Advertisement

Screening for cervical cancer: Will women accept less?

      Background

      U.S. professional organizations increasingly agree that most women require Papanicolaou smear screening every 2 to 3 years rather than annually and that most elderly women may stop screening. We sought to describe the attitudes of women in the United States toward less intense screening, specifically, less frequent screening and eventual cessation of screening.

      Methods

      We conducted a random-digit-dialing telephone survey of women in 2002 (response rate of 75% among eligible women reached by telephone). A nationally representative sample of 360 women aged 40 years or older with no history of cancer was surveyed about their acceptance of less intense screening.

      Results

      Almost all women aged 40 years or older (99%) had had at least one Pap smear; most (59%) were screened annually. When women were asked to choose their preferred frequency for screening, 75% preferred screening at least annually (12% chose screening every 6 months). Less than half (43%) had heard of recommendations advocating less frequent screening. When advised of such recommendations, half of all women believed that they were based on cost. Sixty-nine percent said that they would try to continue being screened annually even if their doctors recommended less frequent screening and advised them of comparable benefits. Only 35% of women thought that there might come a time when they would stop getting Pap smears; of these, almost half would not stop until after age 80 years. The strongest predictor of reluctance to reduce the frequency of screening was a belief that cost was the basis of current screening frequency recommendations.

      Conclusion

      Most women in the United States prefer annual Pap smears and are resistant to the idea of less intense screening. Concern that cost considerations rather than evidence form the basis of screening recommendations may partly explain women’s reluctance to accept less intense screening.

      Keywords

      When Papanicolaou screening for cervical cancer was introduced in the United States in 1941,
      • Papanicolaou G.N.
      • Traut F.H.
      The diagnostic value of vaginal smears in carcinoma of the uterus.
      it became the first systematic effort undertaken to detect cancer early. Since then, vast numbers of women in the United States have participated in what has become an annual screening rite. In its enormous success—both in terms of its widespread acceptance and the sustained reduction in cervical cancer mortality that followed—the Pap smear has come to exemplify a model cancer screening test against which other such tests are now measured. More recently, however, long-accepted Pap smear screening standards have been re-examined.
      Specifically, there have been questions about the optimal intensity of screening for cervical cancer—how often women should be screened, and until what age. Although initially instituted in the United States as an annual test, abundant evidence from large observational studies over the past two decades has led to the widely accepted conclusion that annual screening confers a trivial, if any, advantage over triennial screening.
      IARC Working Group on Evaluation of Cervical Cancer Screening Programmes
      Screening for squamous cervical cancer duration of low risk after negative results of cervical cytology and its implication for screening policies.
      • Sawaya G.F.
      • McConnell J.
      • Kulasingam S.L.
      • et al.
      Risk of cervical cancer associated with extending the interval between cervical-cancer screenings.
      More frequent screening may in fact lead to worse health outcomes, due to the greater number of abnormalities that require intervention.
      Incidence of Pap test abnormalities within 3 years of a normal Pap test—United States, 1991–1998 [editorial note].
      • Saslow D.
      • Runowicz C.D.
      • Solomon D.
      • et al.
      American Cancer Society guideline for the early detection of cervical neoplasia and cancer.
      Other work has suggested negligible benefits of screening elderly women with a history of regular Pap smears.
      • Fahs M.C.
      • Mandelblatt J.
      • Schecter C.
      • Muller C.
      Cost effectiveness of cervical cancer screening for the elderly.
      • Sawaya G.F.
      • Grady D.
      • Kerlikowske K.
      • et al.
      The positive predictive value of cervical smears in previously screened postmenopausal women the Heart and Estrogen/progestin Replacement Study (HERS).
      • Sherlaw-Jones C.
      • Gallivan S.
      • Jenkins D.
      Withdrawing low risk women from cervical screening programmes mathematical modelling study.
      Recommendations in favor of less intense screening, jointly issued in 1988 by all major U.S. professional organizations, have recently been strengthened. In the past 2 years, the U.S. Preventive Services Task Force and the American Cancer Society both clearly advocated less frequent than annual screening for most women,
      • Saslow D.
      • Runowicz C.D.
      • Solomon D.
      • et al.
      American Cancer Society guideline for the early detection of cervical neoplasia and cancer.
      and issued explicit recommendations that elderly women may cease routine screening (at 65 years for the Task Force and 70 years for the American Cancer Society).
      Although guidelines increasingly support less intense screening, survey data show that the majority of U.S. women undergo screening annually and continue periodic screening into old age.
      • Sirovich B.E.
      • Welch H.G.
      The frequency of Pap smear screening in the United States.
      We conducted a survey to learn whether women would be willing to accept a reduction in screening intensity.

