The Joint Commission (TJC) currently accredits approximately 4546 acute care, critical access, and specialty hospitals,1 accounting for approximately 82% of U.S. hospitals (representing 92% of hospital beds). Hospitals seeking to earn and maintain accreditation undergo unannounced on‐site visits by a team of Joint Commission surveyors at least once every 3 years. These surveys address a variety of domains, including the environment of care, infection prevention and control, information management, adherence to a series of national patient safety goals, and leadership.1

The survey process has changed markedly in recent years. Since 2002, accredited hospitals have been required to continuously collect and submit selected performance measure data to The Joint Commission throughout the three‐year accreditation cycle. The tracer methodology, an evaluation method in which surveyors select a patient to follow through the organization in order to assess compliance with selected standards, was instituted in 2004. Soon thereafter, on‐site surveys went from announced to unannounced in 2006.

Despite the 50+ year history of hospital accreditation in the United States, there has been surprisingly little research on the link between accreditation status and measures of hospital quality (both processes and outcomes). It is only recently that a growing number of studies have attempted to examine this relationship. Empirical support for the relationship between accreditation and other quality measures is emerging. Accredited hospitals have been shown to provide better emergency response planning2 and training3 compared to non‐accredited hospitals. Accreditation has been observed to be a key predictor of patient safety system implementation4 and the primary driver of hospitals' patient‐safety initiatives.5 Accredited trauma centers have been associated with significant reductions in patient mortality,6 and accreditation has been linked to better compliance with evidence‐based methadone and substance abuse treatment.7, 8 Accredited hospitals have been shown to perform better on measures of hospital quality in acute myocardial infarction (AMI), heart failure, and pneumonia care.9, 10 Similarly, accreditation has been associated with lower risk‐adjusted in‐hospital mortality rates for congestive heart failure (CHF), stroke, and pneumonia.11, 12 The results of such research, however, have not always been consistent. Several studies have been unable to demonstrate a relationship between accreditation and quality measures. A study of financial and cost‐related outcome measures found no relationship to accreditation,13 and a study comparing medication error rates across different types of organizations found no relationship to accreditation status.14 Similarly, a comparison of accredited versus non‐accredited ambulatory surgical organizations found that patients were less likely to be hospitalized when treated at an accredited facility for colonoscopy procedures, but no such relationship was observed for the other 4 procedures studied.15

While the research to date has been generally supportive of the link between accreditation and other measures of health care quality, the studies were typically limited to only a few measures and/or involved relatively small samples of accredited and non‐accredited organizations. Over the last decade, however, changes in the performance measurement landscape have created previously unavailable opportunities to more robustly examine the relationship between accreditation and other indicators of hospital quality.

At about the same time that The Joint Commission's accreditation process was becoming more vigorous, the Centers for Medicare and Medicaid Services (CMS) began a program of publicly reporting quality data (http://www.hospitalcompare.hhs.gov). The alignment of Joint Commission and CMS quality measures establishes a mechanism through which accredited and non‐accredited hospitals can be compared using the same nationally standardized quality measures. Therefore, we took advantage of this unique circumstancea new and more robust TJC accreditation program and the launching of public quality reportingto examine the relationship between Joint Commission accreditation status and publicly reported hospital quality measures. Moreover, by examining trends in these publicly reported measures over five years and incorporating performance data not found in the Hospital Compare Database, we assessed whether accreditation status was also linked to the pace of performance improvement over time.

By using a population of hospitals and a range of standardized quality measures greater than those used in previous studies, we seek to address the following questions: Is Joint Commission accreditation status truly associated with higher quality care? And does accreditation status help identify hospitals that are more likely to improve their quality and safety over time?

METHODS

RESULTS

For the 16 individual measures used in this study, a total of 4798 hospitals participated in Hospital Compare or reported data to The Joint Commission in 2004 or 2008. Of these, 907 were excluded because the performance data were not available for either 2004 (576 hospitals) or 2008 (331 hospitals) resulting in a missing value for the change in performance score. Therefore, 3891 hospitals (81%) were included in the final analyses. The 907 excluded hospitals were more likely to be rural (50.8% vs 17.5%), be critical access hospitals (53.9% vs 13.9%), have less than 100 beds (77.4% vs 37.6%), be government owned (34.6% vs 22.1%), be for profit (61.4% vs 49.5%), or be unaccredited (79.8% vs 45.8% in 2004; 75.6% vs 12.8% in 2008), compared with the included hospitals (P < 0.001 for all comparisons).

DISCUSSION

While accreditation has face validity and is desired by key stakeholders, it is expensive and time consuming. Stakeholders thus are justified in seeking evidence that accreditation is associated with better quality and safety. Ideally, not only would it be associated with better performance at a single point in time, it would also be associated with the pace of improvement over time.

Our study is the first, to our knowledge, to show the association of accreditation status with improvement in the trajectory of performance over a five‐year period. Taking advantage of the fact that the accreditation process changed substantially at about the same time that TJC and CMS began requiring public reporting of evidence‐based quality measures, we found that hospitals accredited by The Joint Commission had had larger improvements in hospital performance from 2004 to 2008 than non‐accredited hospitals, even though the former started with higher baseline performance levels. This accelerated improvement was broad‐based: Accredited hospitals were more likely to achieve superior performance (greater than 90% adherence to quality measures) in 2008 on 13 of 16 nationally standardized quality‐of‐care measures, three clinical area summary scores, and an overall score compared to hospitals that were not accredited. These results are consistent with other studies that have looked at both process and outcome measures and accreditation.912

It is important to note that the observed accreditation effect reflects a difference between hospitals that have elected to seek one particular self‐regulatory alternative to the more restrictive and extensive public regulatory or licensure requirements with those that have not.39 The non‐accredited hospitals that were included in this study are not considered to be sub‐standard hospitals. In fact, hospitals not accredited by The Joint Commission have also met the standards set by Medicare in the Conditions of Participation, and our study demonstrates that these hospitals achieved reasonably strong performance on publicly reported quality measures (86.8% adherence on the composite measure in 2008) and considerable improvement over the 5 years of public reporting (average improvement on composite measure from 2004 to 2008 of 11.8%). Moreover, there are many paths to improvement, and some non‐accredited hospitals achieve stellar performance on quality measures, perhaps by embracing other methods to catalyze improvement.

That said, our data demonstrate that, on average, accredited hospitals achieve superior performance on these evidence‐based quality measures, and their performance improved more strikingly over time. In interpreting these results, it is important to recognize that, while Joint Commission‐accredited hospitals must report quality data, performance on these measures is not directly factored into the accreditation decision; if this were not so, one could argue that this association is a statistical tautology. As it is, we believe that the 2 measures (accreditation and publicly reported quality measures) are two independent assessments of the quality of an organization, and, while the performance measures may not be a gold standard, a measure of their association does provide useful information about the degree to which accreditation is linked to organizational quality.

There are several potential limitations of the current study. First, while we adjusted for most of the known hospital demographic and organizational factors associated with performance, there may be unidentified factors that are associated with both accreditation and performance. This may not be relevant to a patient or payer choosing a hospital based on accreditation status (who may not care whether accreditation is simply associated with higher quality or actually helps produce such quality), but it is relevant to policy‐makers, who may weigh the value of embracing accreditation versus other maneuvers (such as pay for performance or new educational requirements) as a vehicle to promote high‐quality care.

A second limitation is that the specification of the measures can change over time due to the acquisition of new clinical knowledge, which makes longitudinal comparison and tracking of results over time difficult. There were two measures that had definitional changes that had noticeable impact on longitudinal trends: the AMI measure Primary Percutaneous Coronary Intervention (PCI) Received within 90 Minutes of Hospital Arrival (which in 2004 and 2005 used 120 minutes as the threshold), and the pneumonia measure Antibiotic Within 4 Hours of Arrival (which in 2007 changed the threshold to six hours). Other changes included adding angiotensin‐receptor blocker therapy (ARB) as an alternative to angiotensin‐converting enzyme inhibitor (ACEI) therapy in 2005 to the AMI and heart failure measures ACEI or ARB for left ventricular dysfunction. Other less significant changes have been made to the data collection methods for other measures, which could impact the interpretation of changes in performance over time. That said, these changes influenced both accredited and non‐accredited hospitals equally, and we cannot think of reasons that they would have created differential impacts.

Another limitation is that the 16 process measures provide a limited picture of hospital performance. Although the three conditions in the study account for over 15% of Medicare admissions,19 it is possible that non‐accredited hospitals performed as well as accredited hospitals on other measures of quality that were not captured by the 16 measures. As more standardized measures are added to The Joint Commission and CMS databases, it will be possible to use the same study methodology to incorporate these additional domains.

From the original cohort of 4798 hospitals reporting in 2004 or 2008, 19% were not included in the study due to missing data in either 2004 or 2008. Almost two‐thirds of the hospitals excluded from the study were missing 2004 data and, of these, 77% were critical access hospitals. The majority of these critical access hospitals (97%) were non‐accredited. This is in contrast to the hospitals missing 2008 data, of which only 13% were critical access. Since reporting of data to Hospital Compare was voluntary in 2004, it appears that critical access hospitals chose to wait later to report data to Hospital Compare, compared to acute care hospitals. Since critical access hospitals tended to have lower rates, smaller sample sizes, and be non‐accredited, the results of the study would be expected to slightly underestimate the difference between accredited and non‐accredited hospitals.

Finally, while we have argued that the publicly reported quality measures and TJC accreditation decisions provide different lenses into the quality of a given hospital, we cannot entirely exclude the possibility that there are subtle relationships between these two methods that might be partly responsible for our findings. For example, while performance measure rates do not factor directly into the accreditation decision, it is possible that Joint Commission surveyors may be influenced by their knowledge of these rates and biased in their scoring of unrelated standards during the survey process. While we cannot rule out such biases, we are aware of no research on the subject, and have no reason to believe that such biases may have confounded the analysis.

In summary, we found that Joint Commission‐accredited hospitals outperformed non‐accredited hospitals on nationally standardized quality measures of AMI, heart failure, and pneumonia. The performance gap between Joint Commission‐accredited and non‐accredited hospitals increased over the five years of the study. Future studies should incorporate more robust and varied measures of quality as outcomes, and seek to examine the nature of the observed relationship (ie, whether accreditation is simply a marker of higher quality and more rapid improvement, or the accreditation process actually helps create these salutary outcomes).

