Association between types of social isolation and all-cause mortality among the survivors in the disaster-affected area of the great East Japan earthquake: The RIAS study.
Key claim: Social isolation from friends, but not family members, is significantly associated with an increased risk of all-cause mortality.
Abstract
OBJECTIVE: We aimed to determine whether different types of social isolation after a disaster are associated with all-cause mortality, which currently remains unknown, using data from a prospective cohort study of survivors in an area affected by the Great East Japan Earthquake. METHODS: In total, 9311 participants (3600 men and 5711 women) were followed up from October 1, 2011, to the occurrence of censoring events: death, loss to follow-up, or the end of the follow-up period (December 31, 2017). Social isolation types were assessed using the Lubben Social Network Scale-6 (LSNS-6) Family Subscale and LSNS-6 Friend Subscale (using 5/6 as a cutoff point). Using Cox proportional hazards models, the hazard ratios of all-cause mortality according to the social isolation type (no isolation as the reference) were estimated. RESULTS: Social isolation from friends, but not from family members, was significantly associated with an increased risk of all-cause mortality (hazard ratio, 1.28; 95% confidence interval, 1.10,1.63). CONCLUSIONS: Social isolation from friends contributes to an increased risk of all-cause mortality. More focused preventive healthcare and medical support for premature death after a disaster and the preservation of social networks are required for disaster survivors who are socially isolated from friends.
Why this matters
[human-observational] This prospective disaster-survivor cohort distinguishes friend isolation from family isolation in relation to all-cause mortality, rather than treating social connection as a single exposure or relying on short-term stress markers. [human-observational] The absence of a statistically significant family-isolation association does not establish that family ties are irrelevant or that the two associations differ statistically; residual confounding, illness-related social withdrawal, and the disaster-specific setting limit causal interpretation and generalization.