressing a large majority of personal health care needs.”⁹ Included in the description of integrated is comprehensive, which means providing care for any health problem at any given stage of a patient’s life cycle. Addressing the majority of health care needs refers to the primary care physician receiving all problems that patients bring—unrestricted by organ system—and having the appropriate training to diagnose and manage the majority of them. Caring for a broad spectrum of medical problems encompassing many organ systems is a component of primary care, and this component is what is included in the ACOG definition of extended primary care. Throughout this article the term primary care will be used to represent the broad-spectrum care accepted by the IOM and ACOG as an element of primary care.

Few studies have measured the degree to which OBGs provide primary care.^10-13 Of those that have, many are based on surveys completed by patients or physicians and are focused on women in their reproductive years. Horton and colleagues¹⁰ surveyed a national random sample of 1250 ACOG members in a variety of practices and found that more than 90% of the responding OBGs performed blood pressure screening, breast examinations, mammography, and Papanicolaou tests. Hendrix and coworkers11 reviewed 739 patient encounters from 335 charts of the private practices of faculty in the Department of Obstetrics and Gynecology at Wayne State University and found that of nonobstetrical visits, 80% were for primary gynecologic care and 7% for primary nongynecologic care. Leader and Perales¹² reviewed data from a 1991 economic survey conducted by ACOG of a stratified random sample of 2000 of its members practicing in the United States. Of 1286 respondants, 48% considered themselves primary care providers. A recent study by Jacoby and colleagues¹⁴ used Medicare claims data to examine the scope of care that OBGs provided to their elderly patients. They found that OBGs provided a substantial amount of preventive care but not much nongynecologic care for elderly women.

We extend the work of these investigators by further exploring the breadth of medical conditions for which OBGs provide care to their elderly patients and by examining the degree to which OBGs in both rural and urban areas care for nongynecologic conditions. Our findings can offer some understanding of the potential impact of legislation to designate OBGs as primary care providers for elderly women.

Methods

Data Sources

We used the 1994 Washington State Medicare Part B claims file as the data source for our study. This database, part of the National Claims History File of the Health Care Financing Administration (HCFA), is an administrative data set that captures diagnostic and therapeutic information about services billed to Medicare. The Medicare Part B file contains a series of line items with each representing a discrete billable service for a Medicare beneficiary. These line items included identifiers for the patient receiving the service, the physician providing the service, the diagnosis coded, and the date of the visit. Items submitted by physicians include a unique physician identification number (UPIN) used to designate the specialty of the physician providing these services.

Physician Sample

All physicians practicing in the state of Washington and submitting Medicare claims for patient visits in 1994 were eligible for the study. We focused on the approximately 80% of OBGs who were participating in Medicare Part B. A variety of subspecialties including 5 medical (dermatology, cardiology, gastroenterology, pulmonology, and rheumatology) and 4 surgical (general surgery, orthopedic surgery, otolaryngology, and urology) were selected for comparison. Family practice and internal medicine were included in the descriptive analysis of visit frequency but not in the domain analysis, because all of primary care for elderly women was considered within the domain of these 2 specialties.

Information from the HCFA UPIN National Directory, the American Board of Medical Specialties (ABMS) database, and the American Medical Association (AMA) masterfile were used to link specialty information to the UPINs in the Part B Medicare file. Physicians were designated as a specific specialty type if both the ABMS certification and the primary self-designated specialty captured in the AMA masterfile were the same. In cases where these differed, the physicians were excluded. Including only those physicians who had the same specialty of training as their reported specialty of practice ensured accurate assignment of specialty.

Practice Location

Physicians were designated as practicing in rural or urban areas based on the ZIP codes of their practice addresses. ZIP codes were assigned as rural or urban on the basis of their proximity to a hospital classified as such by the Washington State Office of Rural Health of the Department of Health. The 5 physicians whose practice addresses were unknown or were in both rural and urban areas were excluded from the practice location analysis.