The Joint Commission (TJC) currently accredits approximately 4546 acute care, critical access, and specialty hospitals,1 accounting for approximately 82% of U.S. hospitals (representing 92% of hospital beds). Hospitals seeking to earn and maintain accreditation undergo unannounced on‐site visits by a team of Joint Commission surveyors at least once every 3 years. These surveys address a variety of domains, including the environment of care, infection prevention and control, information management, adherence to a series of national patient safety goals, and leadership.1

The survey process has changed markedly in recent years. Since 2002, accredited hospitals have been required to continuously collect and submit selected performance measure data to The Joint Commission throughout the three‐year accreditation cycle. The tracer methodology, an evaluation method in which surveyors select a patient to follow through the organization in order to assess compliance with selected standards, was instituted in 2004. Soon thereafter, on‐site surveys went from announced to unannounced in 2006.

Despite the 50+ year history of hospital accreditation in the United States, there has been surprisingly little research on the link between accreditation status and measures of hospital quality (both processes and outcomes). It is only recently that a growing number of studies have attempted to examine this relationship. Empirical support for the relationship between accreditation and other quality measures is emerging. Accredited hospitals have been shown to provide better emergency response planning2 and training3 compared to non‐accredited hospitals. Accreditation has been observed to be a key predictor of patient safety system implementation4 and the primary driver of hospitals' patient‐safety initiatives.5 Accredited trauma centers have been associated with significant reductions in patient mortality,6 and accreditation has been linked to better compliance with evidence‐based methadone and substance abuse treatment.7, 8 Accredited hospitals have been shown to perform better on measures of hospital quality in acute myocardial infarction (AMI), heart failure, and pneumonia care.9, 10 Similarly, accreditation has been associated with lower risk‐adjusted in‐hospital mortality rates for congestive heart failure (CHF), stroke, and pneumonia.11, 12 The results of such research, however, have not always been consistent. Several studies have been unable to demonstrate a relationship between accreditation and quality measures. A study of financial and cost‐related outcome measures found no relationship to accreditation,13 and a study comparing medication error rates across different types of organizations found no relationship to accreditation status.14 Similarly, a comparison of accredited versus non‐accredited ambulatory surgical organizations found that patients were less likely to be hospitalized when treated at an accredited facility for colonoscopy procedures, but no such relationship was observed for the other 4 procedures studied.15

While the research to date has been generally supportive of the link between accreditation and other measures of health care quality, the studies were typically limited to only a few measures and/or involved relatively small samples of accredited and non‐accredited organizations. Over the last decade, however, changes in the performance measurement landscape have created previously unavailable opportunities to more robustly examine the relationship between accreditation and other indicators of hospital quality.

At about the same time that The Joint Commission's accreditation process was becoming more vigorous, the Centers for Medicare and Medicaid Services (CMS) began a program of publicly reporting quality data (http://www.hospitalcompare.hhs.gov). The alignment of Joint Commission and CMS quality measures establishes a mechanism through which accredited and non‐accredited hospitals can be compared using the same nationally standardized quality measures. Therefore, we took advantage of this unique circumstancea new and more robust TJC accreditation program and the launching of public quality reportingto examine the relationship between Joint Commission accreditation status and publicly reported hospital quality measures. Moreover, by examining trends in these publicly reported measures over five years and incorporating performance data not found in the Hospital Compare Database, we assessed whether accreditation status was also linked to the pace of performance improvement over time.

By using a population of hospitals and a range of standardized quality measures greater than those used in previous studies, we seek to address the following questions: Is Joint Commission accreditation status truly associated with higher quality care? And does accreditation status help identify hospitals that are more likely to improve their quality and safety over time?

METHODS

RESULTS

For the 16 individual measures used in this study, a total of 4798 hospitals participated in Hospital Compare or reported data to The Joint Commission in 2004 or 2008. Of these, 907 were excluded because the performance data were not available for either 2004 (576 hospitals) or 2008 (331 hospitals) resulting in a missing value for the change in performance score. Therefore, 3891 hospitals (81%) were included in the final analyses. The 907 excluded hospitals were more likely to be rural (50.8% vs 17.5%), be critical access hospitals (53.9% vs 13.9%), have less than 100 beds (77.4% vs 37.6%), be government owned (34.6% vs 22.1%), be for profit (61.4% vs 49.5%), or be unaccredited (79.8% vs 45.8% in 2004; 75.6% vs 12.8% in 2008), compared with the included hospitals (P < 0.001 for all comparisons).

DISCUSSION

While accreditation has face validity and is desired by key stakeholders, it is expensive and time consuming. Stakeholders thus are justified in seeking evidence that accreditation is associated with better quality and safety. Ideally, not only would it be associated with better performance at a single point in time, it would also be associated with the pace of improvement over time.

Our study is the first, to our knowledge, to show the association of accreditation status with improvement in the trajectory of performance over a five‐year period. Taking advantage of the fact that the accreditation process changed substantially at about the same time that TJC and CMS began requiring public reporting of evidence‐based quality measures, we found that hospitals accredited by The Joint Commission had had larger improvements in hospital performance from 2004 to 2008 than non‐accredited hospitals, even though the former started with higher baseline performance levels. This accelerated improvement was broad‐based: Accredited hospitals were more likely to achieve superior performance (greater than 90% adherence to quality measures) in 2008 on 13 of 16 nationally standardized quality‐of‐care measures, three clinical area summary scores, and an overall score compared to hospitals that were not accredited. These results are consistent with other studies that have looked at both process and outcome measures and accreditation.912

It is important to note that the observed accreditation effect reflects a difference between hospitals that have elected to seek one particular self‐regulatory alternative to the more restrictive and extensive public regulatory or licensure requirements with those that have not.39 The non‐accredited hospitals that were included in this study are not considered to be sub‐standard hospitals. In fact, hospitals not accredited by The Joint Commission have also met the standards set by Medicare in the Conditions of Participation, and our study demonstrates that these hospitals achieved reasonably strong performance on publicly reported quality measures (86.8% adherence on the composite measure in 2008) and considerable improvement over the 5 years of public reporting (average improvement on composite measure from 2004 to 2008 of 11.8%). Moreover, there are many paths to improvement, and some non‐accredited hospitals achieve stellar performance on quality measures, perhaps by embracing other methods to catalyze improvement.

That said, our data demonstrate that, on average, accredited hospitals achieve superior performance on these evidence‐based quality measures, and their performance improved more strikingly over time. In interpreting these results, it is important to recognize that, while Joint Commission‐accredited hospitals must report quality data, performance on these measures is not directly factored into the accreditation decision; if this were not so, one could argue that this association is a statistical tautology. As it is, we believe that the 2 measures (accreditation and publicly reported quality measures) are two independent assessments of the quality of an organization, and, while the performance measures may not be a gold standard, a measure of their association does provide useful information about the degree to which accreditation is linked to organizational quality.

There are several potential limitations of the current study. First, while we adjusted for most of the known hospital demographic and organizational factors associated with performance, there may be unidentified factors that are associated with both accreditation and performance. This may not be relevant to a patient or payer choosing a hospital based on accreditation status (who may not care whether accreditation is simply associated with higher quality or actually helps produce such quality), but it is relevant to policy‐makers, who may weigh the value of embracing accreditation versus other maneuvers (such as pay for performance or new educational requirements) as a vehicle to promote high‐quality care.

A second limitation is that the specification of the measures can change over time due to the acquisition of new clinical knowledge, which makes longitudinal comparison and tracking of results over time difficult. There were two measures that had definitional changes that had noticeable impact on longitudinal trends: the AMI measure Primary Percutaneous Coronary Intervention (PCI) Received within 90 Minutes of Hospital Arrival (which in 2004 and 2005 used 120 minutes as the threshold), and the pneumonia measure Antibiotic Within 4 Hours of Arrival (which in 2007 changed the threshold to six hours). Other changes included adding angiotensin‐receptor blocker therapy (ARB) as an alternative to angiotensin‐converting enzyme inhibitor (ACEI) therapy in 2005 to the AMI and heart failure measures ACEI or ARB for left ventricular dysfunction. Other less significant changes have been made to the data collection methods for other measures, which could impact the interpretation of changes in performance over time. That said, these changes influenced both accredited and non‐accredited hospitals equally, and we cannot think of reasons that they would have created differential impacts.

Another limitation is that the 16 process measures provide a limited picture of hospital performance. Although the three conditions in the study account for over 15% of Medicare admissions,19 it is possible that non‐accredited hospitals performed as well as accredited hospitals on other measures of quality that were not captured by the 16 measures. As more standardized measures are added to The Joint Commission and CMS databases, it will be possible to use the same study methodology to incorporate these additional domains.

From the original cohort of 4798 hospitals reporting in 2004 or 2008, 19% were not included in the study due to missing data in either 2004 or 2008. Almost two‐thirds of the hospitals excluded from the study were missing 2004 data and, of these, 77% were critical access hospitals. The majority of these critical access hospitals (97%) were non‐accredited. This is in contrast to the hospitals missing 2008 data, of which only 13% were critical access. Since reporting of data to Hospital Compare was voluntary in 2004, it appears that critical access hospitals chose to wait later to report data to Hospital Compare, compared to acute care hospitals. Since critical access hospitals tended to have lower rates, smaller sample sizes, and be non‐accredited, the results of the study would be expected to slightly underestimate the difference between accredited and non‐accredited hospitals.

Finally, while we have argued that the publicly reported quality measures and TJC accreditation decisions provide different lenses into the quality of a given hospital, we cannot entirely exclude the possibility that there are subtle relationships between these two methods that might be partly responsible for our findings. For example, while performance measure rates do not factor directly into the accreditation decision, it is possible that Joint Commission surveyors may be influenced by their knowledge of these rates and biased in their scoring of unrelated standards during the survey process. While we cannot rule out such biases, we are aware of no research on the subject, and have no reason to believe that such biases may have confounded the analysis.

In summary, we found that Joint Commission‐accredited hospitals outperformed non‐accredited hospitals on nationally standardized quality measures of AMI, heart failure, and pneumonia. The performance gap between Joint Commission‐accredited and non‐accredited hospitals increased over the five years of the study. Future studies should incorporate more robust and varied measures of quality as outcomes, and seek to examine the nature of the observed relationship (ie, whether accreditation is simply a marker of higher quality and more rapid improvement, or the accreditation process actually helps create these salutary outcomes).

The Joint Commission (TJC) currently accredits approximately 4546 acute care, critical access, and specialty hospitals,1 accounting for approximately 82% of U.S. hospitals (representing 92% of hospital beds). Hospitals seeking to earn and maintain accreditation undergo unannounced on‐site visits by a team of Joint Commission surveyors at least once every 3 years. These surveys address a variety of domains, including the environment of care, infection prevention and control, information management, adherence to a series of national patient safety goals, and leadership.1

The survey process has changed markedly in recent years. Since 2002, accredited hospitals have been required to continuously collect and submit selected performance measure data to The Joint Commission throughout the three‐year accreditation cycle. The tracer methodology, an evaluation method in which surveyors select a patient to follow through the organization in order to assess compliance with selected standards, was instituted in 2004. Soon thereafter, on‐site surveys went from announced to unannounced in 2006.

