Ever having had a Pap smear significantly reduces the risk of cervical cancer, according to a study of women in South Africa.1 Compared with women who had never had the test, those who had ever had one had odds of cancer that were 70% lower. The reduction in risk decreased to 60% once 10 years had passed since the last Pap smear, and to 50% after 15 years. The protective effect increased from 60% for one smear to 80% for two or more.
Data for the analysis came from a case-control study on hormonal contraceptives and cervical cancer that had been conducted in the Western Cape between January 1998 and December 2001. The sample included 524 colored and black women younger than 60 who had received a diagnosis of cervical cancer at one of two area hospitals and 1,540 healthy controls matched for decade of age, race, area of residence and hospital. Trained nurses interviewed participants using a standard questionnaire that asked women how many Pap smears they had ever had and when the procedures had been performed. Researchers analyzed the data using logistic regression, controlling for age, area of residence, race, education, parity, age at first sexual activity, contraceptive use and tobacco use.
Among all women in the sample, 50% of those with cervical cancer and 73% of controls had ever had a Pap smear. The odds of developing the disease were 70% lower among women who had ever had the test than among those who had never had the procedure (odds ratio, 0.3). Moreover, the reduction in risk rose from 60% among women who had had one Pap smear (0.4) to 80% among those who had had two or more (0.2).
The protective effect of having had a Pap smear decreased as the time since the last screening increased, from 70% at durations of less than 10 years to 60% at 10-14 years and 50% at 15 years or more. Furthermore, age affected the effect of Pap smear screening on cervical cancer: Having been screened reduced the risk of cancer by 70% among women aged 30 or older (odds ratio, 0.3), but had no significant impact on the risk among younger women.
Among women with cervical cancer, 154 were in the early stages of the disease and 370 had advanced cases. Fifty-seven percent of women with early-stage disease had ever had a Pap smear, compared with 46% of women with late-stage disease. Compared with women in the control group, women with early-stage cancer had significantly reduced odds of having ever had a Pap smear (0.5); the odds were further reduced among those with late-stage cancer (0.3).
In light of their results, the researchers comment that Pap smears appear to be an effective measure to reduce the high incidence of cervical cancer among South African women. They remark that "Assuming the validity of the results, if all women had been screened under the conditions in the present study, an estimated 70% of cases of cervical cancer would have been prevented."
The author of an accompanying commentary, however, points out that this study shares several problems involved in using case-control methods to evaluate screening.2 She notes in particular that the study "depends totally on women's recollection," and argues that "It is highly likely that women tended to report to the interviewer what they perceived as the 'correct' health-related behaviour."--J. Rosenberg
1. Hoffman M et al., Limited Pap screening associated with reduced risk of cervical cancer in South Africa, International Journal of Epidemiology, 2003, 32(4):573-577.
2. Raffle AE, Case-control studies of screening should carry a health warning, International Journal of Epidemiology, 2003, 32(4):577-578.