      Methods

      Sample

      As part of a larger project examining colon, breast, and prostate cancer screening,
      • Schwartz L.M.
      • Woloshin S.
      • Fowler F.J.
      • Welch H.G.
      Enthusiasm for cancer screening in the United States.
      we interviewed a nationally representative sample of women in the United States aged 40 years or older to explore their attitudes about and experiences with screening for cervical cancer. To focus on screening, we excluded women with a history of any non–skin cancer. This project was approved by the institutional review boards at Dartmouth Medical School and the University of Massachusetts.
      We used random digit dialing to obtain a national probability sample of households in the continental United States with telephone service. Aiming for a goal of 500 survey participants, we used a randomly generated list of 4000 telephone numbers, which yielded 1702 working residential telephone numbers. After multiple attempts to each number, interviewers succeeded at 1208 of the residences in completing a 3-minute “screening” interview with an English-speaking adult to identify subjects meeting the study criteria (i.e., a woman aged 40 years or older, or man aged 50 years or older, with no history of any non–skin cancer). At least 1 eligible adult resided in 697 of the households (if more than 1 eligible adult was identified, a computer randomly selected a respondent so that each eligible person had an equal chance of being selected). Of the 500 subjects who completed the interview, 360 were women. The overall response rate, calculated as the number of responses obtained divided by the number of eligible participants reached by telephone, was 72% (500/697). The rate for women was 75% (360/480).
      The distribution of most demographic characteristics in our sample was very similar to those of the U.S. female population aged 40 years or older (Table 1), with the exception of modest differences for age (i.e., the elderly were underrepresented) and education (i.e., persons who were less well educated were underrepresented).
      Table 1Demographic characteristics of the study participants (n = 360) and United States census estimates for women aged 40 years or older
      CharacteristicSample population n/N
      Number of survey respondents in each category over the number who answered the question; item nonresponse 0% to 3%.
      (weighted %)
      Percentages are weighted estimates to account for the sampling strategy and are therefore not simply the ratio of numerator to denominator; percentages may not add to 100% due to rounding.
      2000 U.S. census (%)P value
      P value for the one sample test of proportions comparing the weighted sample population percentage with that of the 2000 U.S. Census.
      Age (years)
       40–49155/355 (38)350.18
       50–59102/355 (31)260.04
       60–6942/355 (14)170.18
       ≥7056/355 (16)230.004
      Race
       White275/350 (79)840.006
       Black43/350 (12)110.34
       Asian11/350 (4)30.45
       Native American/Alaskan Native8/350 (2)10.04
       Other/multiracial13/350 (3)1<0.001
       Hispanic17/354 (5)50.73
      Education
       Less than high school graduate17/357 (5)15<0.001
       High school graduate99/357 (31)370.02
       Some college114/357 (30)270.25
       College graduate or more127/357 (34)21<0.001
      Region
       Northeast68/360 (18)200.51
       South137/360 (38)360.44
       Midwest90/360 (27)240.21
       West65/360 (17)200.11
      * Number of survey respondents in each category over the number who answered the question; item nonresponse 0% to 3%.
      Percentages are weighted estimates to account for the sampling strategy and are therefore not simply the ratio of numerator to denominator; percentages may not add to 100% due to rounding.
      P value for the one sample test of proportions comparing the weighted sample population percentage with that of the 2000 U.S. Census.

