Abstract

STUDY OBJECTIVE--The aim of the study was to evaluate the effectiveness of a centralised population based cervical cytology screening programme. DESIGN--The study was a case-control investigation. SETTING--Cases and controls were confined to the province of Florence. PARTICIPANTS--191 out of 208 cases of cervical cancer in women less than 75 years old at diagnosis in the period 1982-85 were interviewed. For each case three living controls were selected, strictly matched by year of birth and district of residence; in all 573 controls were eventually identified. Of these, 15 had had a hysterectomy (2.6%) and were excluded, and a further 18 (3.2%) did not take part for other reasons, leaving a total of 540 controls. MEASUREMENT AND RESULTS--Screening history was taken from a computerised archive for both cases and controls. A mail questionnaire was used to collect information on several potential confounding variables. For women screened only once in comparison with those never screened, the reduction in risk was about 70% (odds ratio 0.29. 95% confidence limits 0.15-0.55), while the reduction was even greater for those screened twice or more. No trend of increasing risk with increasing interval since last test was shown: considering separately women who had only had one test and those who had had two or more tests, the risk estimates were stable across different time intervals since the last test. CONCLUSIONS--There is a strong protective effect against developing invasive cervical cancer through participation in the screening programme.