Patient Visits and Sample

We aggregated all outpatient physician services (eg, diagnostic tests and procedures) provided to an individual on a single date by the same provider into medical encounters. We used the current procedural terminology (CPT) and the HCFA common procedure coding system (HCPCS) for these services to determine whether they involved face-to-face contact with a physician. Those encounters that included such contact were considered visits. Within each visit, we chose one face-to-face line item—the index line—to identify the primary diagnosis for that visit. This index line either contained the evaluation and management code or, in cases without such a code, the face-to-face line item with the highest allowable charge. The International Classification of Diseases, Ninth Revision (ICD-9) diagnosis codes for these index lines were used to identify the primary diagnosis for each visit.¹⁵

We included women patients of the study physicians who were aged 65 years and older, enrolled in Medicare Part B, and alive throughout 1994. We excluded the 15% of patients enrolled in a managed care plan during the year to select elderly patients with unrestricted access to physicians in any specialty and because patient visit data were not available for those enrolled in those plans.

Designation of Specialty Domain

To interpret the claims-based diagnoses for each visit, we used diagnosis clusters to collapse the ICD-9 system into 120 groupings of individual diagnostic codes.^16,17 The codes in each group share similar pathophysiologic characteristics that often receive similar management. Two physicians reviewed the diagnosis clusters and nonclustered individual ICD-9 diagnoses to identify conditions traditionally cared for by physicians in each specialty. These diagnoses were designated as “in domain”; all others were designated as “out of domain.” Domain assignments were then reviewed and revised by the Medicare Carrier Advisory Committee, which was composed of specialists including all of those represented in our study.

For example, the diagnosis of peptic ulcer disease is considered out of domain for an OBG but in domain for a gastroenterologist, while cervical dysplasia is considered in domain for an OBG but out of domain for a gastroenterologist. We considered out of domain care as a measure of the breadth of care provided. The more out of domain care an OBG provided, the greater the degree of primary care that he or she was providing.

Analysis

We examined the distribution of out of domain visits for individual OBGs and the frequency of all diagnoses and out of domain diagnoses for OBGs in total. We also described visit volumes and percentage of out of domain visits in both rural and urban areas for OBGs and other physician specialists.

Results

Of the 328 Washington physicians in 1994 who submitted Medicare claims and were designated as OBGs by either the ABMS or the AMA masterfile, 285 were designated OBGs by both the ABMS and the AMA masterfile. These 285 OBGs treated 10,522 Medicare patients for 16,743 visits. These patients made a total of 108,720 visits to all physicians during the year. The patients of OBGs averaged 1.6 visits to them and 10.3 visits to any physician over the year. Patients who visited OBGs received 15.4% of their total health care from them.

The visit rate to OBGs (1.6) was the lowest for all the specialties studied; however, it closely resembled rates from other surgical specialties that ranged from 1.8 to 2.4 visits per patient. Medical specialists averaged 1.9 to 3.7 visits per patient, and traditional generalists ranged from 3.7 to 3.8 visits per patient. Of the specialties studied, the percentage of overall health care for patients of physicians of a certain specialty received by those physicians was the third lowest for OBGs (15.4%). This amount was similar to surgical specialists who ranged from 15.3% to 18.9%. Medical specialists ranged from 17.8% to 27.1%, and the patients who saw a family physician or general internist received more than 40% of their total health care from them during the year.

Of the 16,743 visits made to OBGs, 12.2% had diagnoses that were out of domain for the specialty Table 1. Among surgical specialists, general surgeons had the highest percentage of out of domain visits (21.9%), while others had a much lower proportion of out of domain visits (1.5% to 4.5%). Among medical specialists, pulmonologists had the highest percentage of out of domain visits (29.7%); others ranged from 1.2% to 15.0%.

Almost 12% of OBGs had more than 30% of their visits out of domain Figure 1. The median percentage of out of domain visits for individual OBGs was 6.7%. Table 1 shows that rural OBGs provided less out of domain care (8.2%) than their urban counterparts (12.9%). General surgeons were similar to OBGs in providing less out of domain care in the rural setting than urban setting. All the other specialties had the same or more out of domain care in the rural setting than in the urban setting.

