 |
|
Barriers to Follow-up of Abnormal Papanicolaou Smears in an Urban Community Health Center
M. Diane McKee, MD;
Joseph Lurio, MD;
Paul Marantz, MD, MPH;
William Burton, MA;
Michael Mulvihill, DrPH
Arch Fam Med. 1999;8:129-134.
ABSTRACT
| |
Objective To determine factors predictive of failure to return for colposcopy among women with significant abnormalities on Papanicolaou smears in a high-risk clinical population.
Design Telephone survey.
Setting An urban community health center.
Participants Two hundred seventy-nine women randomly selected from all women seen at the health center with abnormal Papanicolaou smears requiring colposcopy during 1993 to 1994. Six (2%) refused participation, and 19% could not be reached for inclusion. Subjects were mostly minority women receiving Medicaid.
Main Outcome Measure Completion of colposcopy.
Results Of the 279 selected women, 79% were interviewed. The rate of adherence with colposcopy was 75% for the respondents. Women who did not know the results of their smear or who incorrectly understood their results were significantly less likely to return for colposcopy (P=.001). Younger women, especially teenagers, were less likely to return (P=.02). Socioeconomic status, education, primary language, health beliefs, fear of cancer, and clinician's gender or discipline were not associated with rate of follow-up. Barriers involving transportation, child care, and insurance also did not predict follow-up.
Conclusions Effective communication of results is the most important factor related to follow-up after abnormal Papanicolaou smear in this setting. In other settings, other factors may be of greater importance.
INTRODUCTION
ALTHOUGH CERVICAL cancer screening by use of the Papanicolaou (Pap) smear has been associated with declining mortality rates from invasive cervical cancer, this disease continues to have substantial impact. In the United States in 1994, 15,000 women were diagnosed as having invasive cervical cancer.1 It has also been noted that the number of women with preinvasive cervical abnormalities has been increasing in recent years, largely because of epidemic exposure to human papillomavirus.2 Effective prevention of cervical cancer depends on early diagnosis and management of these preinvasive lesions.
When Pap smear results suggest precancerous lesions, adequate follow-up involves either repeated Pap smears or colposcopy to identify women who require treatment. Unfortunately, failure to return for follow-up after abnormal cervical cytological findings is a common problem. Poor follow-up may be the result of cultural, financial, or psychological barriers, or may be related to the health care delivery system itself.3 The result of failure to follow up is a delay in the diagnosis and treatment of potentially malignant lesions.
Women in poor and minority communities have been identified as being less likely to utilize screening Pap smears. In addition, they are less likely to follow up after an abnormal Pap smear.4 Typical estimates of the percentage of women who fail to return for follow-up after an abnormal Pap smear range from 30% to 44%5-9 but have been reported at 49%10 and higher in some populations. Lower follow-up rates have been observed for younger women, those lacking health insurance, those with less than a high school education,7 and unmarried women.11 Lower rates have been observed among ethnic groups, including Southeast Asians12 and black and Hispanic women.7 This loss to follow-up is likely to contribute to the substantially higher rate of invasive cervical carcinoma observed among poor and minority women.4
Barriers to follow-up after abnormal cervical cytological findings are least well studied in communities at highest risk of cervical cancer. Paskett et al10 used in-depth interviews with women of mixed socioeconomic status to identify beliefs and values related to the decision to seek follow-up. However, this study included mostly white women and no non-English speakers. In a structured telephone interview with mostly low-income black women, Lerman et al13 identified poor understanding of the purpose of colposcopy, forgetting, and fear as the most important barriers to colposcopy. This study, however, included few Hispanics and was limited to English speakers.
This investigation was conducted to determine the barriers to follow-up among women with significant abnormalities on Pap smear in a high-risk clinical population. The patients studied were largely poor, minority women from an urban community. Extensive patient interviews were conducted in English and Spanish to identify factors predictive of failure to receive adequate follow-up, with an overall goal of developing systems to improve rates of return for colposcopy in this setting.
