Social networks and mortality based on the Komo-Ise cohort study in Japan.

Background: No prospective studies have examined the association between social networks and all-cause and cause-specific mortality among middle-aged Japanese. The study of varied populations may contribute to clarifying the robustness of the observed effects of social networks and extend their generalizability.

Methods: To clarify the association between social networks and mortality among middle-aged and elderly Japanese, a community-based prospective study, the Komo-Ise Study, was conducted in two areas of Gunma Prefecture, Japan. A total of 11 565 subjects aged 40-69 years at baseline in 1993 completed a self-administered questionnaire. During the 7-year follow-up period, 335 men and 155 women died and the relative risk (RR) of each social network item was estimated by the Cox proportional hazard model.

Results: Single women had significantly increased risks of all-cause (multivariate RR = 2.2), and all circulatory system disease (age-area adjusted RR = 2.6) mortality. Men who did not participate in hobbies, club activities, or community groups had significantly higher multivariate RR for all-cause (RR = 1.5), all circulatory system disease (RR = 1.6) and non-cancer and non-circulatory system disease (RR = 2.3) mortality. Urban women who rarely or never met close relatives had significantly elevated risks of all-cause (RR = 2.4), all cancer (RR = 2.6), and non-cancer and non-circulatory system disease (RR = 2.7) mortality after adjustment for established risk factors.

Conclusions: This study provides evidence that social networks are an important predictor of mortality risk for middle-aged and elderly Japanese men and women. Lack of participation, for men, and being single and lack of meeting close relatives, for women, were independent risk factors for mortality.