Despite the 50+ year history of hospital accreditation in the United States, there has been surprisingly little research on the link between accreditation status and measures of hospital quality (both processes and outcomes). It is only recently that a growing number of studies have attempted to examine this relationship. Empirical support for the relationship between accreditation and other quality measures is emerging. Accredited hospitals have been shown to provide better emergency response planning2 and training3 compared to non‐accredited hospitals. Accreditation has been observed to be a key predictor of patient safety system implementation4 and the primary driver of hospitals' patient‐safety initiatives.5 Accredited trauma centers have been associated with significant reductions in patient mortality,6 and accreditation has been linked to better compliance with evidence‐based methadone and substance abuse treatment.7, 8 Accredited hospitals have been shown to perform better on measures of hospital quality in acute myocardial infarction (AMI), heart failure, and pneumonia care.9, 10 Similarly, accreditation has been associated with lower risk‐adjusted in‐hospital mortality rates for congestive heart failure (CHF), stroke, and pneumonia.11, 12 The results of such research, however, have not always been consistent. Several studies have been unable to demonstrate a relationship between accreditation and quality measures. A study of financial and cost‐related outcome measures found no relationship to accreditation,13 and a study comparing medication error rates across different types of organizations found no relationship to accreditation status.14 Similarly, a comparison of accredited versus non‐accredited ambulatory surgical organizations found that patients were less likely to be hospitalized when treated at an accredited facility for colonoscopy procedures, but no such relationship was observed for the other 4 procedures studied.15

While the research to date has been generally supportive of the link between accreditation and other measures of health care quality, the studies were typically limited to only a few measures and/or involved relatively small samples of accredited and non‐accredited organizations. Over the last decade, however, changes in the performance measurement landscape have created previously unavailable opportunities to more robustly examine the relationship between accreditation and other indicators of hospital quality.

At about the same time that The Joint Commission's accreditation process was becoming more vigorous, the Centers for Medicare and Medicaid Services (CMS) began a program of publicly reporting quality data (http://www.hospitalcompare.hhs.gov). The alignment of Joint Commission and CMS quality measures establishes a mechanism through which accredited and non‐accredited hospitals can be compared using the same nationally standardized quality measures. Therefore, we took advantage of this unique circumstancea new and more robust TJC accreditation program and the launching of public quality reportingto examine the relationship between Joint Commission accreditation status and publicly reported hospital quality measures. Moreover, by examining trends in these publicly reported measures over five years and incorporating performance data not found in the Hospital Compare Database, we assessed whether accreditation status was also linked to the pace of performance improvement over time.

By using a population of hospitals and a range of standardized quality measures greater than those used in previous studies, we seek to address the following questions: Is Joint Commission accreditation status truly associated with higher quality care? And does accreditation status help identify hospitals that are more likely to improve their quality and safety over time?

METHODS

RESULTS

For the 16 individual measures used in this study, a total of 4798 hospitals participated in Hospital Compare or reported data to The Joint Commission in 2004 or 2008. Of these, 907 were excluded because the performance data were not available for either 2004 (576 hospitals) or 2008 (331 hospitals) resulting in a missing value for the change in performance score. Therefore, 3891 hospitals (81%) were included in the final analyses. The 907 excluded hospitals were more likely to be rural (50.8% vs 17.5%), be critical access hospitals (53.9% vs 13.9%), have less than 100 beds (77.4% vs 37.6%), be government owned (34.6% vs 22.1%), be for profit (61.4% vs 49.5%), or be unaccredited (79.8% vs 45.8% in 2004; 75.6% vs 12.8% in 2008), compared with the included hospitals (P < 0.001 for all comparisons).

DISCUSSION

While accreditation has face validity and is desired by key stakeholders, it is expensive and time consuming. Stakeholders thus are justified in seeking evidence that accreditation is associated with better quality and safety. Ideally, not only would it be associated with better performance at a single point in time, it would also be associated with the pace of improvement over time.

Our study is the first, to our knowledge, to show the association of accreditation status with improvement in the trajectory of performance over a five‐year period. Taking advantage of the fact that the accreditation process changed substantially at about the same time that TJC and CMS began requiring public reporting of evidence‐based quality measures, we found that hospitals accredited by The Joint Commission had had larger improvements in hospital performance from 2004 to 2008 than non‐accredited hospitals, even though the former started with higher baseline performance levels. This accelerated improvement was broad‐based: Accredited hospitals were more likely to achieve superior performance (greater than 90% adherence to quality measures) in 2008 on 13 of 16 nationally standardized quality‐of‐care measures, three clinical area summary scores, and an overall score compared to hospitals that were not accredited. These results are consistent with other studies that have looked at both process and outcome measures and accreditation.912

It is important to note that the observed accreditation effect reflects a difference between hospitals that have elected to seek one particular self‐regulatory alternative to the more restrictive and extensive public regulatory or licensure requirements with those that have not.39 The non‐accredited hospitals that were included in this study are not considered to be sub‐standard hospitals. In fact, hospitals not accredited by The Joint Commission have also met the standards set by Medicare in the Conditions of Participation, and our study demonstrates that these hospitals achieved reasonably strong performance on publicly reported quality measures (86.8% adherence on the composite measure in 2008) and considerable improvement over the 5 years of public reporting (average improvement on composite measure from 2004 to 2008 of 11.8%). Moreover, there are many paths to improvement, and some non‐accredited hospitals achieve stellar performance on quality measures, perhaps by embracing other methods to catalyze improvement.

That said, our data demonstrate that, on average, accredited hospitals achieve superior performance on these evidence‐based quality measures, and their performance improved more strikingly over time. In interpreting these results, it is important to recognize that, while Joint Commission‐accredited hospitals must report quality data, performance on these measures is not directly factored into the accreditation decision; if this were not so, one could argue that this association is a statistical tautology. As it is, we believe that the 2 measures (accreditation and publicly reported quality measures) are two independent assessments of the quality of an organization, and, while the performance measures may not be a gold standard, a measure of their association does provide useful information about the degree to which accreditation is linked to organizational quality.

There are several potential limitations of the current study. First, while we adjusted for most of the known hospital demographic and organizational factors associated with performance, there may be unidentified factors that are associated with both accreditation and performance. This may not be relevant to a patient or payer choosing a hospital based on accreditation status (who may not care whether accreditation is simply associated with higher quality or actually helps produce such quality), but it is relevant to policy‐makers, who may weigh the value of embracing accreditation versus other maneuvers (such as pay for performance or new educational requirements) as a vehicle to promote high‐quality care.

A second limitation is that the specification of the measures can change over time due to the acquisition of new clinical knowledge, which makes longitudinal comparison and tracking of results over time difficult. There were two measures that had definitional changes that had noticeable impact on longitudinal trends: the AMI measure Primary Percutaneous Coronary Intervention (PCI) Received within 90 Minutes of Hospital Arrival (which in 2004 and 2005 used 120 minutes as the threshold), and the pneumonia measure Antibiotic Within 4 Hours of Arrival (which in 2007 changed the threshold to six hours). Other changes included adding angiotensin‐receptor blocker therapy (ARB) as an alternative to angiotensin‐converting enzyme inhibitor (ACEI) therapy in 2005 to the AMI and heart failure measures ACEI or ARB for left ventricular dysfunction. Other less significant changes have been made to the data collection methods for other measures, which could impact the interpretation of changes in performance over time. That said, these changes influenced both accredited and non‐accredited hospitals equally, and we cannot think of reasons that they would have created differential impacts.

Another limitation is that the 16 process measures provide a limited picture of hospital performance. Although the three conditions in the study account for over 15% of Medicare admissions,19 it is possible that non‐accredited hospitals performed as well as accredited hospitals on other measures of quality that were not captured by the 16 measures. As more standardized measures are added to The Joint Commission and CMS databases, it will be possible to use the same study methodology to incorporate these additional domains.

From the original cohort of 4798 hospitals reporting in 2004 or 2008, 19% were not included in the study due to missing data in either 2004 or 2008. Almost two‐thirds of the hospitals excluded from the study were missing 2004 data and, of these, 77% were critical access hospitals. The majority of these critical access hospitals (97%) were non‐accredited. This is in contrast to the hospitals missing 2008 data, of which only 13% were critical access. Since reporting of data to Hospital Compare was voluntary in 2004, it appears that critical access hospitals chose to wait later to report data to Hospital Compare, compared to acute care hospitals. Since critical access hospitals tended to have lower rates, smaller sample sizes, and be non‐accredited, the results of the study would be expected to slightly underestimate the difference between accredited and non‐accredited hospitals.

Finally, while we have argued that the publicly reported quality measures and TJC accreditation decisions provide different lenses into the quality of a given hospital, we cannot entirely exclude the possibility that there are subtle relationships between these two methods that might be partly responsible for our findings. For example, while performance measure rates do not factor directly into the accreditation decision, it is possible that Joint Commission surveyors may be influenced by their knowledge of these rates and biased in their scoring of unrelated standards during the survey process. While we cannot rule out such biases, we are aware of no research on the subject, and have no reason to believe that such biases may have confounded the analysis.

In summary, we found that Joint Commission‐accredited hospitals outperformed non‐accredited hospitals on nationally standardized quality measures of AMI, heart failure, and pneumonia. The performance gap between Joint Commission‐accredited and non‐accredited hospitals increased over the five years of the study. Future studies should incorporate more robust and varied measures of quality as outcomes, and seek to examine the nature of the observed relationship (ie, whether accreditation is simply a marker of higher quality and more rapid improvement, or the accreditation process actually helps create these salutary outcomes).

With 17 million Americans reporting heavy drinking (5 or more drinks on 5 different occasions in the last month) and 1.7 million hospital discharges in 2006 containing at least 1 alcohol‐related diagnosis, it would be hard to imagine a hospitalist who does not encounter patients with alcohol abuse.1, 2 Estimates from studies looking at the number of risky drinkers among medical inpatients vary widely2% to 60%with more detailed studies suggesting 17% to 25% prevalence.36 Yet despite the large numbers and great costs to the healthcare system, the inpatient treatment of alcohol withdrawal syndrome remains the ugly stepsister to more exciting topics, such as acute myocardial infarction, pulmonary embolism and procedures.7, 8 We hospitalists typically leave the clinical studies, research, and interest on substance abuse to addiction specialists and psychiatrists, perhaps due to our discomfort with these patients, negative attitudes, or belief that there is nothing new in the treatment of alcohol withdrawal syndrome since Dr Leo Henryk Sternbach discovered benzodiazepines in 1957.7, 9 Many of us just admit the alcoholic patient, check the alcohol‐pathway in our order entry system, and stop thinking about it.