Two of the 15 most frequent diagnosis clusters recorded by OBGs were for out of domain care Table 2. The general medical examination was the fifth most frequently reported diagnosis cluster and the most frequent out of domain diagnosis cluster Table 3. Hypertension was the other out of domain diagnosis cluster in the top 15 and comprised 9.0% of all of the out of domain diagnoses. The other out of domain problems diagnosed by OBGs include a full range of primary care conditions.

Discussion

Our study demonstrates that during 1994, the large majority of OBGs provided a limited amount of nongynecologic care to their elderly patients. This is consistent with the findings of other studies examining the scope of OBGs’ practices and suggests that in general OBGs were not serving as primary care providers to their elderly patients.^{11,13-15,17,18}

The visit rates to OBGs were more similar to surgical subspecialties than to the traditional primary care specialties. The number of visits per Medicare patient to OBGs was 1.6, the lowest of the specialties studied and very different from that of traditional primary care specialists (3.7 to 3.8 per patient). Patients of OBGs received 15.4% of their total health care from OBGs, which was much lower than the amount of overall health care that traditional generalists provided to their patients (42.0%-42.9%). This suggests that OBGs were primarily seeing elderly patients in consultation for gynecologic problems. This is consistent with a number of studies that have demonstrated that elderly women are less likely to receive care from OBGs.^10,12,19-22

Our finding that few OBGs provided broad-spectrum care in 1994 suggests that legislation to increase the use of OBGs as primary care providers could affect the way general medical services are delivered to elderly women. Many of these women may have nongynecologic medical conditions requiring treatment and monitoring. If most OBGs do not routinely provide these services, these women will require referral to medical specialists, which could lead to increasing costs, inconvenience, and fragmentation of care.

We did not expect the findings that rural OBGs provided less nongynecologic care than urban OBGs. Because rural areas are often underserved, we hypothesized that rural OBGs would practice as generalists more often and provide more general medical care. We concluded, therefore, that rural OBGs may be in shorter supply than generalists and that they are filling their practices with visits specific to their specialty. In addition, there may be increased competition among urban OBGs for gynecologic visits, so a larger number of patients are seen for nongynecologic problems.

Limitations

Our study has several limitations. We included only the primary diagnosis for each visit. A patient may have presented with both a gynecologic and non-gynecologic problem, but the OBG may have coded the gynecologic diagnosis as the primary one. In addition, patients presenting for a gynecologic complaint may inquire as an aside about a nongynecologic concern that may have been addressed but not coded. This would underestimate the amount of nongynecologic care provided by OBGs as well as the other studied specialists. However, it is unlikely that every time a woman visits an OBG for an upper respiratory infection, joint pain, or glycohemoglobin monitoring she also has an active gynecologic problem. The methodology we used should identify the portion of OBGs who provided a substantial amount of nongynecologic care.

The scope of OBGs’ practices may have changed since the data were collected. The primary reason for such a change would be the implementation of the primary care requirement during residency. Since that change in training occurred in 1996, however, those residents affected are just now entering the work force. Thus, our study’s data are likely to represent current practice patterns. A follow-up study of the scope of OBGs’ practices in 5 to 10 years will help elucidate the effect of the residency primary care requirement on OBGs’ practices.

The assignment of a diagnosis as in or out of domain is subject to interpretation. ICD-9 codes and diagnosis clusters were reviewed separately by 2 physicians to classify the diagnoses. When there was disagreement, the individuals discussed the diagnosis to reach consensus. If they could not agree, the diagnosis was considered out of domain. For example, the general medical examination was considered out of domain because the exact nature of the visit is unclear; however, the diagnosis may have been used for general gynecologic services provided to women without specific diagnoses, such as annual Papanicolaou tests that would be considered in domain. Since this was a conservative approach, it could overestimate the amount of out of domain care provided.