SUBJECTS AND METHODS
This study was conducted at the Montefiore Family Health Center, an urban community health center in the Bronx, NY, that provides acute and comprehensive health care for more than 16,000 patients. The mortality from cervical cancer in this community is 1.5 times greater than the national average.14 Montefiore Family Health Center serves a largely minority, inner-city community, 32% of whom are black, 36% Hispanic, 23% white, and 9% Asian. Fifty-three percent of the Montefiore Family Health Center patients receive Medicaid, 23% are uninsured, 21% are privately insured, and 3% receive Medicare. Primary care is provided by a diverse group of family practitioners (residents, attending physicians, and nurse practitioners), a large number of whom speak Spanish. Translation is readily available in Spanish, Khmer, and Vietnamese for the patients who need it.
Pap smear abnormalities requiring colposcopy were defined as low-grade squamous intraepithelial lesions, high-grade squamous intraepithelial lesions, persistent atypia, or any Pap smear result for which colposcopy is recommended by the pathologist. Women with squamous atypia or atypical squamous cells of uncertain significance were not included unless there were at least 2 such smears. All women with Pap smear abnormalities in 1993 or 1994 (n=564) were identified. By means of a random number table, 350 were selected, of whom 279 met the inclusion criterion of having a Pap smear abnormality requiring colposcopy.
The research team developed the survey instrument after reviewing previous research on follow-up of abnormal Pap smears and adherence with colposcopy. Questions were drawn from existing models of health behavior, including locus of control and the health belief model.15 In addition, the survey included items that addressed a number of potential barriers suggested by previous research,7, 10, 16-18 including fear, embarrassment, economic barriers, and practical barriers such as transportation and child care. Also assessed were health center issues such as waiting time and appointment availability.
The survey was piloted on 12 subjects, drawn from the same pool and not included in the results, and revisions were made to improve clarity. The survey instrument was translated into Spanish, then back-translated to English to avoid ambiguity. A fully bilingual research assistant was trained to perform the chart abstraction and telephone interviews. A protocol was established to assist patients in obtaining follow-up appointments, if needed, when they were contacted for participation in the study.
The revised questionnaire was then administered by telephone after consent was obtained, in English or Spanish as requested by the patient. Women who could not be reached by telephone were sent an initial letter, and nonresponders were sent a second letter. To locate women who may have had inaccurate addresses, notices were also left in the chart stating that the patient could not be reached for inclusion in the study. Several patients were located in this way when they came to the clinic for other indications. A third letter was sent to nonresponders offering a small monetary incentive for participation. For any patients who could not be reached at any available telephone number or address, a computer database search was undertaken by means of the Lexis-Nexis service, matching name, date of birth, and social security number with national data. Finally, the National Death Index was searched to determine if any unreachable patients were deceased.
To determine whether patient self-report of colposcopy is accurate, we verified that colposcopy had been completed by reviewing chart and laboratory data for a sample of the participants (n=41). No patient reported colposcopy who did not actually have it. However, 3 patients from this sample (9%) reported that they did not have colposcopy when they actually did, suggesting that some women had the procedure but did not fully understand it was happening. Thus, for all patients who reported no colposcopy, we verified their report by laboratory and chart records.
Data were analyzed by means of SAS software (SAS Institute, Cary, NC). Patients were categorized as having had follow-up colposcopy or not at the time of the interview. Elapsed time between the abnormal Pap smear and the interview was 17 to 47 months. Univariate predictors of colposcopy were assessed by the  2 test, or Fisher exact test where appropriate.
