Predictors of Follow-up of Atypical and ASCUS Papanicolaou Test Results in a High-Risk Population

Approximately 50 million Papanicolaou (Pap) tests are performed annually in the United States, of which approximately 5% will require further evaluation because of an abnormality.¹ The Bethesda System for reporting cervical/vaginal cytologic diagnoses,² developed in 1988 and revised in 1991, provides uniform diagnostic terminology to improve consistency of reporting. The Bethesda System introduced new terminology for atypical cytology, creating the label “atypical squamous cells of uncertain significance” (ASCUS). Pap test results labelled ASCUS encompass a spectrum of cellular change reflecting a variety of pathologic processes that cannot be more specifically categorized, including reactive changes, inflammation, human papillomavirus (HPV) related changes, suggested dysplasia, and less than optimal slide preparation.² Because studies have consistently shown a small but real risk that an ASCUS Pap result represents an underlying high-grade lesion,^3-6 there has been much debate about their management. Uncertainty about optimal management led to the development of clinical practice guidelines (CPG).^7-9

The National Cancer Institute⁷ (NCI) and a Canadian expert pane¹⁸ suggest that ASCUS and low-grade squamous intraepithelial lesions can be managed either by immediate colposcopy or by repeating cervical cytology. Clinicians may, if they choose, reserve referral to colposcopy for those patients with persistent abnormality on subsequent cytology. This conservative strategy is based on several assumptions:¹⁰ (1) conditions that mimic dysplasia, including reactive changes, are common; (2) spontaneous regression occurs in at least 50% of low-grade lesions; (3) the rate of progression is relatively slow; (4) the chance of carcinoma in situ with an ASCUS Pap test result is low; and (5) persistently negative follow-up Pap results are unlikely to be false negatives. The current practice guideline suggests that if clinicians opt for cytologic follow-up of ASCUS, repeat Pap tests should be done every 4 to 6 months until 3 consecutive normal test results have been obtained. Colposcopy should be performed if there is a subsequent abnormal Pap test result.⁷ The 1996 statement of the American College of Obstetricians and Gynecologists is similar, with the addition that colposcopy be performed for women with a single ASCUS Pap result if they also have any of the following high-risk factors: HPV infection, human immunodeficiency virus (HIV) infection, smoking, or multiple sexual partners.⁹

A disproportionate share of the cervical cancer burden in this country, in terms of both incidence and mortality, is shouldered by low-income minority women.^11-13 Minority and low-income populations have the highest rates of nonadherence with repeat Pap test or colposcopy.^14-18 Thus, the women who are at the highest risk of cervical intraepithelial neoplasia (CIN) may also be the most at risk for suboptimal management after an abnormal Pap test result is obtained. This project assessed current management of ASCUS Pap results in a diverse community at high risk of cervical cancer by comparing actual practice to the NCI guideline. In addition to assessing levels of follow-up in this community, we sought to identify predictors of adherence to that guideline.

Methods

Setting and Population

Data were collected in 7 community health centers serving a low-income, predominantly minority population. The centers represent diversity with regard to practice type (medicine/pediatrics/obstetrics model or family medicine model) and size (20,000-80,000 visits/year). The subjects were women with newly identified atypia or ASCUS Pap test results in 1996; women for whom the index Pap test represented persistent atypia or ASCUS were excluded, as were women with a history of CIN or squamous intraepithelial lesion (SIL). Those who were known to be infected with HIV represent a special category^19,20 that may have been managed differently at the clinician’s discretion; therefore, they were also excluded from the analysis.

Data Collection

After approval by the appropriate institutional review boards, women with atypical results in 1996 were identified by 2 mechanisms: site logs maintained for clinical tracking, and because all sites exclusively use the same laboratory, the central cytology database. Pap test results described as atypia or ASCUS were included, as these represent different terminology for the same spectrum of cytologic abnormalities with continued use of the older term by some cytopathologists. Chart abstraction by a trained research assistant collected information regarding: (1) complicating medical conditions (immunosuppression, history of cancer, substance abuse, pregnancy); (2) history of cervical dysplasia; (3) primary care provider and number of visits to site; (4) the provider’s management plan; (5) all subsequent Pap test and colposcopy results; (6) evidence of communication of the management plan to the subject; and (7) demographics.

Accuracy of chart abstraction was confirmed by the lead author rescreening 20% of the records. The data were entered into a Paradox database (Borland Software Corporation, Scotts Valley, Calif), and statistical analysis was performed using SAS software (SAS Institute, Inc, Cary, NC) and Stata software (Stata Corporation, College Station, Tex).

Analysis

The goal of our analysis was to determine the factors associated with follow-up. We defined 3 mutually exclusive levels of adherence to the CPG: (1) complete adherence (colposcopy done if recommended or 3 repeat tests approximately every 6 months until 3 are normal or colposcopy if there is persistent abnormality); (2) moderate adherence (1 or 2 follow-up tests after the index Pap test); and (3) low adherence (no follow-up cytology or histology, or abnormality on repeat Pap test without colposcopy.)

Univariate analyses were performed relating the level of follow-up to the description of the abnormality (atypia vs ASCUS), patient demographics, and practice and provider characteristics (eg, training, availability of colposcopy on site, and plan of care). In analyses where cell size fell below 5, Fisher exact test or Yates correction (for analyses with row or column size greater than 2) replaced the Pearson chi-square. We performed multivariate analysis using a hierarchical model with patients nested within sites, and ordinal logistic regression to account for the 3 ordered levels of adherence to the CPG. Variables significant at P less than .2 in the univariate analyses were included in the multivariate model to ensure inclusion of potentially important variables. Also, insurance status was included, because others have found significant associations for this variable.^21-23 The provider’s management plan was not included as a variable in the regression model, because it is not sufficiently independent of the adherence levels.