Only one feature of primary care—the breadth of practice—was addressed in this study. We did not examine a number of other features that also characterize primary care—continuity, coordination, and accessibility—as also described by the Committee on the Future of Primary Care of the IOM.⁹ In addition, we did not address quality of care.

Since we limited our study to Washington Medicare beneficiaries aged 65 years and older, we cannot comment on the degree to which OBGs may be providing general medical care to patients younger than 65 years. Younger patients may have different relationships with their physicians and present with different medical issues of varying complexity. We also excluded the 15% of Medicare elderly in Washington who in 1994 were enrolled in a managed care health plan. These results cannot be extrapolated to this population, because nearly all health maintenance organizations restrict access to specialists.

The Effect of Legislation

Our findings raise the question of whether legislation that designates OBGs as primary care providers for elderly women would result in an increase in the use of OBGs as providers of care for problems outside the reproductive system or primarily increase access to OBGs’ specialty services. Passage of legislation such as the Patients’ Bill of Rights Act of 1999 may facilitate elderly women’s obtaining primary gynecologic care, yet it may not have the same effect on their receipt of general medical care. A bill like this would allow women of all ages to designate OBGs as their primary care providers, thus allowing unrestricted and direct access to their services. Studies investigating the care received by elderly women enrolled in private health care plans in which they are able to select OBGs as primary care physicians could provide additional useful information. If the Patients’ Bill of Rights Act of 1999 or similar legislation is passed, studies will be needed to assess its effect on the overall medical care received by elderly women.

Conclusions

In 1994, few OBGs were providing the level of care that ACOG designates as extended primary care and that the IOM considers primary care. Our study does not reflect the degree to which they might have been providing primary preventive care and does not examine OBGs’ abilities to provide care for nongynecologic problems. The extent to which these findings represent current practice is also unknown. Nonetheless, our findings provide a baseline for the type of care provided by OBGs and suggest that without changes in the scope of OBGs’ practice, legislation resulting in more elderly women using OBGs as their primary providers could result in greater fragmentation and costs for their overall medical care.

Alternatively, some OBGs could embrace the movement within their specialty to emphasize treating patients’ general medical problems. OBGs develop close relationships with patients during their reproductive years, and these patients may benefit from a continuing relationship with these physicians as they age. There have been recent changes in obstetrics-gynecology residency requirements, increasing the time spent training in general medicine to better prepare residents to provide nongynecologic care. Other specialties have developed training tracks to specifically prepare physicians to practice primary care. Perhaps this is the time for residency programs in obstetrics-gynecology to do the same for a subset of their residents specifically interested in providing primary care.

Acknowledgments

Our research was supported by grants from the Robert Wood Johnson Foundation, Princeton, NJ, and the Office of Rural Health Policy and the Agency for Health Care Policy and Research of the US Public Health Service, Washington, DC. The views expressed in this article are those of the authors and do not necessarily represent those of the University of Washington, the Health Care Financing Administration, or the Robert Wood Johnson Foundation. The authors would like to acknowledge Peter Houck, MD, for his contributions to the manuscript and Durlin Hickok, MD, MPH, for reviewing the manuscript.

The growth of managed health care organizations and their emphasis on the use of primary care providers as gatekeepers has radically changed the value of a specialty designated as a provider of primary care. Classification as a primary care physician has become important to physicians because it provides a patient base and source of revenue, and it is important to patients because it allows direct access to those physicians.¹ For the most part, general internists, family physicians, and pediatricians are designated as primary care specialists. Increasingly, however, there have been efforts at the state and federal levels to designate obstetrician-gynecologists (OBGs) as primary care providers for women. A bill was introduced in 1999 before both houses of the 106th Congress (S. 6 and H.R. 358, The Patients’ Bill of Rights Act of 1999) that would allow women to choose OBGs as their primary care physicians, and it is still awaiting action.