      Interview protocol

      To learn how the public thinks about screening, we conducted focus groups with adults over 40 years old to discuss cancer screening tests in general, experiences with specific tests, and perceptions of the pros and cons of such testing. The focus groups and all subsequent survey development were done in collaboration with experts at the Center for Survey Research, a professional survey research firm affiliated with the University of Massachusetts Boston. A draft survey instrument was developed based on the results of the focus groups and revised based on the results of 10 cognitive interviews, which were conducted to ensure that the questions were understood and the answers meaningful. The survey was then piloted on 17 eligible adults identified by random digit dialing; interviews were audiotaped to identify questions that were difficult for interviewers to read or for respondents to answer. Final survey revisions were made based on these results.
      The survey consisted of five modules: a general screening module (general questions about the value of early detection) and four cancer-specific modules (covering breast, cervical, prostate, and colorectal cancer screening). The cervical cancer screening module, completed by all women, included questions about women’s screening experiences, current screening practices, history of abnormal test results, testing preferences (e.g., frequency, starting and stopping age), awareness of screening recommendations, risk perceptions about cervical cancer, and willingness to accept less intense screening.
      From December 2001 to July 2002, interviews were conducted by trained, professional interviewers from the University of Massachusetts Center for Survey Research. All interviews took place before the issuance of the most recent U.S. Preventive Services Task Force and American Cancer Society screening guidelines. Interviewers received special training on the purposes and procedures of this particular study; all underwent routine monitoring for quality control and feedback from a supervisor. The interviews took an average of 20 minutes (range, 17 to 54 minutes).

      Statistical analysis

      A subject’s probability of selection was a function of the number of residential telephone lines (i.e., more telephone lines, higher probability of selection) and the number of eligible adults at the residence (i.e., more eligible adults, lower probability of selection). We created weights for each subject surveyed to account for these differential selection probabilities. All percentages reported are weighted to reflect selection probability, using the SVY command series in STATA, version 7.0 (Stata Corporation, College Station, Texas). When the number of respondents in the numerator and denominator are reported next to the percentage, the ratio of these numbers (i.e., the crude percentage) may not equal the weighted percentage. To compare the likelihood of different subgroups of women accepting less intense screening, we used the Pearson chi-squared statistic adjusted for the survey design; for subgroups for which order was relevant (e.g., age groups), we used logistic regression for complex survey data. All reported P values are based on two-sided tests.

      Results

      Screening experience, practices, and risk perception

      Nearly all women (99%) had had at least one Pap smear (Table 2). The majority (59%) reported undergoing screening annually or more often; only 10% were being screened every 2 to 3 years, the recommended interval for average-risk women. Thirty percent had had at least one smear result that required further testing. Of these women, almost half (43%) described the additional testing experience as “very scary” or the “scariest time” of their lives. Half had waited more than 2 weeks to find out they did not have cancer (range, 0 to 24 weeks). Nevertheless, over 99% (102/103) were glad they had undergone screening, even among those for whom the experience was at least “very scary.”
      Table 2Papanicolaou smear screening experiences and risk perception among women in the study (n = 360)
      n/N
      Number of survey respondents in each category over the number who answered the question; item nonresponse 0% to 4%.
      (%
      Percentages are weighted estimates to account for the sampling strategy and are therefore not simply the ratio of numerator to denominator; percentages may not add to 100% due to rounding.
      )
      Experiences
      Except for questions about ever having had a Pap smear or a hysterectomy, Pap smear experience questions were asked only to women who had had at least one smear (n = 355).
      Have had at least one smear355/359 (99)
      Current smear screening schedule
       Yearly or more often209/353 (59)
       Every 2 to 3 years35/353 (10)
       No specific smear schedule37/353 (11)
       Last smear more than 3 years ago72/353 (21)
      Smear-associated discomfort: 0 (not uncomfortable) to 10 (extremely uncomfortable)
       040/353 (11)
       1–3135/353 (39)
       4–7143/353 (40)
       8–1035/353 (10)
      Have had a smear result that required further testing103/355 (30)
       If so, how scary was the experience?
        Not scary at all5/102 (5)
        A little scary30/102 (28)
        Somewhat scary24/102 (25)
        Very scary33/102 (32)
        The scariest time of my life10/102 (11)
      Have had a hysterectomy112/358 (32)
      Risk perception
      Have known someone who has had cervical cancer143/357 (39)
      How many women your age diagnosed with cervical cancer do you think will die within 5 years of diagnosis?
       Most49/345 (13)
       Some/a few264/345 (78)
       Almost none32/345 (9)
      Compared with other women your age, what do you think your risk of developing cervical cancer in the next 10 years is:
       Higher than average23/353 (7)
       Average247/353 (70)
       Lower than average83/353 (23)
      * Number of survey respondents in each category over the number who answered the question; item nonresponse 0% to 4%.
      Percentages are weighted estimates to account for the sampling strategy and are therefore not simply the ratio of numerator to denominator; percentages may not add to 100% due to rounding.
      Except for questions about ever having had a Pap smear or a hysterectomy, Pap smear experience questions were asked only to women who had had at least one smear (n = 355).
      Women did not appear to have an inflated sense of their own risk of developing the disease—just 7% judged themselves to be at above-average risk (Table 2). Women were also realistic about the lethality of the disease. When asked how many women their age treated for cervical cancer would die within 5 years, more than three quarters responded “some” or “a few” (for context, about 30% are expected to die within 5 years).