But in this day of evidence‐based medicine and practice, what is the evidence behind the treatment of alcohol withdrawal, especially in relation to inpatient medicine? Shouldn't we hospitalists be thinking about this question? Hospitalists tend to see 2 types of inpatients with alcohol withdrawal: those solely admitted for withdrawal, and those admitted with active medical issues who then experience alcohol withdrawal. Is there a difference?

The Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM‐IV) defines early alcohol withdrawal as the first 48 hours where there is central nervous system (CNS) stimulation, adrenergic hyperactivity, and the risk of seizures. Late withdrawal, after 48 hours, includes delirium tremens (DTs) and Wernicke's encephalopathy.10 This is based on studies done in the 1950s, where researchers observed patients as they withdrew from alcohol and took notes.11, 12

The goal in treatment of alcohol withdrawal is to minimize symptoms and prevent seizures and DTs which, prior to benzodiazepines, had a mortality rate of 5% to 20%. Before the US Food and Drug Administration (FDA) approval of the first benzodiazepine in 1960 (chlordiazepoxide), physicians treated alcohol withdrawal with ethanol, antipsychotics, or paraldehyde.12 (That is why there is a P in the mnemonic MUDPILES for anion gap acidosis.) The first study to show a real benefit from benzodiazepine was published in 1969, when 537 men in a veterans detoxification unit were randomized to chlordiazepoxide (Librium), chlorpromazine (Thorazine), antihistamine, thiamine, or placebo.12 The primary outcome of DTs and seizures occurred in 10% to 16% of the patients, except for the chlordiazepoxide group where only 2% developed seizures and DTs (there was no P value calculated). Further studies published in the 1970s and early 1980s were too small to demonstrate a benefit. A 1997 meta‐analysis of all these studies, including the 1969 article,12 confirmed benzodiazepines statistically reduced seizures and DTs.13 Which benzodiazepine to use, however, is less clear. Long‐acting benzodiazepines with liver clearance (eg, chlordiazepoxide or diazepam) versus short‐acting with renal clearance (eg, oxazepam or lorazepam) is debated. While there are many strong opinions among clinicians, the same meta‐analysis did not find any difference between them, and a small 2009 study found no difference between a short‐acting and long‐acting benzodiazepine.13, 14

How much benzodiazepine to give and how frequently to dose it was looked at in 2 classic studies.15, 16 Both studies demonstrated that symptom‐triggered dosing of benzodiazepines, based on the Clinical Institute Withdrawal Assessment (CIWA) scale, performed equally well in terms of clinical outcomes, with less medication required as compared with fixed‐dose regimens. Based on these articles, many hospitals created alcohol pathways using solely symptom‐triggered dosing.

The CIWA scale is one of multiple rating scales in the assessment of alcohol withdrawal.17, 18 The CIWA‐Ar is a modified scale that was designed and validated for clinical use in inpatient detoxification centers, and excluded any active medical illness. It has gained popularity, though initial time for staff training and time for administration are limitations to its usefulness. Interestingly, vital signs, which many institutions use in their alcohol withdrawal pathways, were not strongly predictive in the CIWA study of severe withdrawal, seizures, or DTs.17

Finally, what about treatment when the patient does develop seizures or DTs? The evidence on how best to treat alcohol withdrawal seizures comes from a 1999 article which demonstrated a benefit of using lorazepam for recurrent seizures.19, 20 Unfortunately, the treatment for DTs is less clear. A 2004 meta‐analysis on the treatment of delirium tremens found benzodiazepines better than chlorpromazine (Thorazine), but benzodiazepines versus, or in addition to, newer antipsychotics have not been tested. The amount of benzodiazepine to give in DTs is only a Grade C (ie, expert opinion) recommendation: dose for light somnolence.21

All of these studies, however, come back to the basic question: Do they apply to the inpatients that hospitalists care for? A key factor to consider: All of the above‐mentioned studies, including the derivation and validation of the CIWA scale, were done in outpatient centers or inpatient detoxification centers. Patients with active medical illness or comorbidities were excluded. This data may be relevant for the patients admitted solely for alcohol withdrawal, but what about the 60 year old with diabetes, coronary artery disease, and chronic obstructive lung disease admitted for pneumonia who starts to withdraw; or the 72‐year‐old woman who breaks her hip and begins to withdraw on post‐op day 2?

There are 6 relatively recent studies that evaluate PRN (as needed) dosing of benzodiazepines on general medical inpatients.2227 While ideally these articles should apply to a hospitalist's patients, 2 of the studies excluded anyone with acute medical illness.24, 27 From the remaining 4, what do we learn? Weaver and colleagues did a randomized study on general medical patients and found less lorazepam was given with PRN versus fixed dosing.26 Unfortunately, the study was not blinded and there were statistically significant protocol errors. Comorbidity data was not given, leaving us to wonder to which inpatients this applies. Repper‐DeLisi et al. did a retrospective chart review, after implementing an alcohol pathway (not based on the CIWA scale), and did not find a statistical difference in dosing, length of stay, or delirium.25 Foy et al. looked at both medical and surgical patients, and dosed benzodiazepines based on an 18‐item CIWA scale which included vital signs.22 They found that the higher score did correlate with risk of developing severe alcohol withdrawal. However, the scale had limitations. Many patients with illness were at higher risk for severe alcohol withdrawal than their score indicated, and some high scores were believed, in part, due to illness. Jeager et al. did a pre‐comparison and post‐comparison of the implementation of a PRN CIWA protocol by chart review.23 They found a reduction in delirium in patients treated with PRN dosing, but no different in total benzodiazepine given. Because it was chart review, the authors acknowledge that defining delirium tremens was less reliable, and controlling for comorbidities was difficult. The difficult part of delirium in inpatients with alcohol abuse is that the delirium is not always just from DTs.

Two recent studies raised alarm about using a PRN CIWA pathway on patients.28, 29 A 2008 study found that 52% of patients were inappropriately put on a CIWA sliding scale when they either could not communicate or had not been recently drinking, or both.29 (The CIWA scale requires the person be able to answer symptom questions and is not applicable to non‐drinkers.) In 2005, during the implementation of an alcohol pathway at San Francisco General Hospital, an increase in mortality was noted with a PRN CIWA scale on inpatients.28

One of the conundrums for physicians is that whereas alcohol withdrawal has morbidity and mortality risks, benzodiazepine treatment itself has its own risks. Over sedation, respiratory depression, aspiration pneumonia, deconditioning from prolonged sedation, paradoxical agitation and disinhibition are the consequences of the dosing difficulties in alcohol withdrawal. Case reports on astronomical doses required to treat withdrawal (eg, 1600 mg of lorazepam in a day) raise questions of benzodiazepine resistance.30 Hence, multiple studies have been done to find alternatives for benzodiazepines. Our European counterparts lead the way in looking at: carbemazepine, gabapentin, gamma‐hydroxybuterate, corticotropin‐releasing hormone, baclofen, pregabalin, and phenobarbital. Again, the key issue for hospitalists: Are these benzodiazepine alternatives or additives applicable to our patients? These studies are done on outpatients with no concurrent medical illnesses. Yet, logic would suggest that it is the vulnerable hospitalized patients who might benefit the most from reducing the benzodiazepine amount using other agents.

In this issue of the Journal of Hospital Medicine, Lyon et al. provide a glimpse into possible ways to reduce the total benzodiazepine dose for general medical inpatients.31 They randomized inpatients withdrawing from alcohol to baclofen or placebo. Both groups still received PRN lorazepam based on their hospital's CIWA protocol. Prior outpatient studies have shown baclofen benefits patients undergoing alcohol withdrawal and the pathophysiology makes sense; baclofen acts on GABA b receptors. Lyon and collegaues' study results show significant reduction in the amount of benzodiazepine needed with no difference in CIWA scores.31

Is this a practice changer? Well, not yet. The numbers in the study are small and this is only 1 institution. These patients had only moderate alcohol withdrawal and the study was not powered to detect outcomes related to prevention of seizures and delirium tremens. However, the authors should be applauded for looking at alcohol withdrawal in medical inpatients.31 Trying to reduce the harm we cause with our benzodiazepine treatment regimens is a laudable goal. Inpatient alcohol withdrawal, especially for patients with medical comorbidities, is an area ripe for study and certainly deserves to have a spotlight shown on it.

Who better to do this than hospitalists? The Society of Hospital Medicine (SHM) core competency on Alcohol and Drug Withdrawal states, Hospitalists can lead their institutions in evidence based treatment protocols that improve care, reduce costs‐ and length of stay, and facilitate better overall outcomes in patients with substance related withdrawal syndromes.32 Hopefully, Lyon and collegaues' work will lead to the formation of multicenter hospitalist‐initiated studies to provide us with the best evidence for the treatment of inpatient alcohol withdrawal on our patients with comorbidities.31 Given the prevalence and potential severity of alcohol withdrawal in complex inpatients, isn't it time we really knew how to treat them?

With 17 million Americans reporting heavy drinking (5 or more drinks on 5 different occasions in the last month) and 1.7 million hospital discharges in 2006 containing at least 1 alcohol‐related diagnosis, it would be hard to imagine a hospitalist who does not encounter patients with alcohol abuse.1, 2 Estimates from studies looking at the number of risky drinkers among medical inpatients vary widely2% to 60%with more detailed studies suggesting 17% to 25% prevalence.36 Yet despite the large numbers and great costs to the healthcare system, the inpatient treatment of alcohol withdrawal syndrome remains the ugly stepsister to more exciting topics, such as acute myocardial infarction, pulmonary embolism and procedures.7, 8 We hospitalists typically leave the clinical studies, research, and interest on substance abuse to addiction specialists and psychiatrists, perhaps due to our discomfort with these patients, negative attitudes, or belief that there is nothing new in the treatment of alcohol withdrawal syndrome since Dr Leo Henryk Sternbach discovered benzodiazepines in 1957.7, 9 Many of us just admit the alcoholic patient, check the alcohol‐pathway in our order entry system, and stop thinking about it.

But in this day of evidence‐based medicine and practice, what is the evidence behind the treatment of alcohol withdrawal, especially in relation to inpatient medicine? Shouldn't we hospitalists be thinking about this question? Hospitalists tend to see 2 types of inpatients with alcohol withdrawal: those solely admitted for withdrawal, and those admitted with active medical issues who then experience alcohol withdrawal. Is there a difference?

The Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM‐IV) defines early alcohol withdrawal as the first 48 hours where there is central nervous system (CNS) stimulation, adrenergic hyperactivity, and the risk of seizures. Late withdrawal, after 48 hours, includes delirium tremens (DTs) and Wernicke's encephalopathy.10 This is based on studies done in the 1950s, where researchers observed patients as they withdrew from alcohol and took notes.11, 12

The goal in treatment of alcohol withdrawal is to minimize symptoms and prevent seizures and DTs which, prior to benzodiazepines, had a mortality rate of 5% to 20%. Before the US Food and Drug Administration (FDA) approval of the first benzodiazepine in 1960 (chlordiazepoxide), physicians treated alcohol withdrawal with ethanol, antipsychotics, or paraldehyde.12 (That is why there is a P in the mnemonic MUDPILES for anion gap acidosis.) The first study to show a real benefit from benzodiazepine was published in 1969, when 537 men in a veterans detoxification unit were randomized to chlordiazepoxide (Librium), chlorpromazine (Thorazine), antihistamine, thiamine, or placebo.12 The primary outcome of DTs and seizures occurred in 10% to 16% of the patients, except for the chlordiazepoxide group where only 2% developed seizures and DTs (there was no P value calculated). Further studies published in the 1970s and early 1980s were too small to demonstrate a benefit. A 1997 meta‐analysis of all these studies, including the 1969 article,12 confirmed benzodiazepines statistically reduced seizures and DTs.13 Which benzodiazepine to use, however, is less clear. Long‐acting benzodiazepines with liver clearance (eg, chlordiazepoxide or diazepam) versus short‐acting with renal clearance (eg, oxazepam or lorazepam) is debated. While there are many strong opinions among clinicians, the same meta‐analysis did not find any difference between them, and a small 2009 study found no difference between a short‐acting and long‐acting benzodiazepine.13, 14

How much benzodiazepine to give and how frequently to dose it was looked at in 2 classic studies.15, 16 Both studies demonstrated that symptom‐triggered dosing of benzodiazepines, based on the Clinical Institute Withdrawal Assessment (CIWA) scale, performed equally well in terms of clinical outcomes, with less medication required as compared with fixed‐dose regimens. Based on these articles, many hospitals created alcohol pathways using solely symptom‐triggered dosing.

The CIWA scale is one of multiple rating scales in the assessment of alcohol withdrawal.17, 18 The CIWA‐Ar is a modified scale that was designed and validated for clinical use in inpatient detoxification centers, and excluded any active medical illness. It has gained popularity, though initial time for staff training and time for administration are limitations to its usefulness. Interestingly, vital signs, which many institutions use in their alcohol withdrawal pathways, were not strongly predictive in the CIWA study of severe withdrawal, seizures, or DTs.17

Finally, what about treatment when the patient does develop seizures or DTs? The evidence on how best to treat alcohol withdrawal seizures comes from a 1999 article which demonstrated a benefit of using lorazepam for recurrent seizures.19, 20 Unfortunately, the treatment for DTs is less clear. A 2004 meta‐analysis on the treatment of delirium tremens found benzodiazepines better than chlorpromazine (Thorazine), but benzodiazepines versus, or in addition to, newer antipsychotics have not been tested. The amount of benzodiazepine to give in DTs is only a Grade C (ie, expert opinion) recommendation: dose for light somnolence.21

All of these studies, however, come back to the basic question: Do they apply to the inpatients that hospitalists care for? A key factor to consider: All of the above‐mentioned studies, including the derivation and validation of the CIWA scale, were done in outpatient centers or inpatient detoxification centers. Patients with active medical illness or comorbidities were excluded. This data may be relevant for the patients admitted solely for alcohol withdrawal, but what about the 60 year old with diabetes, coronary artery disease, and chronic obstructive lung disease admitted for pneumonia who starts to withdraw; or the 72‐year‐old woman who breaks her hip and begins to withdraw on post‐op day 2?

There are 6 relatively recent studies that evaluate PRN (as needed) dosing of benzodiazepines on general medical inpatients.2227 While ideally these articles should apply to a hospitalist's patients, 2 of the studies excluded anyone with acute medical illness.24, 27 From the remaining 4, what do we learn? Weaver and colleagues did a randomized study on general medical patients and found less lorazepam was given with PRN versus fixed dosing.26 Unfortunately, the study was not blinded and there were statistically significant protocol errors. Comorbidity data was not given, leaving us to wonder to which inpatients this applies. Repper‐DeLisi et al. did a retrospective chart review, after implementing an alcohol pathway (not based on the CIWA scale), and did not find a statistical difference in dosing, length of stay, or delirium.25 Foy et al. looked at both medical and surgical patients, and dosed benzodiazepines based on an 18‐item CIWA scale which included vital signs.22 They found that the higher score did correlate with risk of developing severe alcohol withdrawal. However, the scale had limitations. Many patients with illness were at higher risk for severe alcohol withdrawal than their score indicated, and some high scores were believed, in part, due to illness. Jeager et al. did a pre‐comparison and post‐comparison of the implementation of a PRN CIWA protocol by chart review.23 They found a reduction in delirium in patients treated with PRN dosing, but no different in total benzodiazepine given. Because it was chart review, the authors acknowledge that defining delirium tremens was less reliable, and controlling for comorbidities was difficult. The difficult part of delirium in inpatients with alcohol abuse is that the delirium is not always just from DTs.

Two recent studies raised alarm about using a PRN CIWA pathway on patients.28, 29 A 2008 study found that 52% of patients were inappropriately put on a CIWA sliding scale when they either could not communicate or had not been recently drinking, or both.29 (The CIWA scale requires the person be able to answer symptom questions and is not applicable to non‐drinkers.) In 2005, during the implementation of an alcohol pathway at San Francisco General Hospital, an increase in mortality was noted with a PRN CIWA scale on inpatients.28

One of the conundrums for physicians is that whereas alcohol withdrawal has morbidity and mortality risks, benzodiazepine treatment itself has its own risks. Over sedation, respiratory depression, aspiration pneumonia, deconditioning from prolonged sedation, paradoxical agitation and disinhibition are the consequences of the dosing difficulties in alcohol withdrawal. Case reports on astronomical doses required to treat withdrawal (eg, 1600 mg of lorazepam in a day) raise questions of benzodiazepine resistance.30 Hence, multiple studies have been done to find alternatives for benzodiazepines. Our European counterparts lead the way in looking at: carbemazepine, gabapentin, gamma‐hydroxybuterate, corticotropin‐releasing hormone, baclofen, pregabalin, and phenobarbital. Again, the key issue for hospitalists: Are these benzodiazepine alternatives or additives applicable to our patients? These studies are done on outpatients with no concurrent medical illnesses. Yet, logic would suggest that it is the vulnerable hospitalized patients who might benefit the most from reducing the benzodiazepine amount using other agents.

In this issue of the Journal of Hospital Medicine, Lyon et al. provide a glimpse into possible ways to reduce the total benzodiazepine dose for general medical inpatients.31 They randomized inpatients withdrawing from alcohol to baclofen or placebo. Both groups still received PRN lorazepam based on their hospital's CIWA protocol. Prior outpatient studies have shown baclofen benefits patients undergoing alcohol withdrawal and the pathophysiology makes sense; baclofen acts on GABA b receptors. Lyon and collegaues' study results show significant reduction in the amount of benzodiazepine needed with no difference in CIWA scores.31

Is this a practice changer? Well, not yet. The numbers in the study are small and this is only 1 institution. These patients had only moderate alcohol withdrawal and the study was not powered to detect outcomes related to prevention of seizures and delirium tremens. However, the authors should be applauded for looking at alcohol withdrawal in medical inpatients.31 Trying to reduce the harm we cause with our benzodiazepine treatment regimens is a laudable goal. Inpatient alcohol withdrawal, especially for patients with medical comorbidities, is an area ripe for study and certainly deserves to have a spotlight shown on it.

Who better to do this than hospitalists? The Society of Hospital Medicine (SHM) core competency on Alcohol and Drug Withdrawal states, Hospitalists can lead their institutions in evidence based treatment protocols that improve care, reduce costs‐ and length of stay, and facilitate better overall outcomes in patients with substance related withdrawal syndromes.32 Hopefully, Lyon and collegaues' work will lead to the formation of multicenter hospitalist‐initiated studies to provide us with the best evidence for the treatment of inpatient alcohol withdrawal on our patients with comorbidities.31 Given the prevalence and potential severity of alcohol withdrawal in complex inpatients, isn't it time we really knew how to treat them?

With 17 million Americans reporting heavy drinking (5 or more drinks on 5 different occasions in the last month) and 1.7 million hospital discharges in 2006 containing at least 1 alcohol‐related diagnosis, it would be hard to imagine a hospitalist who does not encounter patients with alcohol abuse.1, 2 Estimates from studies looking at the number of risky drinkers among medical inpatients vary widely2% to 60%with more detailed studies suggesting 17% to 25% prevalence.36 Yet despite the large numbers and great costs to the healthcare system, the inpatient treatment of alcohol withdrawal syndrome remains the ugly stepsister to more exciting topics, such as acute myocardial infarction, pulmonary embolism and procedures.7, 8 We hospitalists typically leave the clinical studies, research, and interest on substance abuse to addiction specialists and psychiatrists, perhaps due to our discomfort with these patients, negative attitudes, or belief that there is nothing new in the treatment of alcohol withdrawal syndrome since Dr Leo Henryk Sternbach discovered benzodiazepines in 1957.7, 9 Many of us just admit the alcoholic patient, check the alcohol‐pathway in our order entry system, and stop thinking about it.

But in this day of evidence‐based medicine and practice, what is the evidence behind the treatment of alcohol withdrawal, especially in relation to inpatient medicine? Shouldn't we hospitalists be thinking about this question? Hospitalists tend to see 2 types of inpatients with alcohol withdrawal: those solely admitted for withdrawal, and those admitted with active medical issues who then experience alcohol withdrawal. Is there a difference?

The Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM‐IV) defines early alcohol withdrawal as the first 48 hours where there is central nervous system (CNS) stimulation, adrenergic hyperactivity, and the risk of seizures. Late withdrawal, after 48 hours, includes delirium tremens (DTs) and Wernicke's encephalopathy.10 This is based on studies done in the 1950s, where researchers observed patients as they withdrew from alcohol and took notes.11, 12

The goal in treatment of alcohol withdrawal is to minimize symptoms and prevent seizures and DTs which, prior to benzodiazepines, had a mortality rate of 5% to 20%. Before the US Food and Drug Administration (FDA) approval of the first benzodiazepine in 1960 (chlordiazepoxide), physicians treated alcohol withdrawal with ethanol, antipsychotics, or paraldehyde.12 (That is why there is a P in the mnemonic MUDPILES for anion gap acidosis.) The first study to show a real benefit from benzodiazepine was published in 1969, when 537 men in a veterans detoxification unit were randomized to chlordiazepoxide (Librium), chlorpromazine (Thorazine), antihistamine, thiamine, or placebo.12 The primary outcome of DTs and seizures occurred in 10% to 16% of the patients, except for the chlordiazepoxide group where only 2% developed seizures and DTs (there was no P value calculated). Further studies published in the 1970s and early 1980s were too small to demonstrate a benefit. A 1997 meta‐analysis of all these studies, including the 1969 article,12 confirmed benzodiazepines statistically reduced seizures and DTs.13 Which benzodiazepine to use, however, is less clear. Long‐acting benzodiazepines with liver clearance (eg, chlordiazepoxide or diazepam) versus short‐acting with renal clearance (eg, oxazepam or lorazepam) is debated. While there are many strong opinions among clinicians, the same meta‐analysis did not find any difference between them, and a small 2009 study found no difference between a short‐acting and long‐acting benzodiazepine.13, 14

How much benzodiazepine to give and how frequently to dose it was looked at in 2 classic studies.15, 16 Both studies demonstrated that symptom‐triggered dosing of benzodiazepines, based on the Clinical Institute Withdrawal Assessment (CIWA) scale, performed equally well in terms of clinical outcomes, with less medication required as compared with fixed‐dose regimens. Based on these articles, many hospitals created alcohol pathways using solely symptom‐triggered dosing.

The CIWA scale is one of multiple rating scales in the assessment of alcohol withdrawal.17, 18 The CIWA‐Ar is a modified scale that was designed and validated for clinical use in inpatient detoxification centers, and excluded any active medical illness. It has gained popularity, though initial time for staff training and time for administration are limitations to its usefulness. Interestingly, vital signs, which many institutions use in their alcohol withdrawal pathways, were not strongly predictive in the CIWA study of severe withdrawal, seizures, or DTs.17

Finally, what about treatment when the patient does develop seizures or DTs? The evidence on how best to treat alcohol withdrawal seizures comes from a 1999 article which demonstrated a benefit of using lorazepam for recurrent seizures.19, 20 Unfortunately, the treatment for DTs is less clear. A 2004 meta‐analysis on the treatment of delirium tremens found benzodiazepines better than chlorpromazine (Thorazine), but benzodiazepines versus, or in addition to, newer antipsychotics have not been tested. The amount of benzodiazepine to give in DTs is only a Grade C (ie, expert opinion) recommendation: dose for light somnolence.21

All of these studies, however, come back to the basic question: Do they apply to the inpatients that hospitalists care for? A key factor to consider: All of the above‐mentioned studies, including the derivation and validation of the CIWA scale, were done in outpatient centers or inpatient detoxification centers. Patients with active medical illness or comorbidities were excluded. This data may be relevant for the patients admitted solely for alcohol withdrawal, but what about the 60 year old with diabetes, coronary artery disease, and chronic obstructive lung disease admitted for pneumonia who starts to withdraw; or the 72‐year‐old woman who breaks her hip and begins to withdraw on post‐op day 2?

There are 6 relatively recent studies that evaluate PRN (as needed) dosing of benzodiazepines on general medical inpatients.2227 While ideally these articles should apply to a hospitalist's patients, 2 of the studies excluded anyone with acute medical illness.24, 27 From the remaining 4, what do we learn? Weaver and colleagues did a randomized study on general medical patients and found less lorazepam was given with PRN versus fixed dosing.26 Unfortunately, the study was not blinded and there were statistically significant protocol errors. Comorbidity data was not given, leaving us to wonder to which inpatients this applies. Repper‐DeLisi et al. did a retrospective chart review, after implementing an alcohol pathway (not based on the CIWA scale), and did not find a statistical difference in dosing, length of stay, or delirium.25 Foy et al. looked at both medical and surgical patients, and dosed benzodiazepines based on an 18‐item CIWA scale which included vital signs.22 They found that the higher score did correlate with risk of developing severe alcohol withdrawal. However, the scale had limitations. Many patients with illness were at higher risk for severe alcohol withdrawal than their score indicated, and some high scores were believed, in part, due to illness. Jeager et al. did a pre‐comparison and post‐comparison of the implementation of a PRN CIWA protocol by chart review.23 They found a reduction in delirium in patients treated with PRN dosing, but no different in total benzodiazepine given. Because it was chart review, the authors acknowledge that defining delirium tremens was less reliable, and controlling for comorbidities was difficult. The difficult part of delirium in inpatients with alcohol abuse is that the delirium is not always just from DTs.

Two recent studies raised alarm about using a PRN CIWA pathway on patients.28, 29 A 2008 study found that 52% of patients were inappropriately put on a CIWA sliding scale when they either could not communicate or had not been recently drinking, or both.29 (The CIWA scale requires the person be able to answer symptom questions and is not applicable to non‐drinkers.) In 2005, during the implementation of an alcohol pathway at San Francisco General Hospital, an increase in mortality was noted with a PRN CIWA scale on inpatients.28

One of the conundrums for physicians is that whereas alcohol withdrawal has morbidity and mortality risks, benzodiazepine treatment itself has its own risks. Over sedation, respiratory depression, aspiration pneumonia, deconditioning from prolonged sedation, paradoxical agitation and disinhibition are the consequences of the dosing difficulties in alcohol withdrawal. Case reports on astronomical doses required to treat withdrawal (eg, 1600 mg of lorazepam in a day) raise questions of benzodiazepine resistance.30 Hence, multiple studies have been done to find alternatives for benzodiazepines. Our European counterparts lead the way in looking at: carbemazepine, gabapentin, gamma‐hydroxybuterate, corticotropin‐releasing hormone, baclofen, pregabalin, and phenobarbital. Again, the key issue for hospitalists: Are these benzodiazepine alternatives or additives applicable to our patients? These studies are done on outpatients with no concurrent medical illnesses. Yet, logic would suggest that it is the vulnerable hospitalized patients who might benefit the most from reducing the benzodiazepine amount using other agents.

In this issue of the Journal of Hospital Medicine, Lyon et al. provide a glimpse into possible ways to reduce the total benzodiazepine dose for general medical inpatients.31 They randomized inpatients withdrawing from alcohol to baclofen or placebo. Both groups still received PRN lorazepam based on their hospital's CIWA protocol. Prior outpatient studies have shown baclofen benefits patients undergoing alcohol withdrawal and the pathophysiology makes sense; baclofen acts on GABA b receptors. Lyon and collegaues' study results show significant reduction in the amount of benzodiazepine needed with no difference in CIWA scores.31

Is this a practice changer? Well, not yet. The numbers in the study are small and this is only 1 institution. These patients had only moderate alcohol withdrawal and the study was not powered to detect outcomes related to prevention of seizures and delirium tremens. However, the authors should be applauded for looking at alcohol withdrawal in medical inpatients.31 Trying to reduce the harm we cause with our benzodiazepine treatment regimens is a laudable goal. Inpatient alcohol withdrawal, especially for patients with medical comorbidities, is an area ripe for study and certainly deserves to have a spotlight shown on it.

Who better to do this than hospitalists? The Society of Hospital Medicine (SHM) core competency on Alcohol and Drug Withdrawal states, Hospitalists can lead their institutions in evidence based treatment protocols that improve care, reduce costs‐ and length of stay, and facilitate better overall outcomes in patients with substance related withdrawal syndromes.32 Hopefully, Lyon and collegaues' work will lead to the formation of multicenter hospitalist‐initiated studies to provide us with the best evidence for the treatment of inpatient alcohol withdrawal on our patients with comorbidities.31 Given the prevalence and potential severity of alcohol withdrawal in complex inpatients, isn't it time we really knew how to treat them?

Continuity of care is considered by many physicians to be of critical importance in providing high‐quality patient care. Most of the research to date has focused on continuity in outpatient primary care. Research on outpatient continuity of care has been facilitated by the fact that a number of measurement tools for outpatient continuity exist.1 Outpatient continuity of care has been linked to better quality of life scores,2 lower costs,3 and less emergency room use.4 As hospital medicine has taken on more and more of the responsibility of inpatient care, primary care doctors have voiced concerns about the impact of hospitalists on overall continuity of care5 and the quality of the doctorpatient relationship.6

Recently, continuity of care in the hospital setting has also received attention. When the Accreditation Council for Graduate Medical Education (ACGME) first proposed restrictions to resident duty hours, the importance of continuity of inpatient care began to be debated in earnest in large part because of the increase in hand‐offs which accompanies discontinuity.7, 8 A recent study of hospitalist communication documented that as many as 13% of hand‐offs at the time of service changes are judged as incomplete by the receiving physician. These incomplete hand‐offs were more likely to be associated with uncertainty regarding the plan of care, as well as perceived near misses or adverse events.9 In addition, several case reports and studies suggest that systems with less continuity may have poorer outcomes.7, 1015

Continuity in the hospital setting is likely to be important for several reasons. First, the acuity of a patient's problem during a hospitalization is likely greater than during an outpatient visit. Thus the complexity of information to be transferred between physicians during a hospital stay is correspondingly greater. Second, the diagnostic uncertainty surrounding many admissions leads to complex thought processes that may be difficult to recreate when handing off patient care to another physician. Finally, knowledge of a patient's hospital course and the likely trajectory of care is facilitated by firsthand knowledge of where the patient has been. All this information can be difficult to distill into a brief sign‐out to another physician who assumes care of the patient.

In the current study, we sought to examine the trends over time in continuity of inpatient care. We chose patients likely to be cared for by general internists: those hospitalized for chronic obstructive pulmonary disease (COPD), pneumonia, and congestive heart failure (CHF). The general internists caring for patients in the hospital could be the patient's primary care physician (PCP), a physician covering for the patient's PCP, a physician assigned at admission by the hospital, or a hospitalist. Our goals were to describe the current level of continuity of care in the hospital setting, to examine whether continuity has changed over time, and to determine factors affecting continuity of care.

Methods

We used a 5% national sample of claims data from Medicare beneficiaries for the years 19962006.16 This included Medicare enrollment files, Medicare Provider Analysis and Review (MEDPAR) files, Medicare Carrier files, and Provider of Services (POS) files.17, 18

Results

Between 1996 and 2006, 528,453 patients hospitalized for COPD, pneumonia, and CHF received care by a generalist physician during their hospital stay. Of these, 64.3% were seen by one generalist physician, 26.9% by two generalist physicians, and 8.8% by three or more generalist physicians during hospitalization.

