A mortality comparison of participants and non-participants in a comprehensive health examination among elderly people living in an urban Japanese community

Background and aims: Recent studies have revealed that there are critical differences between participants and non-participants in health examinations. The aim of this study was to examine mortality differences between participants and non-participants in a comprehensive health examination for prevention of geriatric syndromes among community-dwelling elderly people, using a three-year prospective cohort study. Methods: The study population included 854 adults aged 70 to 84 at baseline. The following items were all studied: the status of participation in the comprehensive health examination as an independent variable, age, gender, number of years of education, living alone, presence of chronic diseases, experience of falls over one year, history of hospitalization over one year, self-rated health, body mass index, instrumental activities of daily living, and subjective well-being as covariates; and all-cause mortality during a three-year follow-up as a dependent variable. Results: In an adjusted Cox's proportional hazard regression model, the mortality risk for participants in the comprehensive health examination was significantly lower than that of non-participants (Risk Ratio (for participants)=0.44, 95% confidence interval=0.24 to 0.78). Conclusions: The present study shows that there is a large mortality difference between participants and non-participants. Our findings suggest two possible interpretations: 1) There is a bias due to self-selection for participation in the trial, which was not eliminated by adjustment for the covariates in the statistical model; 2) There is an intervention effect associated with participation in the comprehensive health examination which reduces the mortality risk.