Traditionally, the specialty of obstetrics-gynecology has considered itself expert in the areas of reproductive health and gynecologic diseases.^2-4 Recently, changing practice philosophy has resulted in an increasing emphasis on providing general medical care.^5-7 In 1993, the American College of Obstetricians and Gynecologists (ACOG) formed a Task Force on Primary and Preventive Healthcare that identified 3 levels of care that can be provided by OBGs: traditional specialty care, primary preventive care, and extended primary care.⁸ OBGs providing primary preventive care take a broader role in health maintenance for women, including health screening and disease prevention. Those providing extended primary care offer primary preventive care and treat medical conditions beyond those pertaining to the reproductive system. Although OBGs have been divided over their role as primary care physicians, a primary care requirement in residency training was implemented in 1996.^3,5

The Committee on the Future of Primary Care of the Institute of Medicine (IOM) defined primary care as “the provision of integrated, accessible health care services by clinicians who are accountable for addressing a large majority of personal health care needs.”⁹ Included in the description of integrated is comprehensive, which means providing care for any health problem at any given stage of a patient’s life cycle. Addressing the majority of health care needs refers to the primary care physician receiving all problems that patients bring—unrestricted by organ system—and having the appropriate training to diagnose and manage the majority of them. Caring for a broad spectrum of medical problems encompassing many organ systems is a component of primary care, and this component is what is included in the ACOG definition of extended primary care. Throughout this article the term primary care will be used to represent the broad-spectrum care accepted by the IOM and ACOG as an element of primary care.

Few studies have measured the degree to which OBGs provide primary care.^10-13 Of those that have, many are based on surveys completed by patients or physicians and are focused on women in their reproductive years. Horton and colleagues¹⁰ surveyed a national random sample of 1250 ACOG members in a variety of practices and found that more than 90% of the responding OBGs performed blood pressure screening, breast examinations, mammography, and Papanicolaou tests. Hendrix and coworkers11 reviewed 739 patient encounters from 335 charts of the private practices of faculty in the Department of Obstetrics and Gynecology at Wayne State University and found that of nonobstetrical visits, 80% were for primary gynecologic care and 7% for primary nongynecologic care. Leader and Perales¹² reviewed data from a 1991 economic survey conducted by ACOG of a stratified random sample of 2000 of its members practicing in the United States. Of 1286 respondants, 48% considered themselves primary care providers. A recent study by Jacoby and colleagues¹⁴ used Medicare claims data to examine the scope of care that OBGs provided to their elderly patients. They found that OBGs provided a substantial amount of preventive care but not much nongynecologic care for elderly women.

We extend the work of these investigators by further exploring the breadth of medical conditions for which OBGs provide care to their elderly patients and by examining the degree to which OBGs in both rural and urban areas care for nongynecologic conditions. Our findings can offer some understanding of the potential impact of legislation to designate OBGs as primary care providers for elderly women.

Methods

Data Sources

We used the 1994 Washington State Medicare Part B claims file as the data source for our study. This database, part of the National Claims History File of the Health Care Financing Administration (HCFA), is an administrative data set that captures diagnostic and therapeutic information about services billed to Medicare. The Medicare Part B file contains a series of line items with each representing a discrete billable service for a Medicare beneficiary. These line items included identifiers for the patient receiving the service, the physician providing the service, the diagnosis coded, and the date of the visit. Items submitted by physicians include a unique physician identification number (UPIN) used to designate the specialty of the physician providing these services.

Physician Sample

All physicians practicing in the state of Washington and submitting Medicare claims for patient visits in 1994 were eligible for the study. We focused on the approximately 80% of OBGs who were participating in Medicare Part B. A variety of subspecialties including 5 medical (dermatology, cardiology, gastroenterology, pulmonology, and rheumatology) and 4 surgical (general surgery, orthopedic surgery, otolaryngology, and urology) were selected for comparison. Family practice and internal medicine were included in the descriptive analysis of visit frequency but not in the domain analysis, because all of primary care for elderly women was considered within the domain of these 2 specialties.