Figure 1 shows the percentage of all patients seen by 1, 2, and 3 or more generalist physicians between 1996 and 2006. The percentage of patients receiving care from one generalist physician declined from 70.7% in 1996 to 59.4% in 2006 (P < 0.001). During the same period, the percentage of patients receiving care from 3 or more generalist physicians increased from 6.5% to 10.7% (P < 0.001). Similar trends were seen for each of the 3 conditions. There was a decrease in overall length of stay during this period, from a mean of 5.7 to 4.9 days (P < 0.001). The increase in the number of generalist physicians providing care during the hospital stay did not correspond to an increase in total number of visits during the hospitalization. The average number of daily visits from a generalist physician was 0.94 (0.30) in 1996 and 0.96 (0.35) in 2006.

Figure 1

Table 1 presents the percentage of patients receiving care from 1, 2, and 3 or more generalist physicians during hospitalization stratified by patient and hospital characteristics. Older adults, females, non‐Hispanic whites, those with higher socioeconomic status, and those with more comorbidities were more likely to receive care by multiple generalist physicians. There was also large variation by geographic region, metropolitan area size, and hospital characteristics. All of these differences were significant at the P < 0.0001 level.

Table 2 presents the results of a multivariable analysis of factors independently associated with experiencing continuity of care. In this analysis, continuity of care was defined as receiving inpatient care from one generalist physician (vs two or more). In the unadjusted models, the odds of experiencing continuity of care decreased by 5.5% per year from 1996 through 2006, and this decrease did not substantially change after adjusting for all other variables (4.8% yearly decrease). Younger patients, females, black patients, and those with low socioeconomic status were slightly more likely to experience continuity of care. As expected, patients admitted on weekends, emergency admissions, and those with intensive care unit (ICU) stays were less likely to experience continuity. There were marked geographic variations in continuity, with continuity approximately half as likely in New England as in the South. Continuity was greatest in smaller metropolitan areas versus rural and large metropolitan areas. Hospital size and teaching status produced only minor variation.

In Table 2 we also show that patients with an established PCP and those who received care from a hospitalist in the hospital were substantially less likely to experience continuity of care. There are several possible interpretations for that finding. For example, it might be that patients admitted to a hospitalist service were likely to see multiple hospitalists. Alternatively, the decreased continuity associated with hospitalists could reflect the fact that some patients cared for predominantly by non‐hospitalists may have seen a hospitalist on call for a sudden change in health status. To further explore these possible explanatory pathways, we constructed three new cohorts: 1) patients receiving all their care from non‐hospitalists, 2) patients receiving all their care from hospitalists, and 3) patients seen by both. As shown in Table 3, in patients seen by non‐hospitalists only, the mean number of generalist physicians seen during hospitalization was slightly greater than in patients cared for only by hospitalists.

We also tested for interactions in Table 2 between admission year and other factors. There was a significant interaction between admission year and having an identifiable PCP in the year prior to admission (Table 2). The odds of experiencing continuity of care decreased more rapidly for patients who did not have a PCP (5.5% per year; 95% CI: 5.2%5.8%) than for those who had one (4.3% per year; 95% CI: 4.1%4.6%).

Discussion

We conducted this study to better understand the degree to which hospitalized patients experience discontinuity of care within a hospital stay and to determine which patients are most likely to experience discontinuity. In our study, we specifically chose admission conditions that would likely be followed primarily by generalist physicians. We found that, over the past decade, discontinuity of care for hospitalized patients has increased substantially, as indicated by the proportion of patients taken care of by more than one generalist physician during a single hospital stay. This occurred even though overall length of stay was decreasing in this same period.

It is perhaps not surprising that inpatient continuity of care has been decreasing in the past 10 years. Outpatient practices are becoming busier, and more doctors are practicing in large group practices, which could lead to several different physicians in the same practice rounding on a hospitalized patient. We have previously demonstrated that hospitalists are caring for an increasing number of patients over this same time period,21 so another possibility is that hospitalist services are being used more often because of this heavy outpatient workload. Our analyses allowed us to test the hypothesis that having hospitalists involved in patient care increases discontinuity.

At first glance, it appears that being cared for by hospitalists may result in worse continuity of care. However, closer scrutiny of the data reveals that the discontinuity ascribed to the hospitalists in the multivariable model appears to be an artifact of defining the hospitalist variable as having been seen by any hospitalist during the hospital stay. This would include patients who saw a hospitalist in addition to their PCP or another non‐hospitalist generalist. When we compared hospitalist‐only care to other generalist care, we could not detect a difference in discontinuity. We know that generalist visits per day to patients has not substantially increased over time, so this discontinuity trend is not explained by having visits by both a hospitalist and the PCP. Therefore, this combination of findings suggests that the increased discontinuity associated with having a hospitalist involved in patient care is likely the result of system issues rather than hospitalist care per se. In fact, patients seem to experience slightly better continuity when they see only hospitalists as opposed to only non‐hospitalists.

What types of systems issues might lead to this finding? Generalists in most settings could choose to involve a hospitalist at any point in the patient's hospital stay. This could occur because of a change in patient acuity requiring the involvement of hospitalists who are present in the hospital more. It is also possible that hospitalists' schedules are created to maximize inpatient continuity of care with individual hospitalists. Even though hospitalists clearly work shifts, the 7 on, 7 off model22 likely results in patients seeing the same physician each day until the switch day. This is in contrast to outpatient primary care doctors whose concentration may be on maintaining continuity within their practice.

As the field of hospital medicine was emerging, many internal medicine physicians from various specialties were concerned about the impact of hospitalists on patient care. In one study, 73% of internal medicine physicians who were not hospitalists thought that hospitalists would worsen continuity of care.23 Primary care and subspecialist internal medicine physicians also expressed the concern that hospitalists could hurt their own relationships with patients,6 presumably because of lost continuity between the inpatient and outpatient settings. However, this fear seems to diminish once hospitalist programs are implemented and primary care doctors have experience with them.23 Our study suggests that the decrease in continuity that has occurred since these studies were published is not likely due to the emergence of hospital medicine, but rather due to other factors that influence who cares for hospitalized patients.

This study had some limitations. Length of stay is an obvious mediator of number of generalist physicians seen. Therefore, the sickest patients are likely to have both a long length of stay and low continuity. We adjusted for this in the multivariable modeling. In addition, given that this study used a large database, certain details are not discernable. For example, we chose to operationalize discontinuity as visits from multiple generalists during a single hospital stay. That is not a perfect definition, but it does represent multiple physicians directing the care of a patient. Importantly, this does not appear to represent continuity with one physician with extra visits from another, as the total number of generalist visits per day did not change over time. It is also possible that patients in the non‐hospitalist group saw physicians only from a single practice, but those details are not included in the database. Finally, we cannot tell what type of hand‐offs were occurring for individual patients during each hospital stay. Despite these disadvantages, using a large database like this one allows for detection of fairly small differences that could still be clinically important.

In summary, hospitalized patients appear to experience less continuity now than 10 years ago. However, the hospitalist model does not appear to play a role in this discontinuity. It is worth exploring in more detail why patients would see both hospitalists and other generalists. This pattern is not surprising, but may have some repercussions in terms of increasing the number of hand‐offs experienced by patients. These could lead to problems with patient safety and quality of care. Future work should explore the reasons for this discontinuity and look at the relationship between inpatient discontinuity outcomes such as quality of care and the doctorpatient relationship.

Continuity of care is considered by many physicians to be of critical importance in providing high‐quality patient care. Most of the research to date has focused on continuity in outpatient primary care. Research on outpatient continuity of care has been facilitated by the fact that a number of measurement tools for outpatient continuity exist.1 Outpatient continuity of care has been linked to better quality of life scores,2 lower costs,3 and less emergency room use.4 As hospital medicine has taken on more and more of the responsibility of inpatient care, primary care doctors have voiced concerns about the impact of hospitalists on overall continuity of care5 and the quality of the doctorpatient relationship.6

Recently, continuity of care in the hospital setting has also received attention. When the Accreditation Council for Graduate Medical Education (ACGME) first proposed restrictions to resident duty hours, the importance of continuity of inpatient care began to be debated in earnest in large part because of the increase in hand‐offs which accompanies discontinuity.7, 8 A recent study of hospitalist communication documented that as many as 13% of hand‐offs at the time of service changes are judged as incomplete by the receiving physician. These incomplete hand‐offs were more likely to be associated with uncertainty regarding the plan of care, as well as perceived near misses or adverse events.9 In addition, several case reports and studies suggest that systems with less continuity may have poorer outcomes.7, 1015

Continuity in the hospital setting is likely to be important for several reasons. First, the acuity of a patient's problem during a hospitalization is likely greater than during an outpatient visit. Thus the complexity of information to be transferred between physicians during a hospital stay is correspondingly greater. Second, the diagnostic uncertainty surrounding many admissions leads to complex thought processes that may be difficult to recreate when handing off patient care to another physician. Finally, knowledge of a patient's hospital course and the likely trajectory of care is facilitated by firsthand knowledge of where the patient has been. All this information can be difficult to distill into a brief sign‐out to another physician who assumes care of the patient.

In the current study, we sought to examine the trends over time in continuity of inpatient care. We chose patients likely to be cared for by general internists: those hospitalized for chronic obstructive pulmonary disease (COPD), pneumonia, and congestive heart failure (CHF). The general internists caring for patients in the hospital could be the patient's primary care physician (PCP), a physician covering for the patient's PCP, a physician assigned at admission by the hospital, or a hospitalist. Our goals were to describe the current level of continuity of care in the hospital setting, to examine whether continuity has changed over time, and to determine factors affecting continuity of care.

Methods

We used a 5% national sample of claims data from Medicare beneficiaries for the years 19962006.16 This included Medicare enrollment files, Medicare Provider Analysis and Review (MEDPAR) files, Medicare Carrier files, and Provider of Services (POS) files.17, 18

Results

Between 1996 and 2006, 528,453 patients hospitalized for COPD, pneumonia, and CHF received care by a generalist physician during their hospital stay. Of these, 64.3% were seen by one generalist physician, 26.9% by two generalist physicians, and 8.8% by three or more generalist physicians during hospitalization.

Figure 1 shows the percentage of all patients seen by 1, 2, and 3 or more generalist physicians between 1996 and 2006. The percentage of patients receiving care from one generalist physician declined from 70.7% in 1996 to 59.4% in 2006 (P < 0.001). During the same period, the percentage of patients receiving care from 3 or more generalist physicians increased from 6.5% to 10.7% (P < 0.001). Similar trends were seen for each of the 3 conditions. There was a decrease in overall length of stay during this period, from a mean of 5.7 to 4.9 days (P < 0.001). The increase in the number of generalist physicians providing care during the hospital stay did not correspond to an increase in total number of visits during the hospitalization. The average number of daily visits from a generalist physician was 0.94 (0.30) in 1996 and 0.96 (0.35) in 2006.

