Epidemiology of cervical cancer in the screened population in Taiwan, 1979-1984.

This study was based on the screening data of April 1979-June 1984 to provide descriptive epidemiology of cervical cancer in Taiwan. During the 63 months, 175,823 women were screened and 1988 (1.13%) suspicious cases were found. The predictive value of a positive test was 50.14% and the specificity of the screening test was 99.44%. The estimated cancer rate among the screened population was 0.57%. The ratio of carcinoma in situ to invasive carcinoma was 1.91. The mean ages were 37.5 years for screened population, 44.1 years for suspects, 46.3 years for cancer cases (44.1 years for carcinoma in situ and 50.8 years for invasive carcinoma), 57.3 years for cervical cancer deaths in 1981 and 58.1 years for uterine cancer deaths in 1981. Univariate descriptive analyses on all women screened revealed the following variables to be positively associated (p less than 0.05) with suspicious Pap smears and cancer confirmed cases: age, number of years married, number of full-term deliveries, number of abortions/miscarriages, and abnormal vaginal bleeding. Negative associations were found with age at first marriage, education level, previous Pap smear and use of condom for contraception. Suspicious rates (but not detected cancer rates) were significantly higher for those women who themselves or whose husbands worked in mining, salt production, fishery or agriculture occupations. They were lowest for those (self or husband) employed in civil service, teaching, military service or police. By residence, suspicious rates were highest for those residing in the most urbanized areas (cities) and lowest for those in the second most urbanized areas (country cities). Thus urbanization level was not significantly involved. Other risk variables reported in the literature such as family history of uterine and other cancers were not significantly associated with either the suspicious or the cancer rate in this study.