Objectives: To analyze prospectively the association between hormone replacement therapy (HRT) and mortality in women before old age.
Design and methods: A group of 11,667 women (91% of the age cohort of the area) aged 52-62 years from the population-based Kuopio Osteoporosis Risk Factor and Prevention Study were followed for 7 years in 1994-2001. Information about HRT use and health events was obtained from two repeated questionnaires in 1989 and 1994. Information about deaths and causes of death from the follow-up period was obtained from the Statistics Finland. Cox's proportional-hazards models were used to calculate risk of death related to the use of HRT.
Results: At the start of follow-up, 2203 women had used HRT > 5 years, 3945 women < or = 5 years and 5519 women had never used it. During the follow-up, 361 deaths occurred. Compared with non-users of HRT, the adjusted hazard ratio (HR) of death from any cause was 1.05 (95% confidence interval (CI) 0.80-1.36) in women who used HRT < or = 5 years and 1.06 (95% CI 0.78-1.46) in women who used HRT > 5 years. The adjusted HR for coronary heart disease (CHD) mortality in women who used HRT < or = 5 years was 0.79 (95% CI 0.36-1.73), and in women who used HRT > 5 years, 2.16 (95% CI 0.93-4.98). For breast cancer mortality the adjusted HR for < or = 5 years of HRT use was 0.96 (95% CI 0.32-2.82) and 2.62 (95% CI 0.98-7.00) for > 5 years of HRT use.
Conclusions: History of HRT use does not affect overall or CHD mortality in women. More than 5 years of HRT use may increase the risk of breast cancer mortality.