Abstract
IMPORTANCE: Body weight fluctuation is associated with greater risks of adverse health outcomes. Whether intensive weight loss interventions affect the association of variability in adiposity measures with adverse health outcomes in individuals with type 2 diabetes has not been studied previously. OBJECTIVE: To evaluate the associations of long-term variability in adiposity indices with cardiovascular disease (CVD) outcomes and whether these associations are affected by an intensive lifestyle intervention among adults with type 2 diabetes. DESIGN, SETTING, AND PARTICIPANTS: This prospective cohort study included participants in the Action for Health in Diabetes (Look AHEAD) trial without CVD at baseline (August 2001 to April 2004). The Look AHEAD study included 16 centers in the United States. Data analysis was performed from December 2020 to June 2021. EXPOSURES: Variability of body mass index (BMI) and waist circumference (WC) across 4 annual visits, assessed using the coefficient of variation (CV), variability independent of the mean (VIM), and standard deviation (SD). MAIN OUTCOMES AND MEASURES: Main outcomes were (1) all-cause mortality, (2) cardiovascular deaths (deaths from myocardial infarction [MI] or stroke), and (3) CVD events (MI, stroke, and/or death from cardiovascular causes). RESULTS: Among 3604 study participants (mean [SD] age, 58.4 [6.6] years; 2240 [62.3%] women; 1364 [37.7%] Black participants; 2404 [66%] White participants), there were 216 CVD events, 33 CVD deaths, and 166 deaths over a median of 6.7 years. In the control group, the hazard ratios (HRs) for the highest quartile (quartile 4) compared with the lowest quartile (quartile 1) of CV of BMI were 4.06 (95% CI, 2.17-7.57), 15.28 (95% CI, 2.89-80.90), and 2.16 (95% CI, 1.21-3.87) for all-cause mortality, CVD mortality, and cardiovascular events, respectively. In the intervention group, the corresponding HRs were 0.99 (95% CI, 0.45-2.16), 1.14 (95% CI, 0.12-10.53), and 0.77 (95% CI, 0.40-1.49) for quartile 4 vs quartile 1. Regarding WC, in the control group, HRs for quartile 4 vs quartile 1 were 1.84 (95% CI, 1.01-3.35), 6.46 (95% CI, 1.16-36.01), and 1.28 (95% CI, 0.72-2.29). In the intervention group, HRs were 1.23 (95% CI, 0.61-2.46), 0.55 (95% CI, 0.15-2.11), and 0.70 (95% CI, 0.39-1.25) for quartile 4 vs quartile 1. CONCLUSIONS AND RELEVANCE: In this cohort study of individuals with type 2 diabetes, higher variability of adiposity indices was associated with significantly increased risk of CVD outcomes and death in the control group but not in the intensive lifestyle intervention group.