RESUMO
OBJECTIVES: To examine associations of ever-use of sex hormones (EUSH) and other factors with endometrial cancer (EC) mortality through a nation-wide Japan Collaborative Cohort Study. METHODS: A total of 63,541 women aged 40-79 years, enrolled in 1988-90 from 45 municipalities of Japan, were followed until 2003 to record their vital status. Using baseline data, the Cox proportional hazard model (age adjusted and multivariate) was used to estimate the hazard ratio (HR) and 95% confidence interval (CI) for EC mortality by selected factors, including EUSH. Bivariate analysis was also conducted to establish associations between EUSH and other factors. RESULTS: The mortality rate from EC was 2.6 per 100,000 person-years during the mean follow-up period of 13.3 years. Prevalence rate of EUSH was 5.2%. Significantly increased risk of EC mortality was found for EUSH with both age adjusted (HR=6.43, 95%CI=2.10-19.67) and multivariate (HR=5.33; 95%CI=1.51-18.82) analyses. Bivariate analysis indicated that history of diabetes mellitus, smoking, drinking, and age at first delivery were positively associated with EUSH, whereas age, number of delivery, number of pregnancy, and age at menarche demonstrated inverse links. CONCLUSIONS: Our results imply that EUSH may increase the risk of EC mortality among Japanese women. However, further studies with more deaths are needed to validate the results.