ISEE 2007 CONFERENCE ABSTRACTS SUPPLEMENT: Abstracts
The prevalence of risk factors associated with cervical cancer may change along time as a consequence of behavioral and/or cultural determinants. This study aimed to determine the prevalence of epidemiologic and clinical aspects related to cervical cancer development in 3 different birth cohorts of Brazilian women in the state of Rio de Janeiro.
Material and Methods:
From October 2004 to May 2006 a cross-sectional study was carried out among 318 Brazilian women who have searched health care in an oncologic reference hospital in Rio de Janeiro, showing histologic diagnosis of negative, precancer, or cancer lesion. Among the enrolled participants, 136 women were born between 1976 and 1988, 138 between 1956 and 1975, and 44 between 1937 and 1955. Antecedents of exposures to known cervical cancer risk factors and clinical characteristics were ascertained through an interview-administered questionnaire; colposcopy and lesion excision were carried out, when indicated. Crude and adjusted prevalence ratios of the studied variables in the different cohorts (1937–1955, 1956–1975, and 1976–1988), with the respective confidence intervals, were calculated using Poisson regression analysis.
Compared with women with negative results, HSIL/cancer patients showed the following prevalence ratios for selected risk factors in the 1976 to 1988 cohort: number of pregnancies (adjusted PR: 1.16, 95% CI 1.01–1.32) and menarche <12 years (adjusted PR: 1.95, 95% CI 1.17–3.25) were statistically associated with cervical cancer in the 1976 to 1988 cohort. Menopause age (adjusted PR: 1.21, 95% CI 1.04–1.41) and current tobacco smoking (adjusted PR: 1.37, 95% CI 1.10–1.37) were independent risk factors for cervical cancer in the 1956 to 1975 cohort.
Risk factors for HSIL/cervical cancer seem to have played different risks among the studied Brazilian women from the 3 different cohorts (1938–1955, 1956–1975, and 1976–1988). The observed results suggest that cervical cancer prevention and early detection programs should have different approaches according to different birth cohorts to be more effective.