[Health practice and total mortality among middle-aged and elderly residents in Sukagawa, Japan].

PURPOSE: The present study aimed to clarify the relationship of total mortality with the HPI (health practices index: the additive index of five health practices) and lifestyle among elderly residents in a Japanese cohort. METHODS: A population-based prospective study was conducted in Sukagawa City of Fukushima Prefecture, Japan. Self-administered questionnaires were distributed and collected via mail. One-third of the city residents aged 40-69 years and over 70 years were randomly selected. Baseline surveys were conducted in February 2001 for the '40-69 cohort' and in July for the 'over 70 cohort'. Among 8746 subjects in the '40-69 cohort' and 2718 in the 'over 70 cohort', 5,657 (64.7%) and 2,019 (74.3%) responded, respectively. The questionnaires consisted of items on 'Alameda seven health practices' (weight, sleeping, smoking, drinking, breakfast, exercise, snacking), medical history, perceived health status, and home bound status of the 'over 70 aged cohort'. We followed the younger cohort for 3 years and 7 months and the elderly cohort for 3 years and 3 months until the end of October 2004, and checked their survival status using a municipal resident registry. Analysis of factors associated with total mortality one year after the baseline survey employed a simple chi2 test, Kaplan-Meier survival analysis, and the Cox proportional-hazards model to compute relative risks (RRs). The HPI was the additive index of five health practices: weight, sleeping, smoking, drinking, exercise. RESULTS: 1. No significant association was observed between lifestyle and total mortality in the '40-69 cohort'. However, there were significant links between total mortality in the 'over 70 cohort' and HPI and exercise. 2. The Kaplan-Meier curves for the 'over 70 cohort' showed higher survival for the group with higher HPI scores. Again, results of the Cox proportional-hazards model showed no factor significantly associated with the total mortality of the '40-69 cohort'. For the 'over 70 cohort', HPI, age, sex, subjective health and housebound status were associated. Medical history did not show any significant influence. For each of 'the Alameda 7 health practices', exercise was associated with all-cause mortality. CONCLUSION: 1. Maintaining good health practices improves life prognosis of the elderly. 2. The HPI is an useful indicator of life prognosis. 3. Promotion of exercise and prevention of becoming housebound are important for improving life prognosis.