Résumé
ABSTRACTBackground: Fluctuating hormone levels are known to influence a woman's mood and well-being. This study aimed to determine whether lifetime hormonal markers are associated with late-life depression symptoms among elderly community-dwelling women.Method: Detailed reproductive histories of 1013 women aged 65 years and over were obtained using questionnaires, and depressive symptoms were assessed using the Centre for Epidemiological Studies Depression Scale. Multivariate logistic regression models were generated to determine whether any lifetime endogenous or exogenous hormonal factors were associated with late-life depression.Results: The prevalence of depressive symptoms was 17%. Age at menopause was associated with depressive symptoms, but only among women with a lower education level. For these women, an earlier age at menopause increased their risk of late-life depression (linear effect, OR = 0.95, 95%CI: 0.91-0.99). The odds of late-life depression were also increased for women who were past (OR = 1.6, 95%CI: 1.1-2.5), but were not current users. On the other hand, long-term oral contraceptive use (>/=10 years) was protective against depression (OR = 0.3, 95%CI: 0.1-0.9). These associations remained significant even after extensive adjustment for a range of potential confounding factors, including sociodemographic factors, mental and physical incapacities, antidepressant use and past depression. The other factors examined - including age at first menses, parity, age at childbirth and surgical menopause - were not associated with late-life depressive symptoms.Conclusions: Lifetime hormonal factors that are significantly associated with depression symptoms in later life have been identified. Further work is needed to determine how potential hormonal interventions could be used in the treatment of late-life depression in certain subgroups of women.