Association of Genome-Wide Genetic Risk for Obesity with the Quality, Quantity, and Timing of Workplace Food Purchases

Abstract

Abstract Objectives Understanding genetic predisposition to certain food choices could inform future tailored interventions to improve dietary intake and prevent cardiometabolic disease. BMI-associated genetic variants have been previously linked with self-reported eating behaviors. The aim of this study was to determine if BMI genetic risk was associated with objectively measured quality, quantity and timing of workplace food purchases. Methods Participants were employees of a large hospital in Boston, MA who enrolled in a randomized trial testing a workplace health promotion intervention. Cafeteria sales data, collected retrospectively for 3 months prior to the start of the intervention, were used to track the quality, quantity, and timing of purchases. A previously validated Healthy Purchasing Score was calculated using a traffic-light labeling system to estimate the healthfulness of the 3-month purchases (range 0–100%, higher = healthier). DNA was extracted from blood, and a BMI genome-wide polygenic score was generated by summing independent BMI-increasing risk alleles beta estimate from a genome-wide association study (Locke et al., 2015). Multivariable linear regression tested associations between BMI genetic score quartiles and workplace purchases, adjusted for age, sex, seasonality, and ancestry. Results In 397 participants, the BMI genetic score accounted for 15.0% of variance in BMI. The highest quartile of the genetic score (associated with greater BMI) was associated with a lower Healthy Purchasing Score compared to the lowest quartile (−5.8 percentage points, P = 0.004). The highest quartile of the BMI genetic score was also associated with purchasing a higher quantity of food items over 3-months (18.73 more items, P = 0.01) and with a later time of breakfast purchases (13.94 minutes later, P = 0.04) and lunch purchases (9.15 minutes later, P = 0.04) compared to the lowest quartile. Conclusions This study provides preliminary evidence for associations between genetic risk for obesity and the quality, quantity, and timing of objectively measured food purchased in a real-life setting. These findings suggest that BMI genetic scores could inform the development of personalized interventions to improve healthy eating and reduce obesity at work and other settings. Funding Sources NIH.