Information from the HCFA UPIN National Directory, the American Board of Medical Specialties (ABMS) database, and the American Medical Association (AMA) masterfile were used to link specialty information to the UPINs in the Part B Medicare file. Physicians were designated as a specific specialty type if both the ABMS certification and the primary self-designated specialty captured in the AMA masterfile were the same. In cases where these differed, the physicians were excluded. Including only those physicians who had the same specialty of training as their reported specialty of practice ensured accurate assignment of specialty.

Practice Location

Physicians were designated as practicing in rural or urban areas based on the ZIP codes of their practice addresses. ZIP codes were assigned as rural or urban on the basis of their proximity to a hospital classified as such by the Washington State Office of Rural Health of the Department of Health. The 5 physicians whose practice addresses were unknown or were in both rural and urban areas were excluded from the practice location analysis.

Patient Visits and Sample

We aggregated all outpatient physician services (eg, diagnostic tests and procedures) provided to an individual on a single date by the same provider into medical encounters. We used the current procedural terminology (CPT) and the HCFA common procedure coding system (HCPCS) for these services to determine whether they involved face-to-face contact with a physician. Those encounters that included such contact were considered visits. Within each visit, we chose one face-to-face line item—the index line—to identify the primary diagnosis for that visit. This index line either contained the evaluation and management code or, in cases without such a code, the face-to-face line item with the highest allowable charge. The International Classification of Diseases, Ninth Revision (ICD-9) diagnosis codes for these index lines were used to identify the primary diagnosis for each visit.¹⁵

We included women patients of the study physicians who were aged 65 years and older, enrolled in Medicare Part B, and alive throughout 1994. We excluded the 15% of patients enrolled in a managed care plan during the year to select elderly patients with unrestricted access to physicians in any specialty and because patient visit data were not available for those enrolled in those plans.

Designation of Specialty Domain

To interpret the claims-based diagnoses for each visit, we used diagnosis clusters to collapse the ICD-9 system into 120 groupings of individual diagnostic codes.^16,17 The codes in each group share similar pathophysiologic characteristics that often receive similar management. Two physicians reviewed the diagnosis clusters and nonclustered individual ICD-9 diagnoses to identify conditions traditionally cared for by physicians in each specialty. These diagnoses were designated as “in domain”; all others were designated as “out of domain.” Domain assignments were then reviewed and revised by the Medicare Carrier Advisory Committee, which was composed of specialists including all of those represented in our study.

For example, the diagnosis of peptic ulcer disease is considered out of domain for an OBG but in domain for a gastroenterologist, while cervical dysplasia is considered in domain for an OBG but out of domain for a gastroenterologist. We considered out of domain care as a measure of the breadth of care provided. The more out of domain care an OBG provided, the greater the degree of primary care that he or she was providing.

Analysis

We examined the distribution of out of domain visits for individual OBGs and the frequency of all diagnoses and out of domain diagnoses for OBGs in total. We also described visit volumes and percentage of out of domain visits in both rural and urban areas for OBGs and other physician specialists.

Results

Of the 328 Washington physicians in 1994 who submitted Medicare claims and were designated as OBGs by either the ABMS or the AMA masterfile, 285 were designated OBGs by both the ABMS and the AMA masterfile. These 285 OBGs treated 10,522 Medicare patients for 16,743 visits. These patients made a total of 108,720 visits to all physicians during the year. The patients of OBGs averaged 1.6 visits to them and 10.3 visits to any physician over the year. Patients who visited OBGs received 15.4% of their total health care from them.

The visit rate to OBGs (1.6) was the lowest for all the specialties studied; however, it closely resembled rates from other surgical specialties that ranged from 1.8 to 2.4 visits per patient. Medical specialists averaged 1.9 to 3.7 visits per patient, and traditional generalists ranged from 3.7 to 3.8 visits per patient. Of the specialties studied, the percentage of overall health care for patients of physicians of a certain specialty received by those physicians was the third lowest for OBGs (15.4%). This amount was similar to surgical specialists who ranged from 15.3% to 18.9%. Medical specialists ranged from 17.8% to 27.1%, and the patients who saw a family physician or general internist received more than 40% of their total health care from them during the year.