Because subjects were clustered within 7 sites (not a simple random sample), we treated site as a random effect with patients nested within the site. An ordered logistic regression model was fit by maximum likelihood estimation, using the program GLLAMM⁶ in Stata version 6.0 The fit of the ordered logistic regression model to the data was tested using the method of Harrell and colleagues.²⁴ Briefly, this method compares observed mean values of the model’s linear combination of the predictor variables to expected values under the assumption that the model (and in particular the proportional odds ratio assumption) is correctly specified.

Results

of the 570 women identified with an atypical or ASCUS Pap test result in 1996, 554 charts were located and reviewed. The charts of 16 individuals could not be located for review. Some women (n=69) identified had a previous atypical or ASCUS Pap test result. The sample of women with no previous ASCUS or atypical result, excluding those with documented HIV infection (n=19) and those with a history of CIN or SIL (n=79), consisted of 387 subjects. The mean age of the subjects at the time of the index Pap test was 32 years (standard deviation=11.7). More than half of the subjects (58.4%) were Hispanic. The majority of subjects had Medicaid (73.1%). The majority of index Pap result abnormalities were described as atypia (61.5%), with the remainder ASCUS (38.5%). A small percentage of women (8.8%) were pregnant at the time of the index Pap test. The subjects were approximately evenly split between receiving care in medicine/pediatrics/obstetrics model sites and family medicine model sites. Colposcopy was provided on site (at least 2 times per week) in 4 of the centers that served 68.7% of the subjects.

Overall, many subjects remained in care in the health centers for a relatively short time. Almost one fourth of the subjects (23.7%) had a final visit to the health center less than 6 months from the index Pap test. Complete data including 2 full years of follow-up from the date of the index Pap test were available for 41% of subjects. Subjects with new atypia had a small chance of SIL on subsequent tests (6.5%, increasing to 10.9% for those subjects with persistent atypia). Results of biopsies for women with new and persistent atypia are presented in Table 1. The majority of subjects had normal or low-grade findings on histology, and no cancer was diagnosed.

Management of atypical results is depicted in the Figure 1. Providers were credited with a “correct” plan if they recommended either a repeat test in 6 months or less, or colposcopy. “Incorrect” management included any other recommendation or failure to document any plan (18.9%). Providers were most likely to recommend a repeat test in 3 months (31.8%) or 6 months (46.1%). We considered communication to subjects “complete” if the notes reflected discussion of the result at a visit (58.3%) or if a completed telephone call was documented (3.9%). Communication was considered attempted if a letter was sent (47.2%) or a call attempted (8.8%) and documented. In 8.3% of charts no documentation of attempted or completed communication was present. There were no significant associations of provider plan or communication of results with demographic variables, pregnancy, or abnormality type. Family physicians were more likely than midlevel providers or other physicians to generate a correct plan (87.3%, 80.8%, 69.8%, respectively; P=.002). Family physicians documented completed communication of results in 77.8% of cases, compared with 70.1% for other physicians and 57.4% for midlevel providers.

Providers were more likely to recommend colposcopy when the index Pap test result was reported as ASCUS rather than atypia (22.1% vs 11.4%, P=.016). Providers were much more likely to choose colposcopy if the pathologist made a recommendation for colposcopy rather than a repeat test (85.2% vs 8.9%, P=.001) and if the report included “favored SIL” in the result (75.0% vs 11.5%, P=.001.) When colposcopy was recommended, 70.8% completed the procedure. After a first atypical Pap test, 36% had no follow-up test; 27% had 1 follow-up test; 20% had 2 tests; and 17% had 3 subsequent tests. Subjects were found to have complete, moderate, and low adherence levels as follows: 26.9%, 27.6%, and 45.5%.

Univariate associations with levels of adherence are shown in Table 2. Complete adherence was more likely when the result was discussed at a visit, when the provider’s plan was documented, when the provider was other than a family physician, and if the plan was colposcopy as opposed to repeat cytology. Complete adherence was associated with remaining in care at the centers after 1997. It was more likely for results reported as ASCUS as opposed to atypia. Large differences are seen in the number of women with complete adherence by site, with the percentage ranging from 7% to 57%.

In the ordered logistic regression analysis, 4 factors were found to be predictive of the quality of adherence: the description of the abnormality (atypia vs ASCUS), availability of colposcopy on-site, discussion of the plan at a visit, and site. The detailed results are shown in Table 3. The adjusted odds ratio for complete versus moderate and low adherence (or equivalently moderate and complete vs low) associated with having an ASCUS lesion versus atypia is 1.56; for attending a clinic where colposcopy is referred off-site, 0.39; and for having discussed the follow-up plan, 3.44. The effects of ethnicity, type of health insurance, provider training, and pregnancy were not statistically significant.

