BACKGROUND Recent studies have found that long-term changes in weight during adulthood are associated with a high risk of mortality. The objective of this study was to characterize body mass index (BMI) trajectories during adulthood and to examine the association between BMI trajectories and risk of death in the Japanese population. METHODS The data were extracted from Japan Public Health Center-based Prospective Study-a population-based prospective cohort study in Japan with participants aged 40-69 years followed over 20 years. The participants were categorized into multiple BMI trajectory groups using the latent class growth model. The Cox proportional-hazards model was conducted using all-cause mortality and cause-specific mortality as outcomes and the identified BMI trajectory groups as a predictor. In total, 65 520 participants were included in the analysis. RESULTS Six BMI trajectory groups were identified: underweight stable (Group 1), low-to-high normal (Group 2), high-to-low normal (Group 3), normal to overweight (Group 4), overweight to normal (Group 5) and normal to obese (Group 6). Our Cox models showed a higher hazard (risk) of all-cause mortality among participants in the BMI-declining groups [Group 3, adjusted hazard ratio (aHR): 1.10, 95% CI: 1.05-1.16; Group 5, aHR: 1.16, 95% CI: 1.08-1.26], underweight stable group (Group 1, aHR: 1.27, 95% CI: 1.21-1.33) and normal to obese group (Group 6, aHR: 1.22, 95% CI: 1.13-1.33) than Group 2 (low-to-high normal BMI trajectory). CONCLUSIONS Stable underweight and weight loss were associated with a high risk of mortality, both of which were uniquely observed in a Japanese population.

中文翻译：

---

使用基于人群的前瞻性队列研究来了解日本的体重指数轨迹和死亡风险：基于日本公共卫生中心的前瞻性研究。

背景技术最近的研究发现，成年期间体重的长期变化与高死亡风险相关。本研究的目的是描绘成年期体重指数 (BMI) 轨迹的特征，并研究 BMI 轨迹与日本人群死亡风险之间的关联。方法 数据摘自日本公共卫生中心的前瞻性研究，这是一项在日本进行的基于人群的前瞻性队列研究，参与者年龄为 40-69 岁，随访时间超过 20 年。使用潜在类别增长模型将参与者分为多个 BMI 轨迹组。Cox 比例风险模型是使用全因死亡率和特定原因死亡率作为结果，并使用已确定的 BMI 轨迹组作为预测因子来进行的。分析总共包括 65 520 名参与者。结果 确定了六个 BMI 轨迹组：体重不足稳定（第 1 组）、从低到高正常（第 2 组）、从高到低正常（第 3 组）、正常到超重（第 4 组）、超重到正常（第 3 组） 5) 和正常至肥胖（第 6 组）。我们的 Cox 模型显示，BMI 下降组的参与者全因死亡的风险较高[第 3 组，调整后风险比 (aHR)：1.10，95% CI：1.05-1.16；第 5 组，aHR：1.16，95% CI：1.08-1.26]、体重不足稳定组（第 1 组，aHR：1.27，95% CI：1.21-1.33）和正常至肥胖组（第 6 组，aHR：1.22，95%） CI：1.13-1.33）高于第 2 组（从低到高的正常 BMI 轨迹）。结论 稳定的体重不足和体重减轻与高死亡风险相关，这两种情况在日本人群中都是独一无二的。