RESULTS
Of the randomly selected 279 women with Pap smear results indicating a need for colposcopy, 6 (2%) refused participation in the study and 54 (19%) were unreachable at any available telephone number or address. Women who could not be reached for inclusion in the study were similar to those who were interviewed with regard to age, insurance status, income, and severity of Pap smear result. They differed only in the percentage of women who identified themselves as black (54% compared with 34% in the study population) and Hispanic (43% compared with 58% in the study population). Seventeen (6%) of the 279 women were deceased, all from complications of acquired immunodeficiency syndrome. Interviews were completed for 202 women. Table 1 describes the participants. The majority of participants (75%) had low-grade Pap smear abnormalities. Subjects included women who identified themselves as Hispanic, black, white, and Asian, with a mean age of 30.3 years. The majority of participants were insured by Medicaid. More than half had at least a high school education.
|
|
|
|
Table 1. Characteristics of Study Participants*
|
|
|
Of the 202 interviewed participants, 151 (75%) had colposcopy. As described in Table 2, older patients were more likely to have had colposcopy than younger participants. Teenagers were significantly less likely to have follow-up (40%) than women aged 18 to 29 years (73%) and women aged 30 years or older (81%) (P=.02). Women with higher-grade lesions (high-grade squamous intraepithelial lesions and cancer) were more likely to have had colposcopy than women with lower-grade lesions (87% vs 72%), but this finding only approaches statistical significance (P=.06). No significant difference in adherence with colposcopy was found for women of differing race or ethnicity. Women adherent and nonadherent with colposcopy did not differ with regard to education level or number of children.
|
|
|
|
Table 2. Adherence With Colposcopy by Selected Variables*
|
|
|
As shown in Table 3, women who reported that they did not know the result of their Pap smear (14% of respondents) were much less likely to have had colposcopy (36% vs 81%; P=.001). Women who reported their results correctly as abnormal were more likely to have had colposcopy (83% vs 59%; P=.02) than were women who incorrectly reported that they had normal results. Participants who reported that they received a letter informing them of the Pap smear result were more adherent (87% vs 66%; P=.001). Thirty-seven (18%) of the study participants did not have an accurate address in the registration database. Eighty-six (45%) of the participants did not have an accurate telephone number recorded, and these women were slightly less likely to have had colposcopy (69% vs 79%), although this result did not reach statistical significance (P=.10).
|
|
|
|
Table 3. Barriers to Colposcopy
|
|
|
Thirty-seven (18%) of the participants reported that they were more comfortable communicating in Spanish. However, these patients were not less likely to complete colposcopy, nor did having a Spanish-speaking provider affect the likelihood of adherence for Hispanic patients. Language did not appear to be a common barrier, as 92% of respondents disagreed with the statement that "no one speaks my language to explain." However, for the small number of women (n=14) who disagreed with the statement that it is "easy to get the doctor to understand my needs," the rate of return for colposcopy was substantially lower (50%; P=.03).
Provider gender and type (nurse practitioner or family physician) were not associated with return for colposcopy. Women whose primary providers were residents had a rate of return for colposcopy similar to that of women whose primary providers were attending physicians. Practical barriers, such as the need for child care, time off work, and difficulty getting to the health center, were not associated with rates of follow-up, nor were economic barriers, such as income and insurance.
Many women reported fears related to having an abnormal Pap smear. The majority reported fear that an abnormal Pap smear may mean cancer (80%). As shown in Table 4, questions about pain or side effects of colposcopy were answered primarily by women who had already undergone colposcopy. Among the women who had had colposcopy, the majority (72%) agreed with the statement that colposcopy may be painful, and 38% agreed with the statement that colposcopy may have side effects.
|
|
|
|
Table 4. Fears Associated With Having an Abnormal Papanicolaou Smear Result
|
|
|
COMMENT
The most important finding in this study is that lack of effective communication of abnormal results or their implications occurs and is associated with low rates of return for colposcopy. Women who either do not know their Pap smear result or do not understand that their Pap smear was abnormal are at very high risk of no follow-up. Importantly, only 14% of the participants reported that they did not know the result of their Pap smear. However, 10% of the women who reported a result reported it incorrectly. When women recalled receiving a letter, they were substantially more likely to have follow-up.
