Abstract
Objectives To investigate gender differences in the relationship between sleep and depression in the elderly.
Methods Residents of a village in Kumamoto Prefecture, Japan (563 people) aged≥65 years were given a self-administered questionnaire survey between June and July 2010. To evaluate levels of sleep and depression, the Pittsburgh Sleep Quality Index (PSQI) and the Geriatric Depression Scale—Short Form (GDS-SF) were used. Adjustment factors assessed age, income, residence status, social role, medical history, nursing situation, and cognitive function. As with previous studies, respondents were evaluated as having a disturbed sleep or depression if the sum of their PSQI and GDS-SF scores was 6 or more. We examined the relationship between sleep and depression using multiple logistic regression analysis, with presence of depression in each gender introduced as a dependent variable. Given that some degree of correlation was expected among adjustment factors in the model, we constructed a model that introduced the adjustment factors stepwise to confirm the robustness of the model and any effect of multicollinearity.
Results Overall (n=424), a significant relationship was found between disturbed sleep and depression in all models. The odds ratio was 1.92 (95% confidence interval: 1.11–3.32) in the final model, controlling all adjustment factors. In men (n=198), although the relationship between disturbed sleep and depression was significant before adjustment, no significance was observed after adjustment. In women (n=226), however, a significant relationship was observed between disturbed sleep and depression in both models. After the introduction of all variables as adjustment factors, the odds ratio was 2.28 (95% confidence interval: 1.11–4.69).
Conclusion Our findings suggest a significant relationship between disturbed sleep and depression in elderly women, but not in men. While the reasons for this gender difference are unclear at present, various factors, such as the endocrine system, are likely to be involved. Future studies should take this gender difference into account in order to enact more effective measures for preventing depression in the elderly.