Figure 1

Table 1 presents the percentage of patients receiving care from 1, 2, and 3 or more generalist physicians during hospitalization stratified by patient and hospital characteristics. Older adults, females, non‐Hispanic whites, those with higher socioeconomic status, and those with more comorbidities were more likely to receive care by multiple generalist physicians. There was also large variation by geographic region, metropolitan area size, and hospital characteristics. All of these differences were significant at the P < 0.0001 level.

Table 2 presents the results of a multivariable analysis of factors independently associated with experiencing continuity of care. In this analysis, continuity of care was defined as receiving inpatient care from one generalist physician (vs two or more). In the unadjusted models, the odds of experiencing continuity of care decreased by 5.5% per year from 1996 through 2006, and this decrease did not substantially change after adjusting for all other variables (4.8% yearly decrease). Younger patients, females, black patients, and those with low socioeconomic status were slightly more likely to experience continuity of care. As expected, patients admitted on weekends, emergency admissions, and those with intensive care unit (ICU) stays were less likely to experience continuity. There were marked geographic variations in continuity, with continuity approximately half as likely in New England as in the South. Continuity was greatest in smaller metropolitan areas versus rural and large metropolitan areas. Hospital size and teaching status produced only minor variation.

In Table 2 we also show that patients with an established PCP and those who received care from a hospitalist in the hospital were substantially less likely to experience continuity of care. There are several possible interpretations for that finding. For example, it might be that patients admitted to a hospitalist service were likely to see multiple hospitalists. Alternatively, the decreased continuity associated with hospitalists could reflect the fact that some patients cared for predominantly by non‐hospitalists may have seen a hospitalist on call for a sudden change in health status. To further explore these possible explanatory pathways, we constructed three new cohorts: 1) patients receiving all their care from non‐hospitalists, 2) patients receiving all their care from hospitalists, and 3) patients seen by both. As shown in Table 3, in patients seen by non‐hospitalists only, the mean number of generalist physicians seen during hospitalization was slightly greater than in patients cared for only by hospitalists.

We also tested for interactions in Table 2 between admission year and other factors. There was a significant interaction between admission year and having an identifiable PCP in the year prior to admission (Table 2). The odds of experiencing continuity of care decreased more rapidly for patients who did not have a PCP (5.5% per year; 95% CI: 5.2%5.8%) than for those who had one (4.3% per year; 95% CI: 4.1%4.6%).

Discussion

We conducted this study to better understand the degree to which hospitalized patients experience discontinuity of care within a hospital stay and to determine which patients are most likely to experience discontinuity. In our study, we specifically chose admission conditions that would likely be followed primarily by generalist physicians. We found that, over the past decade, discontinuity of care for hospitalized patients has increased substantially, as indicated by the proportion of patients taken care of by more than one generalist physician during a single hospital stay. This occurred even though overall length of stay was decreasing in this same period.

It is perhaps not surprising that inpatient continuity of care has been decreasing in the past 10 years. Outpatient practices are becoming busier, and more doctors are practicing in large group practices, which could lead to several different physicians in the same practice rounding on a hospitalized patient. We have previously demonstrated that hospitalists are caring for an increasing number of patients over this same time period,21 so another possibility is that hospitalist services are being used more often because of this heavy outpatient workload. Our analyses allowed us to test the hypothesis that having hospitalists involved in patient care increases discontinuity.

At first glance, it appears that being cared for by hospitalists may result in worse continuity of care. However, closer scrutiny of the data reveals that the discontinuity ascribed to the hospitalists in the multivariable model appears to be an artifact of defining the hospitalist variable as having been seen by any hospitalist during the hospital stay. This would include patients who saw a hospitalist in addition to their PCP or another non‐hospitalist generalist. When we compared hospitalist‐only care to other generalist care, we could not detect a difference in discontinuity. We know that generalist visits per day to patients has not substantially increased over time, so this discontinuity trend is not explained by having visits by both a hospitalist and the PCP. Therefore, this combination of findings suggests that the increased discontinuity associated with having a hospitalist involved in patient care is likely the result of system issues rather than hospitalist care per se. In fact, patients seem to experience slightly better continuity when they see only hospitalists as opposed to only non‐hospitalists.

What types of systems issues might lead to this finding? Generalists in most settings could choose to involve a hospitalist at any point in the patient's hospital stay. This could occur because of a change in patient acuity requiring the involvement of hospitalists who are present in the hospital more. It is also possible that hospitalists' schedules are created to maximize inpatient continuity of care with individual hospitalists. Even though hospitalists clearly work shifts, the 7 on, 7 off model22 likely results in patients seeing the same physician each day until the switch day. This is in contrast to outpatient primary care doctors whose concentration may be on maintaining continuity within their practice.

As the field of hospital medicine was emerging, many internal medicine physicians from various specialties were concerned about the impact of hospitalists on patient care. In one study, 73% of internal medicine physicians who were not hospitalists thought that hospitalists would worsen continuity of care.23 Primary care and subspecialist internal medicine physicians also expressed the concern that hospitalists could hurt their own relationships with patients,6 presumably because of lost continuity between the inpatient and outpatient settings. However, this fear seems to diminish once hospitalist programs are implemented and primary care doctors have experience with them.23 Our study suggests that the decrease in continuity that has occurred since these studies were published is not likely due to the emergence of hospital medicine, but rather due to other factors that influence who cares for hospitalized patients.

This study had some limitations. Length of stay is an obvious mediator of number of generalist physicians seen. Therefore, the sickest patients are likely to have both a long length of stay and low continuity. We adjusted for this in the multivariable modeling. In addition, given that this study used a large database, certain details are not discernable. For example, we chose to operationalize discontinuity as visits from multiple generalists during a single hospital stay. That is not a perfect definition, but it does represent multiple physicians directing the care of a patient. Importantly, this does not appear to represent continuity with one physician with extra visits from another, as the total number of generalist visits per day did not change over time. It is also possible that patients in the non‐hospitalist group saw physicians only from a single practice, but those details are not included in the database. Finally, we cannot tell what type of hand‐offs were occurring for individual patients during each hospital stay. Despite these disadvantages, using a large database like this one allows for detection of fairly small differences that could still be clinically important.

In summary, hospitalized patients appear to experience less continuity now than 10 years ago. However, the hospitalist model does not appear to play a role in this discontinuity. It is worth exploring in more detail why patients would see both hospitalists and other generalists. This pattern is not surprising, but may have some repercussions in terms of increasing the number of hand‐offs experienced by patients. These could lead to problems with patient safety and quality of care. Future work should explore the reasons for this discontinuity and look at the relationship between inpatient discontinuity outcomes such as quality of care and the doctorpatient relationship.

Continuity of care is considered by many physicians to be of critical importance in providing high‐quality patient care. Most of the research to date has focused on continuity in outpatient primary care. Research on outpatient continuity of care has been facilitated by the fact that a number of measurement tools for outpatient continuity exist.1 Outpatient continuity of care has been linked to better quality of life scores,2 lower costs,3 and less emergency room use.4 As hospital medicine has taken on more and more of the responsibility of inpatient care, primary care doctors have voiced concerns about the impact of hospitalists on overall continuity of care5 and the quality of the doctorpatient relationship.6

Recently, continuity of care in the hospital setting has also received attention. When the Accreditation Council for Graduate Medical Education (ACGME) first proposed restrictions to resident duty hours, the importance of continuity of inpatient care began to be debated in earnest in large part because of the increase in hand‐offs which accompanies discontinuity.7, 8 A recent study of hospitalist communication documented that as many as 13% of hand‐offs at the time of service changes are judged as incomplete by the receiving physician. These incomplete hand‐offs were more likely to be associated with uncertainty regarding the plan of care, as well as perceived near misses or adverse events.9 In addition, several case reports and studies suggest that systems with less continuity may have poorer outcomes.7, 1015

Continuity in the hospital setting is likely to be important for several reasons. First, the acuity of a patient's problem during a hospitalization is likely greater than during an outpatient visit. Thus the complexity of information to be transferred between physicians during a hospital stay is correspondingly greater. Second, the diagnostic uncertainty surrounding many admissions leads to complex thought processes that may be difficult to recreate when handing off patient care to another physician. Finally, knowledge of a patient's hospital course and the likely trajectory of care is facilitated by firsthand knowledge of where the patient has been. All this information can be difficult to distill into a brief sign‐out to another physician who assumes care of the patient.

In the current study, we sought to examine the trends over time in continuity of inpatient care. We chose patients likely to be cared for by general internists: those hospitalized for chronic obstructive pulmonary disease (COPD), pneumonia, and congestive heart failure (CHF). The general internists caring for patients in the hospital could be the patient's primary care physician (PCP), a physician covering for the patient's PCP, a physician assigned at admission by the hospital, or a hospitalist. Our goals were to describe the current level of continuity of care in the hospital setting, to examine whether continuity has changed over time, and to determine factors affecting continuity of care.

Methods

We used a 5% national sample of claims data from Medicare beneficiaries for the years 19962006.16 This included Medicare enrollment files, Medicare Provider Analysis and Review (MEDPAR) files, Medicare Carrier files, and Provider of Services (POS) files.17, 18

Results

Between 1996 and 2006, 528,453 patients hospitalized for COPD, pneumonia, and CHF received care by a generalist physician during their hospital stay. Of these, 64.3% were seen by one generalist physician, 26.9% by two generalist physicians, and 8.8% by three or more generalist physicians during hospitalization.

Figure 1 shows the percentage of all patients seen by 1, 2, and 3 or more generalist physicians between 1996 and 2006. The percentage of patients receiving care from one generalist physician declined from 70.7% in 1996 to 59.4% in 2006 (P < 0.001). During the same period, the percentage of patients receiving care from 3 or more generalist physicians increased from 6.5% to 10.7% (P < 0.001). Similar trends were seen for each of the 3 conditions. There was a decrease in overall length of stay during this period, from a mean of 5.7 to 4.9 days (P < 0.001). The increase in the number of generalist physicians providing care during the hospital stay did not correspond to an increase in total number of visits during the hospitalization. The average number of daily visits from a generalist physician was 0.94 (0.30) in 1996 and 0.96 (0.35) in 2006.

Figure 1

Table 1 presents the percentage of patients receiving care from 1, 2, and 3 or more generalist physicians during hospitalization stratified by patient and hospital characteristics. Older adults, females, non‐Hispanic whites, those with higher socioeconomic status, and those with more comorbidities were more likely to receive care by multiple generalist physicians. There was also large variation by geographic region, metropolitan area size, and hospital characteristics. All of these differences were significant at the P < 0.0001 level.