      Attitudes toward screening intensity and recommendations

      When women were asked to select their preferred screening frequency, independent of cost, 75% chose to be screened annually or more often (12% preferred screening every 6 months) (Table 3). Forty-three percent had heard or read of recommendations in favor of screening every 2 or 3 years. After being informed about such recommendations, women indicated their beliefs about each of two potential rationales behind less frequent screening. Forty percent thought that less frequent screening was recommended because scientific evidence shows that performing Pap smears less often saves as many lives as performing them every year. Half thought that the recommendations were made because of cost.
      Table 3Papanicolaou smear screening preferences, knowledge, and attitudes among women in the study (n = 360)
      n/N
      Number of survey respondents in each category over the number who answered the question; item nonresponse 0% to 9%.
      (%
      Percentages are weighted estimates to account for the sampling strategy and are therefore not simply the ratio of numerator to denominator; percentages may not add to 100% due to rounding.
      )
      Preferences
      Pap smear preference questions were asked only to women who had had at least one smear (n = 355).
      If cost was not a concern, would you prefer to be screened:
       Every 6 months or more often48/349 (12)
       Every year220/349 (63)
       Every 2 years or less often81/349 (25)
      Do you think there might be a time when you will stop being screened?
       No226/340 (65)
       Yes114/340 (35)
        If so, when?
         Before age 70 years34/104 (32)
         Between age 70 and 80 years26/104 (24)
         After age 80 years44/104 (44)
      How would you respond if your doctor told you that the benefit of screening would be the same if you had them less often than you do now, and recommended that you have them less often?
      Includes only the 209 women currently undergoing annual or more frequent screening.
       I would try to keep being screened as often as I do now148/209 (69)
       I would agree to have screening less often61/209 (31)
      Knowledge and attitudes
      Have heard or read about recommendations in favor of getting screened every 2 to 3 years instead of every year149/358 (43)
      Think recommendations in favor of screening every 2 or 3 years are based on scientific evidence122/329 (40)
      Think recommendations in favor of Pap smears every 2 or 3 years are based on cost171/343 (50)
      * Number of survey respondents in each category over the number who answered the question; item nonresponse 0% to 9%.
      Percentages are weighted estimates to account for the sampling strategy and are therefore not simply the ratio of numerator to denominator; percentages may not add to 100% due to rounding.
      Pap smear preference questions were asked only to women who had had at least one smear (n = 355).
      § Includes only the 209 women currently undergoing annual or more frequent screening.

      Acceptance of stopping among all women screened

      Of all women who had ever undergone screening, 65% thought that there would never come a time when they would stop having Pap smears. Of the minority of women who might someday stop, almost half would not stop until after age 80 years (Table 3).
      Several factors were associated with a woman’s intent to continue screening indefinitely (Figure 1). Younger women were more likely than older women to think that they would never stop screening. Women’s risk perception also predicted their intention to continue screening indefinitely: women with higher self-described risk of cervical cancer, women who had known someone with cervical cancer, and women who had required further testing after a Pap smear were all more likely to believe that they would never stop screening. Women currently screened annually (or more often) were much more likely to think that they would never stop screening than those screened less frequently (79% [160/206] vs. 46% [65/132], P <0.0001).
      Figure thumbnail gr1
      Figure 1Percentage (number) of women who say they will never stop getting Pap smears, according to age, risk perception, and perceived motivation for screening guidelines. Data presented include women who have been screened at least once. Percentages are weighted estimates to account for the sampling strategy and are therefore not simply the ratio of numerator to denominator. Percentages may not add to 100% due to rounding. An asterisk denotes that the difference between groups is significant at P <0.05.