Of the 16,743 visits made to OBGs, 12.2% had diagnoses that were out of domain for the specialty Table 1. Among surgical specialists, general surgeons had the highest percentage of out of domain visits (21.9%), while others had a much lower proportion of out of domain visits (1.5% to 4.5%). Among medical specialists, pulmonologists had the highest percentage of out of domain visits (29.7%); others ranged from 1.2% to 15.0%.

Almost 12% of OBGs had more than 30% of their visits out of domain Figure 1. The median percentage of out of domain visits for individual OBGs was 6.7%. Table 1 shows that rural OBGs provided less out of domain care (8.2%) than their urban counterparts (12.9%). General surgeons were similar to OBGs in providing less out of domain care in the rural setting than urban setting. All the other specialties had the same or more out of domain care in the rural setting than in the urban setting.

Two of the 15 most frequent diagnosis clusters recorded by OBGs were for out of domain care Table 2. The general medical examination was the fifth most frequently reported diagnosis cluster and the most frequent out of domain diagnosis cluster Table 3. Hypertension was the other out of domain diagnosis cluster in the top 15 and comprised 9.0% of all of the out of domain diagnoses. The other out of domain problems diagnosed by OBGs include a full range of primary care conditions.

Discussion

Our study demonstrates that during 1994, the large majority of OBGs provided a limited amount of nongynecologic care to their elderly patients. This is consistent with the findings of other studies examining the scope of OBGs’ practices and suggests that in general OBGs were not serving as primary care providers to their elderly patients.^{11,13-15,17,18}

The visit rates to OBGs were more similar to surgical subspecialties than to the traditional primary care specialties. The number of visits per Medicare patient to OBGs was 1.6, the lowest of the specialties studied and very different from that of traditional primary care specialists (3.7 to 3.8 per patient). Patients of OBGs received 15.4% of their total health care from OBGs, which was much lower than the amount of overall health care that traditional generalists provided to their patients (42.0%-42.9%). This suggests that OBGs were primarily seeing elderly patients in consultation for gynecologic problems. This is consistent with a number of studies that have demonstrated that elderly women are less likely to receive care from OBGs.^10,12,19-22

Our finding that few OBGs provided broad-spectrum care in 1994 suggests that legislation to increase the use of OBGs as primary care providers could affect the way general medical services are delivered to elderly women. Many of these women may have nongynecologic medical conditions requiring treatment and monitoring. If most OBGs do not routinely provide these services, these women will require referral to medical specialists, which could lead to increasing costs, inconvenience, and fragmentation of care.

We did not expect the findings that rural OBGs provided less nongynecologic care than urban OBGs. Because rural areas are often underserved, we hypothesized that rural OBGs would practice as generalists more often and provide more general medical care. We concluded, therefore, that rural OBGs may be in shorter supply than generalists and that they are filling their practices with visits specific to their specialty. In addition, there may be increased competition among urban OBGs for gynecologic visits, so a larger number of patients are seen for nongynecologic problems.

Limitations

Our study has several limitations. We included only the primary diagnosis for each visit. A patient may have presented with both a gynecologic and non-gynecologic problem, but the OBG may have coded the gynecologic diagnosis as the primary one. In addition, patients presenting for a gynecologic complaint may inquire as an aside about a nongynecologic concern that may have been addressed but not coded. This would underestimate the amount of nongynecologic care provided by OBGs as well as the other studied specialists. However, it is unlikely that every time a woman visits an OBG for an upper respiratory infection, joint pain, or glycohemoglobin monitoring she also has an active gynecologic problem. The methodology we used should identify the portion of OBGs who provided a substantial amount of nongynecologic care.

