In 1998, 465 women from 25 to 49 years of age, inhabitants of the municipality of São Luís, Maranhão, Brazil were interviewed to estimate Pap smear coverage, interval between tests and to identify factors associated with not performing Pap smears. A three-stage household cluster survey was performed; 95% confidence intervals corrected by design effect, and crude and adjusted odds ratio estimates were calculated by logistic regression. Pap smear coverage at least once in a lifetime was 82.4% (95% CI 76.6% - 87.0%) and nearly reached the minimum value of the 85% needed to impact the incidence and mortality of cervical cancer. The interval between tests was low, given that 65.8% of women repeated the test within one year. After adjustment for confounding factors, the risks of not having performed a Pap smear test at least once in a lifetime were higher for women who did not live with a companion, who had five to eight years of schooling, who had not been to a medical appointment in the past three months and who lived in households whose head of family was engaged in a manual and non-qualified occupation. Women who had two to four sexual partners in the past three months were at a lower risk of not having a cervical smear. In this Northeastern state capital, Pap smear coverage was similar to other national studies. However, some women were unnecessarily tested at short intervals, wasting resources and contributing to reduce the access of more vulnerable groups who presented higher risks for not having had a Pap smear test.
Cervix neoplasms; Vaginal smears; Cross-sectional studies; Brazil