These findings suggest that many women who have been informed do not fully understand the meaning of these results and may not understand the need for follow-up. This result is similar to the finding by Lerman et al,13 who reported that lack of understanding of the purpose of colposcopy is an important barrier. Interestingly, lack of English fluency did not correlate with lack of understanding. The large majority of subjects interviewed in this study described the health center as a place of respect, where they were able to communicate their needs. It is not possible to determine how much this rate of follow-up is influenced by "formal" communication, such as letters, compared with the informal communication fostered by continuity between women and their providers. The latter appears to be important as well, because those few women (7%) who reported that they were not able to communicate their needs were substantially less likely to return for colposcopy.
Forty-five percent of the participants did not have correct telephone numbers and 18% did not have correct addresses in the registration database. It is likely that these issues contribute to some women not being informed. However, the rate of follow-up colposcopy was not significantly different for women with and without correct registration data, while there was an association with understanding that a result was abnormal. This suggests that clinicians have added responsibility to ensure that women not only are informed but also understand the meaning of their results. Maintaining accurate information to contact women is especially challenging in this setting. It is noteworthy that during the process of contacting potential participants, we learned that many patients gave false registration information. This may be because of the nature of this community health center practice, where care is provided to many undocumented aliens, and many individuals are strongly motivated to avoid getting bills because of limited financial resources. Patients who are inclined to give false information might at least provide accurate emergency contact information if the intended use of this information is more clearly communicated by the health center at the time of registration.
THIS STUDY was conducted in a setting designed to maximize follow-up, which emphasizes culturally and language-appropriate clinical care and where a centralized Pap smear tracking and notification system was already in place. Follow-up with recommended colposcopy after abnormal Pap smear in this center is better than reported rates for similar high-risk populations.6, 10, 13 Although this issue was not directly addressed by our data, we hypothesize that several factors may contribute to the better-than-expected overall return rate that we observed. Colposcopy is provided on site, and the center offers sliding scale fees for uninsured patients. Thus, 2 possible barriers, referral to another site and cost, did not seem to contribute to lack of follow-up. In addition, the large majority of women receiving care in this center reported that they were treated with respect, and that they were able to communicate their needs (even if not English speakers.) A clinical setting that is less responsive to its diverse population might be expected to have lower levels of follow-up.
This study confirms the findings of other investigators7, 11 who identified younger patients as being at higher risk of poor follow-up. Teenagers appear to be at especially high risk. Lerman et al13 identified several practical barriers to adherence; however, our data did not confirm that practical barriers such as child care, transportation, or cost were associated with failure to follow up in this setting. Our data are consistent with those of Lerman et al in that our participants did not report the belief that follow-up was not needed if they were asymptomatic. The large majority of our participants believed that early detection was important.
Previous investigators have found that women with an abnormal Pap smear are afraid of cancer and of colposcopy and its side effects.13, 16-17 Similar fears were commonly reported by participants in this study. However, unlike these previous studies, our data do not demonstrate that fear of cancer influences the rate of follow-up. Unfortunately, many women in our study, including the majority of those who had not had colposcopy, did not respond to the questions about fear of pain and side effects from colposcopy. It is likely that women unfamiliar with colposcopy were hesitant to speculate. Because this project did not interview and then prospectively assess rates of follow-up, we are unable to determine whether fear of colposcopy is a barrier to follow-up. Interestingly, many women who had had colposcopy had persistent fear of its potential side effects, suggesting that the education that takes place at the time of colposcopy needs to address this concern more effectively.
There are other limitations to this study that should be considered. We could not reach 19% of the eligible study subjects. Their rate of follow-up may be lower than that for women who could be reached to participate. However, we were not able to accurately assess the rate of completed colposcopy among women not reached, because we could not determine how many had follow-up care elsewhere. If all the unreachable group did not have colposcopy, the rate of follow-up would be 58%, compared with 81% if all of the unreachable group had colposcopy. Comparing the demographics of the unlocated group with those of our final sample, the only significant difference found was a higher proportion of black women and fewer Hispanics among those not located.