The scope of OBGs’ practices may have changed since the data were collected. The primary reason for such a change would be the implementation of the primary care requirement during residency. Since that change in training occurred in 1996, however, those residents affected are just now entering the work force. Thus, our study’s data are likely to represent current practice patterns. A follow-up study of the scope of OBGs’ practices in 5 to 10 years will help elucidate the effect of the residency primary care requirement on OBGs’ practices.

The assignment of a diagnosis as in or out of domain is subject to interpretation. ICD-9 codes and diagnosis clusters were reviewed separately by 2 physicians to classify the diagnoses. When there was disagreement, the individuals discussed the diagnosis to reach consensus. If they could not agree, the diagnosis was considered out of domain. For example, the general medical examination was considered out of domain because the exact nature of the visit is unclear; however, the diagnosis may have been used for general gynecologic services provided to women without specific diagnoses, such as annual Papanicolaou tests that would be considered in domain. Since this was a conservative approach, it could overestimate the amount of out of domain care provided.

Only one feature of primary care—the breadth of practice—was addressed in this study. We did not examine a number of other features that also characterize primary care—continuity, coordination, and accessibility—as also described by the Committee on the Future of Primary Care of the IOM.⁹ In addition, we did not address quality of care.

Since we limited our study to Washington Medicare beneficiaries aged 65 years and older, we cannot comment on the degree to which OBGs may be providing general medical care to patients younger than 65 years. Younger patients may have different relationships with their physicians and present with different medical issues of varying complexity. We also excluded the 15% of Medicare elderly in Washington who in 1994 were enrolled in a managed care health plan. These results cannot be extrapolated to this population, because nearly all health maintenance organizations restrict access to specialists.

The Effect of Legislation

Our findings raise the question of whether legislation that designates OBGs as primary care providers for elderly women would result in an increase in the use of OBGs as providers of care for problems outside the reproductive system or primarily increase access to OBGs’ specialty services. Passage of legislation such as the Patients’ Bill of Rights Act of 1999 may facilitate elderly women’s obtaining primary gynecologic care, yet it may not have the same effect on their receipt of general medical care. A bill like this would allow women of all ages to designate OBGs as their primary care providers, thus allowing unrestricted and direct access to their services. Studies investigating the care received by elderly women enrolled in private health care plans in which they are able to select OBGs as primary care physicians could provide additional useful information. If the Patients’ Bill of Rights Act of 1999 or similar legislation is passed, studies will be needed to assess its effect on the overall medical care received by elderly women.

Conclusions

In 1994, few OBGs were providing the level of care that ACOG designates as extended primary care and that the IOM considers primary care. Our study does not reflect the degree to which they might have been providing primary preventive care and does not examine OBGs’ abilities to provide care for nongynecologic problems. The extent to which these findings represent current practice is also unknown. Nonetheless, our findings provide a baseline for the type of care provided by OBGs and suggest that without changes in the scope of OBGs’ practice, legislation resulting in more elderly women using OBGs as their primary providers could result in greater fragmentation and costs for their overall medical care.

Alternatively, some OBGs could embrace the movement within their specialty to emphasize treating patients’ general medical problems. OBGs develop close relationships with patients during their reproductive years, and these patients may benefit from a continuing relationship with these physicians as they age. There have been recent changes in obstetrics-gynecology residency requirements, increasing the time spent training in general medicine to better prepare residents to provide nongynecologic care. Other specialties have developed training tracks to specifically prepare physicians to practice primary care. Perhaps this is the time for residency programs in obstetrics-gynecology to do the same for a subset of their residents specifically interested in providing primary care.

Acknowledgments

Our research was supported by grants from the Robert Wood Johnson Foundation, Princeton, NJ, and the Office of Rural Health Policy and the Agency for Health Care Policy and Research of the US Public Health Service, Washington, DC. The views expressed in this article are those of the authors and do not necessarily represent those of the University of Washington, the Health Care Financing Administration, or the Robert Wood Johnson Foundation. The authors would like to acknowledge Peter Houck, MD, for his contributions to the manuscript and Durlin Hickok, MD, MPH, for reviewing the manuscript.