Discussion

The introduction of the ASCUS label in the early 1990s created controversy and led to guidelines from the NCI7 and others^8,9 based primarily on expert opinion. Data from a large randomized clinical trial25 currently underway may resolve some of the questions about the effectiveness of repeat cytology versus immediate colposcopy in the management of low-grade dysplasia. Importantly, lack of evidence for the guideline may adversely affect compliance. Implementation of either strategy is difficult, especially in communities with many risk factors for poor follow-up. Our results indicate low adherence to the NCI interim guideline in an urban primary care setting serving predominately low-income minority women. Of concern is that almost half our sample, compared with 30% of subjects in a Canadian study,²⁶ had no follow-up cytology or histology. Our findings highlight some of the difficulties encountered in managing low-grade abnormalities in a setting challenged by high provider turnover, staffing difficulties, incorrect phone and address information, and financial and language barriers. Long-term repeated cytology is difficult to accomplish where loss to follow-up is common and competing demands of managing multiple chronic illnesses may interfere. Our findings are similar to those of other investigators^27,28 in that colposcopy, when recommended, was not completed by approximately one third of the women.

Our findings highlight the importance of strict use of Bethesda terminology by cytopathologists, since the older terminology may be misunderstood and less aggressively followed by primary care clinicians. We chose to include women with either atypical or ASCUS results, because they are different labels for the same spectrum of abnormalities. We found that providers were slightly more likely to recommend colposcopy for ASCUS and ultimately that women with atypical results were more likely to have low adherence to the CPG. This probably reflects that many primary care providers do not view the guideline as applicable to results reported as atypia or communicate with patients about atypical results differently.

There were substantial differences in rates of follow-up by site that are likely related to several factors. Although tracking of abnormal Pap test results was required at all sites the intensity varied, with the most successful site staffed by registered nurses who worked closely with the colposcopy providers in maintaining a manual card file. Sites where colposcopy was recommended more often had overall higher proportions of women with complete adherence to the CPG than those sites where serial cytology was the follow-up strategy of choice. Family physicians were more likely to generate a correct plan yet had worse overall adherence. This is likely due to a preference to follow up women by serial cytology, which proves difficult in settings where patients often stay in care for less than 2 years. However, the finding persists even after controlling for the provider plan, suggesting that other factors are also contributing to account for the differences between sites. Importantly, providers were more likely to opt for colposcopy, and women were more likely to complete it if the service was available in the subject’s health center.

We are unable to determine all of the causes of inadequate follow-up. Systemic factors, however, are clearly implicated by the present data, as indicated by the substantial variation by site and the number of charts in which a clear provider plan was not indicated or where no indication that the subject had been informed of the result was documented. When documentation was present that results were discussed with patients during visits patients were much more likely to have good follow-up, again suggesting that good communication of results is imperative. Our results also suggest that reducing barriers to colposcopy, by such means as providing the service on-site, may be effective in achieving optimal follow-up when colposcopy is recommended. Recommending colposcopy for all women with an initial ASCUS or atypical Pap test result will result in significant stress on available colposcopic resources and expose large numbers of women to an expensive, uncomfortable, and distressing procedure unnecessarily. However, providers need to carefully consider the risk of loss to follow-up in making a recommendation, take extra effort to insure that women are informed of and understand their Pap test result, and establish clinical tracking efforts to meet the challenge of serial cytology in high-risk settings.

Limitations

Our findings are limited in that the standard of care to which we compared practice is based on a guideline that is not evidence based. Other limitations of the project are largely due to constraints imposed by the abstraction of retrospective data from primary care charts. Information about how women were informed of results is sketchy, and failure to document may result in underestimation of actual patient contact. The extent to which women had follow-up if they no longer received care in the health centers cannot be estimated. Some women may have had follow-up with other providers outside our system. Demographic information was obtained from site registration databases rather than self-report and may include some misclassification. Some important information is not captured consistently in the site’s computers, including changes in insurance and primary care providers. Provider turnover was extremely high in these health centers during the interval studied, though lists of providers are not available to create a precise measure to include in our model.

Acknowledgments

This work was supported by a clinical research training grant from the American Cancer Society.

Approximately 50 million Papanicolaou (Pap) tests are performed annually in the United States, of which approximately 5% will require further evaluation because of an abnormality.¹ The Bethesda System for reporting cervical/vaginal cytologic diagnoses,² developed in 1988 and revised in 1991, provides uniform diagnostic terminology to improve consistency of reporting. The Bethesda System introduced new terminology for atypical cytology, creating the label “atypical squamous cells of uncertain significance” (ASCUS). Pap test results labelled ASCUS encompass a spectrum of cellular change reflecting a variety of pathologic processes that cannot be more specifically categorized, including reactive changes, inflammation, human papillomavirus (HPV) related changes, suggested dysplasia, and less than optimal slide preparation.² Because studies have consistently shown a small but real risk that an ASCUS Pap result represents an underlying high-grade lesion,^3-6 there has been much debate about their management. Uncertainty about optimal management led to the development of clinical practice guidelines (CPG).^7-9

The National Cancer Institute⁷ (NCI) and a Canadian expert pane¹⁸ suggest that ASCUS and low-grade squamous intraepithelial lesions can be managed either by immediate colposcopy or by repeating cervical cytology. Clinicians may, if they choose, reserve referral to colposcopy for those patients with persistent abnormality on subsequent cytology. This conservative strategy is based on several assumptions:¹⁰ (1) conditions that mimic dysplasia, including reactive changes, are common; (2) spontaneous regression occurs in at least 50% of low-grade lesions; (3) the rate of progression is relatively slow; (4) the chance of carcinoma in situ with an ASCUS Pap test result is low; and (5) persistently negative follow-up Pap results are unlikely to be false negatives. The current practice guideline suggests that if clinicians opt for cytologic follow-up of ASCUS, repeat Pap tests should be done every 4 to 6 months until 3 consecutive normal test results have been obtained. Colposcopy should be performed if there is a subsequent abnormal Pap test result.⁷ The 1996 statement of the American College of Obstetricians and Gynecologists is similar, with the addition that colposcopy be performed for women with a single ASCUS Pap result if they also have any of the following high-risk factors: HPV infection, human immunodeficiency virus (HIV) infection, smoking, or multiple sexual partners.⁹

