BACKGROUND: It is unclear whether weight change since young adulthood affects the risk of mortality due to cardiovascular disease (CVD). The aim of this study was to investigate weight change since age 20 in relation to the risk of CVD mortality. METHODS AND RESULTS: A total of 41,364 eligible Japanese men and women aged 40-79 years participated in the Ohsaki Cohort Study baseline survey in 1994. Hazard ratios (HRs) and 95% confidence intervals (CIs) for CVD mortality were calculated according to weight change since age 20 (loss ≥ 10.0 kg; loss 5.0-9.9 kg; stable [± 4.9 kg]; gain 5.0-9.9 kg; gain ≥ 10.0 kg). During 13.3 years of follow-up, 1,756 participants died of CVD. The association between weight change and CVD mortality was L-shaped in men and U-shaped in women; the multivariate HR (95% CI) for men with weight loss ≥ 10.0 kg was 1.52 (1.25-1.85), and that for women with weight loss ≥ 10.0 kg and weight gain ≥ 10.0 kg was 1.62 (1.25-2.11) and 1.36 (1.09-1.69), respectively. Cross-classification analysis based on body mass index (BMI) at age 20 and weight change tended to be U-shaped, except for men whose BMI had been <25 kg/m(2) at age 20, in which case it was L-shaped. CONCLUSIONS: Weight loss since young adulthood is associated with excess risk of mortality due to CVD in men, while a U-shaped relationship was observed for women.
BACKGROUND: It is unclear whether weight change since young adulthood affects the risk of mortality due to cardiovascular disease (CVD). The aim of this study was to investigate weight change since age 20 in relation to the risk of CVD mortality. METHODS AND RESULTS: A total of 41,364 eligible Japanese men and women aged 40-79 years participated in the Ohsaki Cohort Study baseline survey in 1994. Hazard ratios (HRs) and 95% confidence intervals (CIs) for CVD mortality were calculated according to weight change since age 20 (loss ≥ 10.0 kg; loss 5.0-9.9 kg; stable [± 4.9 kg]; gain 5.0-9.9 kg; gain ≥ 10.0 kg). During 13.3 years of follow-up, 1,756 participants died of CVD. The association between weight change and CVD mortality was L-shaped in men and U-shaped in women; the multivariate HR (95% CI) for men with weight loss ≥ 10.0 kg was 1.52 (1.25-1.85), and that for women with weight loss ≥ 10.0 kg and weight gain ≥ 10.0 kg was 1.62 (1.25-2.11) and 1.36 (1.09-1.69), respectively. Cross-classification analysis based on body mass index (BMI) at age 20 and weight change tended to be U-shaped, except for men whose BMI had been <25 kg/m(2) at age 20, in which case it was L-shaped. CONCLUSIONS:Weight loss since young adulthood is associated with excess risk of mortality due to CVD in men, while a U-shaped relationship was observed for women.