      Acceptance of less frequent screening among women screened annually

      Because recommendations for reducing screening frequency apply only to women presently undergoing annual or more frequent screening, we focused on these women (59% of those ever screened) in assessing acceptance of less frequent screening. Each woman was asked how she would respond if her doctor recommended that she be screened less often and advised her of equal benefits with less frequent screening. Among women currently being screened annually or more often, more than two thirds (69%) would try to keep having Papanicolaou tests as often as they do now rather than agreeing to a decrease (Table 3).
      A number of factors predicted lack of acceptance of less frequent screening (Figure 2). Younger women were more likely than older women to try to continue annual (or more frequent) screening. Risk perception also played a role. Women who had known someone with cervical cancer were more likely to try to continue annual screening than were women who had not. Prior knowledge of screening guidelines was not associated with the likelihood of rejecting less frequent screening (69% [56/77] of those who had heard of the guideline would try to continue annual screening vs. 68% [91/131] of those who had not, P = 0.92). Women who believed that cost considerations underlie screening recommendations were much more likely than other women to try to continue annual screening, whereas women who believed that the recommendations were based on scientific evidence were more accepting of less frequent screening. (The responses of the small number of women who believed that the recommendations were based on both cost and evidence mirrored those of women who believed that they were based on cost alone.)
      Figure thumbnail gr2
      Figure 2Percentage (number) of women who say they would try to keep having Pap smears at their current screening frequency despite a physician’s recommendation to have them less often, according to age, risk perception, and perceived motivation for screening guidelines. Data presented include women currently screened annually or more often. Percentages are weighted estimates to account for the sampling strategy and are therefore not simply the ratio of numerator to denominator. Percentages may not add to 100% due to rounding. An asterisk indicates that the difference between groups is significant at P <0.05.

      Attitudes among women who have undergone hysterectomy

      Women who had had a hysterectomy—most of whom are not at risk of cervical cancer—were as likely as those who had not had the procedure to plan to continue screening indefinitely (59% [61/103] vs. 68% [163/235] of women without a hysterectomy, P = 0.14), and equally likely to try to continue annual screening (66% [36/52] vs. 70% [111/156], among women currently screened annually or more often, P = 0.63). In addition, stratification of the analyses described above (of predictors of rejecting less intense screening) by hysterectomy status revealed almost no differences between the two groups.