Because there was a range of 17 to 47 months from Pap smear to interview, it is possible that recall bias could affect results. This effect on responses regarding attitude and belief is likely to be minimal, and the final outcome of colposcopy is accurate because we did not rely exclusively on self-report. Women's recall of how they were informed of their result may diminish over time; however, because of the strength of the association between recalling a letter and having colposcopy, it is unlikely that correcting for recall bias would alter the significance of this result. Because the setting of this study is an urban community health center, the findings may not be generalizable to other settings. However, the results are of interest because of the unique and traditionally understudied population, where there is a disproportionate burden of cervical cancer.
Even in a setting with systems in place to maximize follow-up, there was still a substantial number of women without adequate follow-up after an abnormal Pap smear. These women, already identified with potential for a malignant cervical lesion, were at very high risk of cervical cancer. In this community, where the burden of cervical cancer is considerably higher than the national average, the loss to follow-up of one quarter of women who need colposcopy is significant. Timely follow-up of abnormal results permits the opportunity to evaluate and treat cervical lesions with the least invasive means and with the lowest risk of progression to invasive disease. Any program to improve rates of follow-up should begin with improving the communication of abnormal results to women. Such communication should take into account the substantial fear of cancer found among these women. However, fear may not necessarily be a barrier to follow-up if women are informed, especially if results are communicated in a setting of respect and trust.
The purpose of this project was to determine how to intervene to improve the quality of care provided to women with abnormal Pap smears in a diverse inner-city setting. Many potential contributors to poor follow-up were considered, including access problems, language and other communication barriers, psychological barriers, and practical impediments. These results suggest that efforts to improve rates of return for colposcopy in this setting should focus on the notification of women with abnormal results and, more important, better assessment of whether women understand the meaning of these results. The content of the notification message should be unambiguous as well as culturally and linguistically appropriate. Finally, specific efforts are needed to ensure that adolescents with abnormal Pap smears return for further evaluation.
AUTHOR INFORMATION
Accepted for publication February 14, 1998.
This project was supported by a quality improvement grant from the New York State Department of Health, Albany.
We thank Nancy Hernandez for her hard work and Sylvia Solis for her administrative support.
Reprints: M. Diane McKee, MD, Department of Family Medicine, 902 Belfer Bldg, 1300 Morris Park Ave, Bronx, NY 10461 (e-mail: mckee{at}aecom.yu.edu).
From the Departments of Family Medicine (Drs McKee, Lurio, and Mulvihill and Mr Burton) and Epidemiology and Social Medicine (Dr Marantz), Albert Einstein College of Medicine, Bronx, NY.
REFERENCES
| |
1. American Cancer Society. Cancer Facts and Figures. Atlanta, Ga: American Cancer Society; 1994.
2. Koutsky L, Galloway D, Holmes K. Epidemiology of genital human papillomavirus infection. Epidemiol Rev. 1988;10:122-163.
FREE FULL TEXT
3. McKee D. Strategies to improve follow-up of abnormal Pap smears. Arch Fam Med. 1997;6:574-577.
FREE FULL TEXT
4. National Cancer Institute Cancer Screening Consortium for Underserved Women. Breast and cancer screening among underserved women. Arch Fam Med. 1995;4:617-624.
FREE FULL TEXT
5. Laedtke TW, Dignan M. Compliance with therapy for cervical dysplasia among women of low socioeconomic status. South Med J. 1992;85:5-8.
ISI
| PUBMED
6. Cartwright PS, Reed G. No-show behavior in a county hospital clinic. Am J Gynecol Health. 1990;6:15-21.
7. Marcus AC, Crane LA, Kaplan CP, Reading AE, Savage E, Gunning J. Improving adherence to screening follow-up among women with abnormal Pap smears: results from a large clinic-based trial of three intervention strategies. Med Care. 1992;30:216-229.