A disproportionate share of the cervical cancer burden in this country, in terms of both incidence and mortality, is shouldered by low-income minority women.^11-13 Minority and low-income populations have the highest rates of nonadherence with repeat Pap test or colposcopy.^14-18 Thus, the women who are at the highest risk of cervical intraepithelial neoplasia (CIN) may also be the most at risk for suboptimal management after an abnormal Pap test result is obtained. This project assessed current management of ASCUS Pap results in a diverse community at high risk of cervical cancer by comparing actual practice to the NCI guideline. In addition to assessing levels of follow-up in this community, we sought to identify predictors of adherence to that guideline.

Methods

Setting and Population

Data were collected in 7 community health centers serving a low-income, predominantly minority population. The centers represent diversity with regard to practice type (medicine/pediatrics/obstetrics model or family medicine model) and size (20,000-80,000 visits/year). The subjects were women with newly identified atypia or ASCUS Pap test results in 1996; women for whom the index Pap test represented persistent atypia or ASCUS were excluded, as were women with a history of CIN or squamous intraepithelial lesion (SIL). Those who were known to be infected with HIV represent a special category^19,20 that may have been managed differently at the clinician’s discretion; therefore, they were also excluded from the analysis.

Data Collection

After approval by the appropriate institutional review boards, women with atypical results in 1996 were identified by 2 mechanisms: site logs maintained for clinical tracking, and because all sites exclusively use the same laboratory, the central cytology database. Pap test results described as atypia or ASCUS were included, as these represent different terminology for the same spectrum of cytologic abnormalities with continued use of the older term by some cytopathologists. Chart abstraction by a trained research assistant collected information regarding: (1) complicating medical conditions (immunosuppression, history of cancer, substance abuse, pregnancy); (2) history of cervical dysplasia; (3) primary care provider and number of visits to site; (4) the provider’s management plan; (5) all subsequent Pap test and colposcopy results; (6) evidence of communication of the management plan to the subject; and (7) demographics.

Accuracy of chart abstraction was confirmed by the lead author rescreening 20% of the records. The data were entered into a Paradox database (Borland Software Corporation, Scotts Valley, Calif), and statistical analysis was performed using SAS software (SAS Institute, Inc, Cary, NC) and Stata software (Stata Corporation, College Station, Tex).

Analysis

The goal of our analysis was to determine the factors associated with follow-up. We defined 3 mutually exclusive levels of adherence to the CPG: (1) complete adherence (colposcopy done if recommended or 3 repeat tests approximately every 6 months until 3 are normal or colposcopy if there is persistent abnormality); (2) moderate adherence (1 or 2 follow-up tests after the index Pap test); and (3) low adherence (no follow-up cytology or histology, or abnormality on repeat Pap test without colposcopy.)

Univariate analyses were performed relating the level of follow-up to the description of the abnormality (atypia vs ASCUS), patient demographics, and practice and provider characteristics (eg, training, availability of colposcopy on site, and plan of care). In analyses where cell size fell below 5, Fisher exact test or Yates correction (for analyses with row or column size greater than 2) replaced the Pearson chi-square. We performed multivariate analysis using a hierarchical model with patients nested within sites, and ordinal logistic regression to account for the 3 ordered levels of adherence to the CPG. Variables significant at P less than .2 in the univariate analyses were included in the multivariate model to ensure inclusion of potentially important variables. Also, insurance status was included, because others have found significant associations for this variable.^21-23 The provider’s management plan was not included as a variable in the regression model, because it is not sufficiently independent of the adherence levels.

Because subjects were clustered within 7 sites (not a simple random sample), we treated site as a random effect with patients nested within the site. An ordered logistic regression model was fit by maximum likelihood estimation, using the program GLLAMM⁶ in Stata version 6.0 The fit of the ordered logistic regression model to the data was tested using the method of Harrell and colleagues.²⁴ Briefly, this method compares observed mean values of the model’s linear combination of the predictor variables to expected values under the assumption that the model (and in particular the proportional odds ratio assumption) is correctly specified.

Results

of the 570 women identified with an atypical or ASCUS Pap test result in 1996, 554 charts were located and reviewed. The charts of 16 individuals could not be located for review. Some women (n=69) identified had a previous atypical or ASCUS Pap test result. The sample of women with no previous ASCUS or atypical result, excluding those with documented HIV infection (n=19) and those with a history of CIN or SIL (n=79), consisted of 387 subjects. The mean age of the subjects at the time of the index Pap test was 32 years (standard deviation=11.7). More than half of the subjects (58.4%) were Hispanic. The majority of subjects had Medicaid (73.1%). The majority of index Pap result abnormalities were described as atypia (61.5%), with the remainder ASCUS (38.5%). A small percentage of women (8.8%) were pregnant at the time of the index Pap test. The subjects were approximately evenly split between receiving care in medicine/pediatrics/obstetrics model sites and family medicine model sites. Colposcopy was provided on site (at least 2 times per week) in 4 of the centers that served 68.7% of the subjects.