      Discussion

      We found that more than half of women in the United States undergo Pap smear screening for cervical cancer at least once a year, and even more would choose annual screening if offered the choice. Even with a recommendation from their doctors, less than one third of women presently undergoing annual screening would agree to be screened less often, despite comparable benefits with less frequent screening. Most women could not imagine a time later in life when they would stop undergoing screening. Even women who are largely not at risk of cervical cancer—those who have undergone hysterectomy—are reluctant to accept less intense screening. The strongest predictors of lack of acceptance of less intense screening were concern that screening recommendations were based on cost, and a current pattern of annual or more frequent screening.
      Our study has several limitations. First, our survey was conducted by telephone, and conclusions can apply only to the approximately 95% of the U.S. population residing in households with a telephone. Second, because our survey included only women aged 40 years or older, we are unable to report the screening attitudes of younger women, who presently undergo screening at the highest rates.
      • Sirovich B.E.
      • Welch H.G.
      The frequency of Pap smear screening in the United States.
      Our age restriction was imposed because our survey also focused on screening for three other cancers that are not relevant to the population under 40 years old. However, older women are precisely the group for whom decreased screening intensity is most relevant—both because they are least likely to have had a recent abnormal smear
      • Rolnick S.
      • Laferla J.J.
      • Wehrle D.
      • et al.
      Pap smear screening in a health maintenance organization 1986–1990.
      • Raffle A.E.
      • Alden B.
      • Mackenzie E.F.D.
      Detection rates for abnormal cervical smears what are we screening for?.
      and because they are closest to a recommended stopping age. Lastly, some may be concerned about the generalizability of our results, based on our response rate of 75% (among women known to be eligible to participate). If those not responding to the survey are less likely to be screened, then we may have overestimated the extent and frequency of screening in the population and, possibly, women’s preferences for screening intensity. Any such bias, however, is likely to be small for two reasons. First, based on the results of large federal surveys, our sample is highly representative of U.S. women, both in terms of demographic characteristics, and Papanicolaou smear screening history (using the National Health Interview Survey, we have found that 96% of women aged 40 years or older had had a Papanicolaou smear, 29% had had a hysterectomy, and 53% were undergoing annual screening).
      National Center for Health Statistics (2002)
      Second, when we repeated our analysis, forcing our sample to mirror the demographic characteristics of the U.S. population (using so-called “poststratification weights”), our results changed only minimally.
      Despite the broad agreement among professional organizations in support of reduced Pap smear screening intensity (both frequency and duration), U.S. women clearly do not welcome less intense screening. Our findings add to previous work documenting general reluctance to undergo less screening. A 1995 change in a Minnesota health maintenance organization’s Pap smear guideline from routine annual to triennial screening led to skeptical or negative reactions by the majority of female members surveyed, particularly those who had been undergoing annual screening.
      • Rolnick S.J.
      • LaFerla J.J.
      • Jackson J.
      • et al.
      Impact of a new cervical Pap smear screening guideline on member perceptions and comfort levels.
      Attitudes toward colon cancer screening have been similar—preliminary data show that almost three quarters of elderly persons surveyed planned to continue screening for the rest of their lives; more than a third would do so even if their doctor recommended against it.
      • Lewis C.L.
      • Watson L.C.
      • Kistler C.E.
      • et al.
      Older adults’ perceptions of colon cancer screening decisions in late life.
      Our study has allowed us to expand on previous work by exploring possible explanations for the observed reluctance to accept less intense screening. We suspected that, to some extent, reluctance would be due to lack of awareness of screening guidelines. We did not find this to be the case. Although we found less than half of women knew of guidelines recommending less frequent screening, knowledge of the guidelines was not associated with acceptance of less intense screening. We did find evidence supporting the role of habit: women being screened frequently (annually or more often) were extremely likely to plan to do so indefinitely. Beyond this, we found that the two most powerful drivers of attitudes toward screening intensity were a woman’s perceived risk of cervical cancer and her perceived rationale for the screening recommendations.
      Although not surprising, our finding that women’s perceived risk of cervical cancer is a major determinant of their reluctance to reduce screening intensity has important implications. Risk perception itself is determined by a number of factors—in our data, one of the most powerful determinants of higher perceived risk of cervical cancer was having had an abnormal test result. Between 20% and 30% of women fall into this group,
      National Center for Health Statistics (2002)
      the overwhelming majority of whom will never have cervical cancer or a lesion that would progress to cancer. Because the frequency of Pap smear testing determines the annual volume of abnormal smears (and the number of women who will have one),
      • Saslow D.
      • Runowicz C.D.
      • Solomon D.
      • et al.
      American Cancer Society guideline for the early detection of cervical neoplasia and cancer.
      • Sirovich B.E.
      • Welch H.G.
      The frequency of Pap smear screening in the United States.
      • Colditz G.A.
      • Hoaglin D.C.
      • Berkey C.S.
      Cancer incidence and mortality the priority of screening frequency and population coverage.
      a self-fulfilling cycle ensues: more annual screening leads to more abnormal results, and more abnormal results lead to more women with increased risk perception who engage in more annual screening (and plan never to stop). Reducing high-intensity screening is therefore particularly difficult. It involves interrupting this positive feedback cycle
      • Ransohoff D.F.
      • McNaughton Collins M.
      • Fowler F.J.
      Why is prostate cancer screening so common when the evidence is so uncertain? A system without negative feedback.
      in addition to breaking the ingrained habit of annual screening.
      The strongest predictor of a woman’s not accepting a reduction in screening frequency was her belief that cost considerations rather than evidence drive screening recommendations. Moreover, such skepticism about the motivation for screening guidelines also tended to predict a woman’s intention to never stop being screened. Thus, any public education campaigns regarding changes in screening guidelines should be explicit about the evidence guiding those revisions, and what role, if any, cost considerations had in formulating the recommendations. In particular, such programs should mention the possibility that higher-intensity screening may lead to worse health outcomes—including unnecessary interventions for low-grade lesions and psychological distress.
      Incidence of Pap test abnormalities within 3 years of a normal Pap test—United States, 1991–1998 [editorial note].
      • Saslow D.
      • Runowicz C.D.
      • Solomon D.
      • et al.
      American Cancer Society guideline for the early detection of cervical neoplasia and cancer.
      Although our findings may be useful in guiding educational efforts aimed at women and their physicians, they also point out how difficult it will likely be to gain acceptance for less intense screening. Even among women who believed that screening recommendations are based solely on scientific evidence, just over half would agree to give up their annual screening habit, and even fewer think they would eventually stop screening.