ISI
| PUBMED
8. Lane DS. Compliance with referrals from a cancer-screening project. J Fam Pract. 1983;15:811-817.
9. Nathoo V. Investigation of non-responders at a cervical cancer screening clinic in Manchester. BMJ. 1988;296:1041-1042.
10. Paskett ED, White E, Carter W, Chu J. Improving follow-up after an abnormal Pap smear: a randomized controlled trial. Prev Med. 1990;19:630-641.
FULL TEXT
|
ISI
| PUBMED
11. Michielutte R, Diseker RA, Young L, May WJ. Noncompliance in screening follow-up among family planning clinic patients with cervical dysplasia. Prev Med. 1985;14:248-257.
FULL TEXT
|
ISI
| PUBMED
12. Carey P, Gjerdingen DK. Follow-up of abnormal Papanicolaou smears among women of different races. J Fam Pract. 1993;37:583-587.
ISI
| PUBMED
13. Lerman C, Hanjani P, Caputo C, et al. Telephone counseling improves adherence to colposcopy among lower-income minority women. J Clin Oncol. 1992;10:330-333.
ABSTRACT
14. National Cancer Institute. SEER Cancer Statistics Review: 1973-1990. Bethesda, Md: National Cancer Institute; 1993. NIH publication 93-2789.
15. Haynes RB, Taylor DW, Sackett DC. Compliance in Health Care. Baltimore, Md: Johns Hopkins University Press; 1979.
16. Beresford JS, Gervaise PA. The emotional impact of abnormal Pap smears on patients referred for colposcopy. Colposc Gynecol Laser Surg. 1986;2:83-87.
17. Lerman C, Miller S, Scarborough R, et al. Adverse psychologic consequences of positive cytologic cervical screening. Am J Obstet Gynecol. 1991;165:658-662.
ISI
| PUBMED
18. Hubbell FA, Chavez LR, Mishra SI, Valdez RB. Beliefs about sexual behavior and other predictors of Papanicolaou screening among Latinas and Anglo women. Arch Intern Med. 1996;156:2353-2358.
FREE FULL TEXT
RELATED ARTICLE
The Archives of Family Medicine Continuing Medical Education Program
Arch Fam Med. 1999;8(2):107-109.
FULL TEXT
THIS ARTICLE HAS BEEN CITED BY OTHER ARTICLES
|
Women's Initial Experience of Abnormal Papanicolaou Smear
Mitchell and Hall
J Holist Nurs 2009;27:93-102.
ABSTRACT
Follow-Up of Abnormal Pap Smears Among Incarcerated Women
Clarke et al.
J Correct Health Care 2007;13:22-26.
ABSTRACT
Feasibility of Management of High-Grade Cervical Lesions in a Single Visit: A Randomized Controlled Trial
Brewster et al.
JAMA 2005;294:2182-2187.
ABSTRACT
| FULL TEXT
Cervical Cancer Prevention: Making Programs More Appropriate and Pragmatic
Blumenthal and Gaffikin
JAMA 2005;294:2225-2228.
FULL TEXT
Effect of patient socioeconomic status on perceptions of first- and second-year medical students
Woo et al.
CMAJ 2004;170:1915-1919.
ABSTRACT
| FULL TEXT
What Factors Are Associated with Diagnostic Follow-Up after Abnormal Mammograms? Findings from a U.S. National Survey
Yabroff et al.
Cancer Epidemiol. Biomarkers Prev. 2004;13:723-732.
ABSTRACT
| FULL TEXT
Women's Experiences of Abnormal Cervical Cytology: Illness Representations, Care Processes, and Outcomes
Karasz et al.
Ann Fam Med 2003;1:196-202.
ABSTRACT
| FULL TEXT
Is the Promise of Cancer-Screening Programs Being Compromised? Quality of Follow-Up Care after Abnormal Screening Results
Yabroff et al.
Med Care Res Rev 2003;60:294-331.
ABSTRACT
|
