The potentially modifiable burden of incident heart failure due to obesity: the atherosclerosis risk in communities study

Abstract

The authors estimated the generalized impact fraction (GIF) for heart failure (HF) related to obesity, representing the proportion of incident HF events that could be prevented from reductions in obesity and/or overweight. The Atherosclerosis Risk in Communities Study is a biracial population-based cohort study of persons aged 45-64 years from 4 US communities with a median 14 years of follow-up (1987-2003) for incident, hospitalized, or fatal HF. Body mass index (BMI; weight (kg)/height (m)(2)) was measured at baseline (1987-1989) and categorized as normal weight (BMI <25), overweight (BMI 25-29.9), or obese (BMI ≥30). After exclusion of prevalent HF, missing BMI, and poorly represented racial groups, the sample size was 14,642. The GIF and attributable fraction were calculated using a case-load weighted-sum method. A 95% distribution of the GIF was estimated from bootstrapped data sets. A 30% hypothetical reduction in obesity/overweight would potentially prevent 8.5% (95% simulation interval: 6.1, 10.7) of incident HF events. The attributable fraction, which assumes complete elimination of obesity/overweight, was 28% (95% simulation interval: 20, 36)-approximately 3 times larger than the most optimistic GIF calculated here. Investigators studying exposures that are unlikely to be eradicated given current prevention efforts, such as obesity, should consider estimating the GIF to avoid overestimates of population impact.

Figure 1.

The median generalized impact fraction and attributable fraction (with 2.5%–97.5% simulation intervals) derived from 10,000 bootstrap data sets using the case-load weighted-sum method, given 10 scenarios of reduced prevalence of obesity and overweight, Atherosclerosis Risk in Communities Study, 1987–2003. Scenarios for the reduced prevalence of overweight and obesity are as follows: 1) 15% reduction in obesity, shift obese to overweight; 2) 15% reduction in obesity, shift obese to normal weight; 3) 15% reduction in obesity and overweight, shift down 1 category; 4) 15% reduction in obesity and overweight, down to normal weight; 5) 30% reduction in obesity, shift obese to overweight; 6) 30% reduction in obesity, shift obese to normal weight; 7) 30% reduction in obesity and overweight, shift down 1 category; 8) 30% reduction in obesity and overweight, shift to normal weight; 9) attributable fraction, complete elimination of obesity; 10) attributable fraction, complete elimination of obesity and overweight. Bars, 95% simulation interval.