Overall, many subjects remained in care in the health centers for a relatively short time. Almost one fourth of the subjects (23.7%) had a final visit to the health center less than 6 months from the index Pap test. Complete data including 2 full years of follow-up from the date of the index Pap test were available for 41% of subjects. Subjects with new atypia had a small chance of SIL on subsequent tests (6.5%, increasing to 10.9% for those subjects with persistent atypia). Results of biopsies for women with new and persistent atypia are presented in Table 1. The majority of subjects had normal or low-grade findings on histology, and no cancer was diagnosed.

Management of atypical results is depicted in the Figure 1. Providers were credited with a “correct” plan if they recommended either a repeat test in 6 months or less, or colposcopy. “Incorrect” management included any other recommendation or failure to document any plan (18.9%). Providers were most likely to recommend a repeat test in 3 months (31.8%) or 6 months (46.1%). We considered communication to subjects “complete” if the notes reflected discussion of the result at a visit (58.3%) or if a completed telephone call was documented (3.9%). Communication was considered attempted if a letter was sent (47.2%) or a call attempted (8.8%) and documented. In 8.3% of charts no documentation of attempted or completed communication was present. There were no significant associations of provider plan or communication of results with demographic variables, pregnancy, or abnormality type. Family physicians were more likely than midlevel providers or other physicians to generate a correct plan (87.3%, 80.8%, 69.8%, respectively; P=.002). Family physicians documented completed communication of results in 77.8% of cases, compared with 70.1% for other physicians and 57.4% for midlevel providers.

Providers were more likely to recommend colposcopy when the index Pap test result was reported as ASCUS rather than atypia (22.1% vs 11.4%, P=.016). Providers were much more likely to choose colposcopy if the pathologist made a recommendation for colposcopy rather than a repeat test (85.2% vs 8.9%, P=.001) and if the report included “favored SIL” in the result (75.0% vs 11.5%, P=.001.) When colposcopy was recommended, 70.8% completed the procedure. After a first atypical Pap test, 36% had no follow-up test; 27% had 1 follow-up test; 20% had 2 tests; and 17% had 3 subsequent tests. Subjects were found to have complete, moderate, and low adherence levels as follows: 26.9%, 27.6%, and 45.5%.

Univariate associations with levels of adherence are shown in Table 2. Complete adherence was more likely when the result was discussed at a visit, when the provider’s plan was documented, when the provider was other than a family physician, and if the plan was colposcopy as opposed to repeat cytology. Complete adherence was associated with remaining in care at the centers after 1997. It was more likely for results reported as ASCUS as opposed to atypia. Large differences are seen in the number of women with complete adherence by site, with the percentage ranging from 7% to 57%.

In the ordered logistic regression analysis, 4 factors were found to be predictive of the quality of adherence: the description of the abnormality (atypia vs ASCUS), availability of colposcopy on-site, discussion of the plan at a visit, and site. The detailed results are shown in Table 3. The adjusted odds ratio for complete versus moderate and low adherence (or equivalently moderate and complete vs low) associated with having an ASCUS lesion versus atypia is 1.56; for attending a clinic where colposcopy is referred off-site, 0.39; and for having discussed the follow-up plan, 3.44. The effects of ethnicity, type of health insurance, provider training, and pregnancy were not statistically significant.

Discussion

The introduction of the ASCUS label in the early 1990s created controversy and led to guidelines from the NCI7 and others^8,9 based primarily on expert opinion. Data from a large randomized clinical trial25 currently underway may resolve some of the questions about the effectiveness of repeat cytology versus immediate colposcopy in the management of low-grade dysplasia. Importantly, lack of evidence for the guideline may adversely affect compliance. Implementation of either strategy is difficult, especially in communities with many risk factors for poor follow-up. Our results indicate low adherence to the NCI interim guideline in an urban primary care setting serving predominately low-income minority women. Of concern is that almost half our sample, compared with 30% of subjects in a Canadian study,²⁶ had no follow-up cytology or histology. Our findings highlight some of the difficulties encountered in managing low-grade abnormalities in a setting challenged by high provider turnover, staffing difficulties, incorrect phone and address information, and financial and language barriers. Long-term repeated cytology is difficult to accomplish where loss to follow-up is common and competing demands of managing multiple chronic illnesses may interfere. Our findings are similar to those of other investigators^27,28 in that colposcopy, when recommended, was not completed by approximately one third of the women.

Our findings highlight the importance of strict use of Bethesda terminology by cytopathologists, since the older terminology may be misunderstood and less aggressively followed by primary care clinicians. We chose to include women with either atypical or ASCUS results, because they are different labels for the same spectrum of abnormalities. We found that providers were slightly more likely to recommend colposcopy for ASCUS and ultimately that women with atypical results were more likely to have low adherence to the CPG. This probably reflects that many primary care providers do not view the guideline as applicable to results reported as atypia or communicate with patients about atypical results differently.

There were substantial differences in rates of follow-up by site that are likely related to several factors. Although tracking of abnormal Pap test results was required at all sites the intensity varied, with the most successful site staffed by registered nurses who worked closely with the colposcopy providers in maintaining a manual card file. Sites where colposcopy was recommended more often had overall higher proportions of women with complete adherence to the CPG than those sites where serial cytology was the follow-up strategy of choice. Family physicians were more likely to generate a correct plan yet had worse overall adherence. This is likely due to a preference to follow up women by serial cytology, which proves difficult in settings where patients often stay in care for less than 2 years. However, the finding persists even after controlling for the provider plan, suggesting that other factors are also contributing to account for the differences between sites. Importantly, providers were more likely to opt for colposcopy, and women were more likely to complete it if the service was available in the subject’s health center.

