S Tsugane1, S Sasaki, Y Tsubono. 1. Epidemiology and Biostatistics Division, National Cancer Center Research Institute East, Kashiwa, Japan. stsugane@east.ncc.go.jp
Abstract
OBJECTIVE: Although obesity is a major health problem in Western countries, its impact may differ in another culture. This paper examines the association between body mass index (BMI; kg/m2) and mortality in Japan, where the mortality profile and BMI distribution differ substantially from Western countries. DESIGN: The JPHC Study cohort I, a population-based prospective study in four public health center areas, started in 1990 and was followed-up to the end of 1999. SUBJECTS: A total of 19,500 men and 21,315 women aged 40-59 y who submitted their body weight and height and did not report any serious disease at baseline. MEASUREMENT: Risk of death by category of BMI. RESULTS: During 10 y of follow-up, 943 and 483 deaths were documented in the men and women, respectively. The association between BMI and all-cause mortality was U-shaped: compared with a category of 23.0-24.9, the statistically significant elevations in relative risk were observed in both under- and overweight categories (2.26 in 14.0-18.9, 1.57 in 19.0-20.9, 1.33 in 21.0-22.9 and 1.38 in 27.0-29.9, 1.97 in 30.0-39.9 in men, 1.94 in 14.0-18.9 and 1.91 in 30.0-39.9 in women) after adjustment for possible confounders and weight change. The U-shaped association did not change after excluding deaths occurring in the first 5 y of follow-up or even after excluding subjects who reported a weight loss of 5 kg or more since 20-y-old, or after stratifying subjects by smoking status. CONCLUSION: Both under- and overweight are important determinants of premature death among the Japanese population.
OBJECTIVE: Although obesity is a major health problem in Western countries, its impact may differ in another culture. This paper examines the association between body mass index (BMI; kg/m2) and mortality in Japan, where the mortality profile and BMI distribution differ substantially from Western countries. DESIGN: The JPHC Study cohort I, a population-based prospective study in four public health center areas, started in 1990 and was followed-up to the end of 1999. SUBJECTS: A total of 19,500 men and 21,315 women aged 40-59 y who submitted their body weight and height and did not report any serious disease at baseline. MEASUREMENT: Risk of death by category of BMI. RESULTS: During 10 y of follow-up, 943 and 483 deaths were documented in the men and women, respectively. The association between BMI and all-cause mortality was U-shaped: compared with a category of 23.0-24.9, the statistically significant elevations in relative risk were observed in both under- and overweight categories (2.26 in 14.0-18.9, 1.57 in 19.0-20.9, 1.33 in 21.0-22.9 and 1.38 in 27.0-29.9, 1.97 in 30.0-39.9 in men, 1.94 in 14.0-18.9 and 1.91 in 30.0-39.9 in women) after adjustment for possible confounders and weight change. The U-shaped association did not change after excluding deaths occurring in the first 5 y of follow-up or even after excluding subjects who reported a weight loss of 5 kg or more since 20-y-old, or after stratifying subjects by smoking status. CONCLUSION: Both under- and overweight are important determinants of premature death among the Japanese population.
Authors: Andrew O Odegaard; Mark A Pereira; Woon-Puay Koh; Myron D Gross; Sue Duval; Mimi C Yu; Jian-Min Yuan Journal: PLoS One Date: 2010-11-15 Impact factor: 3.240