Risk Factors for Death among the Functionally Independent Elderly Living in Japan: A 3-Year Prospective Cohort Study

To examine risk factors for death in Japanese elderly men and women who live ordinary lives in a community, we performed a prospective cohort study in Kishimoto Town, Tottori Prefecture, Japan. In September 2001, the town population of the elderly who were 65 years and older and who lived functionally independently with or without family was 1383. We delivered questionnaires to them, and collected 1313 (94.9%) answers. Of the 1313, 29 were highly suggestive to dependent living. Subjects of the baseline survey consisted of 1284 independent elderly, excluding the 29. The subjects themselves reported answers to our questionnaire items concerning demographic variables (age, family), history of disease, homebound condition (place of daily activities), activity of daily living (walking, excreting), lifestyle (tobacco), psychology and mental status (subjective health), quality of life (domestic role) and physical status (body pain, experience of falling). They were followed-up for 3 years until August 2004. We examined correlations between their deaths during the follow-up (total number of deaths, 79: 49 men and 30 women) and potential factors for death using the Cox proportional hazard model. By the multivariate analysis of these variables, we observed that elderly men had four risk factors for death: age (hazard ratio and 95% con?dence interval; 1.09 and 1.04?1.14), subjective health (2.45 and 1.40?4.30), domestic role (2.21 and 1.22?4.01) and tobacco (1.96 and 1.10?3.48). Elderly women had two risk factors, age (1.13 and 1.07?1.20) and physical ability or skill in handling banking duties (one of competence indexes for elderly activities of daily living) (2.45 and 1.12?5.39). Most noticeably, in the present elderly Japanese living functionally independently, death was signi?cantly correlated with psychosocial factors (subjective health, domestic role) rather than physical factors (restriction of going outside due to incontinence).