      Acknowledgment

      We thank Floyd J. Fowler, PhD, and his team at the Center for Survey Research in Boston, Massachusetts, for their expert assistance in designing, evaluating, and conducting the survey. We also once again express our gratitude to H. Gilbert Welch.

      References

        • Papanicolaou G.N.
        • Traut F.H.
        The diagnostic value of vaginal smears in carcinoma of the uterus.
        Am J Obstet Gynecol. 1941; 42: 193-206
        • IARC Working Group on Evaluation of Cervical Cancer Screening Programmes
        Screening for squamous cervical cancer.
        BMJ. 1986; 293: 659-664
        • Sawaya G.F.
        • McConnell J.
        • Kulasingam S.L.
        • et al.
        Risk of cervical cancer associated with extending the interval between cervical-cancer screenings.
        N Engl J Med. 2003; 349: 1501-1509
      1. Incidence of Pap test abnormalities within 3 years of a normal Pap test—United States, 1991–1998 [editorial note].
        MMWR Morb Mortal Wkly Rep. 2000; 49: 1001-1003
        • Saslow D.
        • Runowicz C.D.
        • Solomon D.
        • et al.
        American Cancer Society guideline for the early detection of cervical neoplasia and cancer.
        CA Cancer J Clin. 2002; 52: 342-362
        • Fahs M.C.
        • Mandelblatt J.
        • Schecter C.
        • Muller C.
        Cost effectiveness of cervical cancer screening for the elderly.
        Ann Intern Med. 1992; 117: 520-527
        • Sawaya G.F.
        • Grady D.
        • Kerlikowske K.
        • et al.
        The positive predictive value of cervical smears in previously screened postmenopausal women.
        Ann Intern Med. 2000; 133: 942-950
        • Sherlaw-Jones C.
        • Gallivan S.
        • Jenkins D.
        Withdrawing low risk women from cervical screening programmes.
        BMJ. 1999; 318: 356-361
      2. Guide to Clinical Preventive Services, Third Edition: Periodic Updates. Agency for Healthcare Research and Quality, Rockville, Maryland2004 (AHRQ Publication No. 04-IP003. January. Available at: http://www.ahrq.gov/clinic/uspstf/uspscerv.htm Accessed October 22, 2004)
        • Sirovich B.E.
        • Welch H.G.
        The frequency of Pap smear screening in the United States.
        J Gen Intern Med. 2004; 19: 243-250
        • Schwartz L.M.
        • Woloshin S.
        • Fowler F.J.
        • Welch H.G.
        Enthusiasm for cancer screening in the United States.
        JAMA. 2004; 291: 71-78
        • Rolnick S.
        • Laferla J.J.
        • Wehrle D.
        • et al.
        Pap smear screening in a health maintenance organization.
        Prev Med. 1996; 25: 156-161
        • Raffle A.E.
        • Alden B.
        • Mackenzie E.F.D.
        Detection rates for abnormal cervical smears.
        Lancet. 1995; 345: 1469-1473
        • National Center for Health Statistics (2002)
        Data File Documentation, National Health Interview Survey, 2000. National Center for Health Statistics, Hyattsville, Maryland2000 (machine readable data file and documentation) Available at: ftp://ftp.cdc.gov/pub/Health_Statistics/NCHS/Datasets/NHIS/2000/
        • Rolnick S.J.
        • LaFerla J.J.
        • Jackson J.
        • et al.
        Impact of a new cervical Pap smear screening guideline on member perceptions and comfort levels.
        Prev Med. 1999; 28: 530-534
        • Lewis C.L.
        • Watson L.C.
        • Kistler C.E.
        • et al.
        Older adults’ perceptions of colon cancer screening decisions in late life.
        J Gen Intern Med. 2003; 18: 195
        • Colditz G.A.
        • Hoaglin D.C.
        • Berkey C.S.
        Cancer incidence and mortality.
        Milbank Q. 1997; 75: 147-173
        • Ransohoff D.F.
        • McNaughton Collins M.
        • Fowler F.J.
        Why is prostate cancer screening so common when the evidence is so uncertain? A system without negative feedback.
        Am J Med. 2002; 113: 663-667