Hormone therapy and mortality during a 14‐year follow‐up of 14 324 Norwegian women

OBJECTIVES: We evaluated mortality from cardiovascular disease (CVD), coronary heart disease (CHD) and all causes in relation to use of any hormone therapy (HT) and HT with oestradiol and norethisterone or levonorgestrel. DESIGN: Population-based cohort study. SETTING AND SUBJECTS: Women in three Norwegian counties were invited to a health survey in 1985-88 and 82.8% participated. In all 14 324 post- or perimenopausal women aged 35-62 years, including 702 HT users with a mean age of 48.8 years, were followed for 14 years. RESULTS: Women using HT had mortality from all causes and CVD comparable with that of nonusers. The relative risk (RRs) for CVD mortality amongst all women were 0.69 (95% CI: 0.35-1.33) for users of HT, and 0.96 (95% CI: 0.43-2.17) for users of HT with norethisterone or levonorgestrel. Amongst women free of self-reported cardiovascular health problems at baseline all-cause, CVD and CHD mortality tended to be lower amongst users of HT whilst HT use was linked with increased mortality amongst women with cardiovascular health problems. CONCLUSIONS: In this cohort of women around the usual age of menopause all-cause or CVD mortality amongst users of HT, most often oestradiol combined with norethisterone or levonorgestrel, was not markedly different from that of nonusers. Early CHD events amongst HT users prior to the baseline survey, together with selective inclusion of healthy subjects, may in part explain protective effects of HT on CHD reported from previous observational studies.