高齢者総合機能評価は健診よりも健康寿命喪失を予測する:JAGESコホート研究

Daichi Okabe, Taishi Tsuji, Katsunori Kondo

In Japan, the Kihon checklist, which a useful part of the Comprehensive Geriatric Assessment (CGA), is performed using questionnaire. On the other hand, specific health checkup screens are available for some basic diseases, such as diabetes and dyslipidemia, which can cause stroke, the largest cause of long-term care need. However, to date, no report has compared CGA and health checkups in older people for their ability to predict healthy life expectancy; therefore, this study was undertaken to do that. Data from the Japan Gerontological Evaluation Study (JAGES) 2010, a self-administered mail survey of independent people aged 65 years or older, were used. A total of 9,756 participants in six cities responded to the questionnaire, underwent a health check, and were followed up for 3 years.Cox proportional hazards models were used to estimate hazard ratios (HR) for the eventual need for long-term care level 2 or more or death, adjusting for sex, age, drinking or smoking habits, educational years, and equivalent income. Explanatory variables in the Kihon checklist included seven risks, namely, general frailty, functional disability, malnutrition, oral dysfunction, seclusion, cognitive impairment, and depression, and in specific health checkups, 15 required items including metabolic syndrome. The incident rate of long-term care level 2 or more or death was 19.4/1,000 person-years. All risks in the Kihon checklist, excluding oral dysfunction, were significant (range of HRs: 1.44-3.63). Six items in the specific health checkups (urine protein, low BMI, AST, HDL, FPG, and HbA1c) were significant (range of HRs: 1.37-2.07). Metabolic syndrome was not significant (HR: 1.05). Therefore, CGA performed using questionnaire predicts healthy life expectancy better than a health checkup based on a blood test.