Longitudinal Trajectories of Body Roundness Index and Risk of Incident Hypertension in Non-Diabetic Elderly Adults: A Prospective Nationwide Cohort Study Involving Two Cohorts

Abstract

Background: Central adiposity is a key determinant of hypertension, yet most studies rely on single-time-point measures or include individuals with diabetes, potentially confounding associations. The relationship between long-term body roundness index (BRI) trajectories and incident hypertension in non-diabetic populations remains unclear. Methods: We conducted a longitudinal analysis of non-diabetic middle-aged and older adults from the China Health and Retirement Longitudinal Study (CHARLS), with external validation using the English Longitudinal Study of Ageing (ELSA). In CHARLS, BRI measurements from Waves 1–3 (2011–2016) were used to construct trajectories, and incident hypertension was assessed at Wave 4 (2018–2020). In ELSA, trajectories were constructed using BRI measurements from Waves 2, 4, and 6 (2004–2013), and incident hypertension was assessed at Wave 8 (2016–2017). Associations were examined using multivariable logistic regression, with subgroup and sensitivity analyses. Machine learning models were additionally developed for multivariable hypertension risk prediction. Results: Three BRI trajectories were identified in both cohorts: low-level inverted U-shaped, moderate-level inverted U-shaped, and high-level increasing trajectories. Higher BRI trajectories were associated with a graded increase in incident hypertension risk in CHARLS. Compared with the low-level inverted U-shaped trajectory, the adjusted odds ratios (95% confidence intervals) were 1.40 (1.13–1.73) for the moderate-level inverted U-shaped trajectory and 1.76 (1.21–2.56) for the high-level increasing trajectory. Directionally consistent findings were observed in ELSA, particularly for the high-level increasing trajectory. Findings remained robust in subgroup and sensitivity analyses. Machine learning analyses showed that BRI trajectory provided complementary predictive information when combined with clinical and biomarker variables. Conclusions: Longitudinal BRI patterns were associated with incident hypertension in non-diabetic adults. Repeated assessment of BRI may provide complementary information for identifying individuals at elevated hypertension risk, although its causal and clinical implications require further investigation.