Body weight history from young adulthood and subsequent blood pressure trajectories in a Japanese employment-based health-check cohort: a retrospective longitudinal study
Abstract
Whether long-term body weight history across adulthood, rather than body weight at a single time point, is associated with later blood pressure (BP) trajectories remains uncertain. We examined this association in a large longitudinal health checkup cohort. We analyzed repeated systolic and diastolic blood pressure measurements from 406,814 participants in a Japanese employment-based health-check cohort using sex-stratified linear mixed-effects models with age-by-group interactions and adjustment for sleep, exercise habits, physical activity, and walking speed. Participants were classified into three groups: history of underweight (BMI < 18.5 kg/m² at least once; n = 80,890), normal weight only (BMI 18.5 to < 25.0 kg/m² throughout the 20s; n = 228,343), and history of obesity (BMI ≥ 25.0 kg/m² at least once; n = 97,581). Sensitivity analyses among 402,651 participants additionally adjusted for eating speed, breakfast skipping, and late-evening meals. The underweight-history and obesity-history groups were further subdivided by the proportion of follow-up time spent underweight or obese (0–25%, 25–50%, 50–75%, and 75–100%). Participants with a history of obesity during their 20s showed higher BP trajectories than those with normal weight only, and these differences increased with age. In contrast, participants with a history of underweight showed lower BP trajectories than the normal-weight-only group. Age-by-group interactions were significant in women and men for both systolic and diastolic BP (all P < 0.001). Additional adjustment for eating behaviors had minimal influence on these associations. In the nine-group analysis, the groups with 50–75% and 75–100% of follow-up time spent obese showed higher BP trajectories than the normal-weight-only group, whereas the groups with 50–75% and 75–100% of follow-up time spent underweight showed lower BP trajectories. These findings suggest that shorter durations of obesity and underweight may be associated with BP trajectories closer to those of the normal-weight-only group. Overall, long-term body weight history may be more informative than body weight at a single time point when evaluating later BP burden.