We are unable to determine all of the causes of inadequate follow-up. Systemic factors, however, are clearly implicated by the present data, as indicated by the substantial variation by site and the number of charts in which a clear provider plan was not indicated or where no indication that the subject had been informed of the result was documented. When documentation was present that results were discussed with patients during visits patients were much more likely to have good follow-up, again suggesting that good communication of results is imperative. Our results also suggest that reducing barriers to colposcopy, by such means as providing the service on-site, may be effective in achieving optimal follow-up when colposcopy is recommended. Recommending colposcopy for all women with an initial ASCUS or atypical Pap test result will result in significant stress on available colposcopic resources and expose large numbers of women to an expensive, uncomfortable, and distressing procedure unnecessarily. However, providers need to carefully consider the risk of loss to follow-up in making a recommendation, take extra effort to insure that women are informed of and understand their Pap test result, and establish clinical tracking efforts to meet the challenge of serial cytology in high-risk settings.

Limitations

Our findings are limited in that the standard of care to which we compared practice is based on a guideline that is not evidence based. Other limitations of the project are largely due to constraints imposed by the abstraction of retrospective data from primary care charts. Information about how women were informed of results is sketchy, and failure to document may result in underestimation of actual patient contact. The extent to which women had follow-up if they no longer received care in the health centers cannot be estimated. Some women may have had follow-up with other providers outside our system. Demographic information was obtained from site registration databases rather than self-report and may include some misclassification. Some important information is not captured consistently in the site’s computers, including changes in insurance and primary care providers. Provider turnover was extremely high in these health centers during the interval studied, though lists of providers are not available to create a precise measure to include in our model.

Acknowledgments

This work was supported by a clinical research training grant from the American Cancer Society.

Approximately 50 million Papanicolaou (Pap) tests are performed annually in the United States, of which approximately 5% will require further evaluation because of an abnormality.¹ The Bethesda System for reporting cervical/vaginal cytologic diagnoses,² developed in 1988 and revised in 1991, provides uniform diagnostic terminology to improve consistency of reporting. The Bethesda System introduced new terminology for atypical cytology, creating the label “atypical squamous cells of uncertain significance” (ASCUS). Pap test results labelled ASCUS encompass a spectrum of cellular change reflecting a variety of pathologic processes that cannot be more specifically categorized, including reactive changes, inflammation, human papillomavirus (HPV) related changes, suggested dysplasia, and less than optimal slide preparation.² Because studies have consistently shown a small but real risk that an ASCUS Pap result represents an underlying high-grade lesion,^3-6 there has been much debate about their management. Uncertainty about optimal management led to the development of clinical practice guidelines (CPG).^7-9

The National Cancer Institute⁷ (NCI) and a Canadian expert pane¹⁸ suggest that ASCUS and low-grade squamous intraepithelial lesions can be managed either by immediate colposcopy or by repeating cervical cytology. Clinicians may, if they choose, reserve referral to colposcopy for those patients with persistent abnormality on subsequent cytology. This conservative strategy is based on several assumptions:¹⁰ (1) conditions that mimic dysplasia, including reactive changes, are common; (2) spontaneous regression occurs in at least 50% of low-grade lesions; (3) the rate of progression is relatively slow; (4) the chance of carcinoma in situ with an ASCUS Pap test result is low; and (5) persistently negative follow-up Pap results are unlikely to be false negatives. The current practice guideline suggests that if clinicians opt for cytologic follow-up of ASCUS, repeat Pap tests should be done every 4 to 6 months until 3 consecutive normal test results have been obtained. Colposcopy should be performed if there is a subsequent abnormal Pap test result.⁷ The 1996 statement of the American College of Obstetricians and Gynecologists is similar, with the addition that colposcopy be performed for women with a single ASCUS Pap result if they also have any of the following high-risk factors: HPV infection, human immunodeficiency virus (HIV) infection, smoking, or multiple sexual partners.⁹

A disproportionate share of the cervical cancer burden in this country, in terms of both incidence and mortality, is shouldered by low-income minority women.^11-13 Minority and low-income populations have the highest rates of nonadherence with repeat Pap test or colposcopy.^14-18 Thus, the women who are at the highest risk of cervical intraepithelial neoplasia (CIN) may also be the most at risk for suboptimal management after an abnormal Pap test result is obtained. This project assessed current management of ASCUS Pap results in a diverse community at high risk of cervical cancer by comparing actual practice to the NCI guideline. In addition to assessing levels of follow-up in this community, we sought to identify predictors of adherence to that guideline.

Methods

Setting and Population

Data were collected in 7 community health centers serving a low-income, predominantly minority population. The centers represent diversity with regard to practice type (medicine/pediatrics/obstetrics model or family medicine model) and size (20,000-80,000 visits/year). The subjects were women with newly identified atypia or ASCUS Pap test results in 1996; women for whom the index Pap test represented persistent atypia or ASCUS were excluded, as were women with a history of CIN or squamous intraepithelial lesion (SIL). Those who were known to be infected with HIV represent a special category^19,20 that may have been managed differently at the clinician’s discretion; therefore, they were also excluded from the analysis.

