Abstract
BACKGROUND: Many survivors of the Great East Japan Earthquake that occurred in 2011 were at risk of deteriorating health, especially elderly people living in disaster-stricken areas. The objectives of this prospective study were: a) to clarify the different lifestyle and psychosocial factors associated with frailty by sex among the non-disabled elderly survivors, and b) to describe the differences in characteristics stratified by the degree of disaster-related housing damage. METHODS: We followed 2261 Japanese survivors aged ≥65 years (45.3% male; mean age, 71.7 years) without disability or frailty who completed a self-administered questionnaire at baseline. All participants completed a baseline questionnaire in 2011 and at least one identical follow-up questionnaire between 2012 and 2015 regarding lifestyle (smoking status, alcohol intake, physical activity, sedentary lifestyle, and dietary intake) and psychosocial factors (self-rated health, standard of living, psychological distress, and social networks). Frailty was defined as a score of ≥5 on the Kihon Checklist, which is used by the Japanese government to certify the need for long-term care insurance. Adjusted odds ratios and 95% confidence intervals with frailty as the dichotomous dependent variable and health factors as the independent variables were calculated using a multilevel model for repeated measures by sex, followed by stratification analyses by the degree of housing damage. RESULTS: Over the 4-year study period, 510 participants (22.6%) developed frailty. In the post-disaster setting, many of the psychosocial factors remained more prevalent 4 years later among survivors with extensive housing damage. The presence of risk factors regarding the development of frailty differed by the degree of housing damage. Among men, psychological distress, in parallel with a poor social network, was related to frailty among only the participants with extensive housing damage and those living in temporary housing, whereas among women, worsening psychological distress was associated only with no damage and no displaced survivors. Among women with extensive damage and displacement, health outcomes such as overweight and diabetes and poor social networks were strongly related to frailty. CONCLUSIONS: Lifestyle and psychosocial factors associated with the risk of frailty differ by sex and the degree of housing damage.