Data Collection

After approval by the appropriate institutional review boards, women with atypical results in 1996 were identified by 2 mechanisms: site logs maintained for clinical tracking, and because all sites exclusively use the same laboratory, the central cytology database. Pap test results described as atypia or ASCUS were included, as these represent different terminology for the same spectrum of cytologic abnormalities with continued use of the older term by some cytopathologists. Chart abstraction by a trained research assistant collected information regarding: (1) complicating medical conditions (immunosuppression, history of cancer, substance abuse, pregnancy); (2) history of cervical dysplasia; (3) primary care provider and number of visits to site; (4) the provider’s management plan; (5) all subsequent Pap test and colposcopy results; (6) evidence of communication of the management plan to the subject; and (7) demographics.

Accuracy of chart abstraction was confirmed by the lead author rescreening 20% of the records. The data were entered into a Paradox database (Borland Software Corporation, Scotts Valley, Calif), and statistical analysis was performed using SAS software (SAS Institute, Inc, Cary, NC) and Stata software (Stata Corporation, College Station, Tex).

Analysis

The goal of our analysis was to determine the factors associated with follow-up. We defined 3 mutually exclusive levels of adherence to the CPG: (1) complete adherence (colposcopy done if recommended or 3 repeat tests approximately every 6 months until 3 are normal or colposcopy if there is persistent abnormality); (2) moderate adherence (1 or 2 follow-up tests after the index Pap test); and (3) low adherence (no follow-up cytology or histology, or abnormality on repeat Pap test without colposcopy.)

Univariate analyses were performed relating the level of follow-up to the description of the abnormality (atypia vs ASCUS), patient demographics, and practice and provider characteristics (eg, training, availability of colposcopy on site, and plan of care). In analyses where cell size fell below 5, Fisher exact test or Yates correction (for analyses with row or column size greater than 2) replaced the Pearson chi-square. We performed multivariate analysis using a hierarchical model with patients nested within sites, and ordinal logistic regression to account for the 3 ordered levels of adherence to the CPG. Variables significant at P less than .2 in the univariate analyses were included in the multivariate model to ensure inclusion of potentially important variables. Also, insurance status was included, because others have found significant associations for this variable.^21-23 The provider’s management plan was not included as a variable in the regression model, because it is not sufficiently independent of the adherence levels.

Because subjects were clustered within 7 sites (not a simple random sample), we treated site as a random effect with patients nested within the site. An ordered logistic regression model was fit by maximum likelihood estimation, using the program GLLAMM⁶ in Stata version 6.0 The fit of the ordered logistic regression model to the data was tested using the method of Harrell and colleagues.²⁴ Briefly, this method compares observed mean values of the model’s linear combination of the predictor variables to expected values under the assumption that the model (and in particular the proportional odds ratio assumption) is correctly specified.

Results

of the 570 women identified with an atypical or ASCUS Pap test result in 1996, 554 charts were located and reviewed. The charts of 16 individuals could not be located for review. Some women (n=69) identified had a previous atypical or ASCUS Pap test result. The sample of women with no previous ASCUS or atypical result, excluding those with documented HIV infection (n=19) and those with a history of CIN or SIL (n=79), consisted of 387 subjects. The mean age of the subjects at the time of the index Pap test was 32 years (standard deviation=11.7). More than half of the subjects (58.4%) were Hispanic. The majority of subjects had Medicaid (73.1%). The majority of index Pap result abnormalities were described as atypia (61.5%), with the remainder ASCUS (38.5%). A small percentage of women (8.8%) were pregnant at the time of the index Pap test. The subjects were approximately evenly split between receiving care in medicine/pediatrics/obstetrics model sites and family medicine model sites. Colposcopy was provided on site (at least 2 times per week) in 4 of the centers that served 68.7% of the subjects.

Overall, many subjects remained in care in the health centers for a relatively short time. Almost one fourth of the subjects (23.7%) had a final visit to the health center less than 6 months from the index Pap test. Complete data including 2 full years of follow-up from the date of the index Pap test were available for 41% of subjects. Subjects with new atypia had a small chance of SIL on subsequent tests (6.5%, increasing to 10.9% for those subjects with persistent atypia). Results of biopsies for women with new and persistent atypia are presented in Table 1. The majority of subjects had normal or low-grade findings on histology, and no cancer was diagnosed.

Management of atypical results is depicted in the Figure 1. Providers were credited with a “correct” plan if they recommended either a repeat test in 6 months or less, or colposcopy. “Incorrect” management included any other recommendation or failure to document any plan (18.9%). Providers were most likely to recommend a repeat test in 3 months (31.8%) or 6 months (46.1%). We considered communication to subjects “complete” if the notes reflected discussion of the result at a visit (58.3%) or if a completed telephone call was documented (3.9%). Communication was considered attempted if a letter was sent (47.2%) or a call attempted (8.8%) and documented. In 8.3% of charts no documentation of attempted or completed communication was present. There were no significant associations of provider plan or communication of results with demographic variables, pregnancy, or abnormality type. Family physicians were more likely than midlevel providers or other physicians to generate a correct plan (87.3%, 80.8%, 69.8%, respectively; P=.002). Family physicians documented completed communication of results in 77.8% of cases, compared with 70.1% for other physicians and 57.4% for midlevel providers.

Providers were more likely to recommend colposcopy when the index Pap test result was reported as ASCUS rather than atypia (22.1% vs 11.4%, P=.016). Providers were much more likely to choose colposcopy if the pathologist made a recommendation for colposcopy rather than a repeat test (85.2% vs 8.9%, P=.001) and if the report included “favored SIL” in the result (75.0% vs 11.5%, P=.001.) When colposcopy was recommended, 70.8% completed the procedure. After a first atypical Pap test, 36% had no follow-up test; 27% had 1 follow-up test; 20% had 2 tests; and 17% had 3 subsequent tests. Subjects were found to have complete, moderate, and low adherence levels as follows: 26.9%, 27.6%, and 45.5%.

Univariate associations with levels of adherence are shown in Table 2. Complete adherence was more likely when the result was discussed at a visit, when the provider’s plan was documented, when the provider was other than a family physician, and if the plan was colposcopy as opposed to repeat cytology. Complete adherence was associated with remaining in care at the centers after 1997. It was more likely for results reported as ASCUS as opposed to atypia. Large differences are seen in the number of women with complete adherence by site, with the percentage ranging from 7% to 57%.

In the ordered logistic regression analysis, 4 factors were found to be predictive of the quality of adherence: the description of the abnormality (atypia vs ASCUS), availability of colposcopy on-site, discussion of the plan at a visit, and site. The detailed results are shown in Table 3. The adjusted odds ratio for complete versus moderate and low adherence (or equivalently moderate and complete vs low) associated with having an ASCUS lesion versus atypia is 1.56; for attending a clinic where colposcopy is referred off-site, 0.39; and for having discussed the follow-up plan, 3.44. The effects of ethnicity, type of health insurance, provider training, and pregnancy were not statistically significant.

Discussion

The introduction of the ASCUS label in the early 1990s created controversy and led to guidelines from the NCI7 and others^8,9 based primarily on expert opinion. Data from a large randomized clinical trial25 currently underway may resolve some of the questions about the effectiveness of repeat cytology versus immediate colposcopy in the management of low-grade dysplasia. Importantly, lack of evidence for the guideline may adversely affect compliance. Implementation of either strategy is difficult, especially in communities with many risk factors for poor follow-up. Our results indicate low adherence to the NCI interim guideline in an urban primary care setting serving predominately low-income minority women. Of concern is that almost half our sample, compared with 30% of subjects in a Canadian study,²⁶ had no follow-up cytology or histology. Our findings highlight some of the difficulties encountered in managing low-grade abnormalities in a setting challenged by high provider turnover, staffing difficulties, incorrect phone and address information, and financial and language barriers. Long-term repeated cytology is difficult to accomplish where loss to follow-up is common and competing demands of managing multiple chronic illnesses may interfere. Our findings are similar to those of other investigators^27,28 in that colposcopy, when recommended, was not completed by approximately one third of the women.

Our findings highlight the importance of strict use of Bethesda terminology by cytopathologists, since the older terminology may be misunderstood and less aggressively followed by primary care clinicians. We chose to include women with either atypical or ASCUS results, because they are different labels for the same spectrum of abnormalities. We found that providers were slightly more likely to recommend colposcopy for ASCUS and ultimately that women with atypical results were more likely to have low adherence to the CPG. This probably reflects that many primary care providers do not view the guideline as applicable to results reported as atypia or communicate with patients about atypical results differently.

There were substantial differences in rates of follow-up by site that are likely related to several factors. Although tracking of abnormal Pap test results was required at all sites the intensity varied, with the most successful site staffed by registered nurses who worked closely with the colposcopy providers in maintaining a manual card file. Sites where colposcopy was recommended more often had overall higher proportions of women with complete adherence to the CPG than those sites where serial cytology was the follow-up strategy of choice. Family physicians were more likely to generate a correct plan yet had worse overall adherence. This is likely due to a preference to follow up women by serial cytology, which proves difficult in settings where patients often stay in care for less than 2 years. However, the finding persists even after controlling for the provider plan, suggesting that other factors are also contributing to account for the differences between sites. Importantly, providers were more likely to opt for colposcopy, and women were more likely to complete it if the service was available in the subject’s health center.

We are unable to determine all of the causes of inadequate follow-up. Systemic factors, however, are clearly implicated by the present data, as indicated by the substantial variation by site and the number of charts in which a clear provider plan was not indicated or where no indication that the subject had been informed of the result was documented. When documentation was present that results were discussed with patients during visits patients were much more likely to have good follow-up, again suggesting that good communication of results is imperative. Our results also suggest that reducing barriers to colposcopy, by such means as providing the service on-site, may be effective in achieving optimal follow-up when colposcopy is recommended. Recommending colposcopy for all women with an initial ASCUS or atypical Pap test result will result in significant stress on available colposcopic resources and expose large numbers of women to an expensive, uncomfortable, and distressing procedure unnecessarily. However, providers need to carefully consider the risk of loss to follow-up in making a recommendation, take extra effort to insure that women are informed of and understand their Pap test result, and establish clinical tracking efforts to meet the challenge of serial cytology in high-risk settings.

Limitations

Our findings are limited in that the standard of care to which we compared practice is based on a guideline that is not evidence based. Other limitations of the project are largely due to constraints imposed by the abstraction of retrospective data from primary care charts. Information about how women were informed of results is sketchy, and failure to document may result in underestimation of actual patient contact. The extent to which women had follow-up if they no longer received care in the health centers cannot be estimated. Some women may have had follow-up with other providers outside our system. Demographic information was obtained from site registration databases rather than self-report and may include some misclassification. Some important information is not captured consistently in the site’s computers, including changes in insurance and primary care providers. Provider turnover was extremely high in these health centers during the interval studied, though lists of providers are not available to create a precise measure to include in our model.

Acknowledgments

This work was supported by a clinical research training grant from the American Cancer Society.