DOI: 10.1161/jaha.126.049396 ISSN: 2047-9980

Obesity Polygenic Risk and Healthy Lifestyle Interactions on Weight Trajectories in Women and Men

Aishi Ghosh, Meghan Lau, Anika Fernandes, Brian E. Cade, Elizabeth W. Karlson, Susan Redline, Richa Saxena, Tianyi Huang, Hassan S. Dashti

Background

Genetics and environmental factors contribute to obesity risk, but the extent to which healthy behaviors can offset genetic susceptibility remains unclear. We examined the interaction between obesity polygenic risk and a composite healthy lifestyle score on body mass index (BMI) trajectories in women and men.

Methods

We analyzed 13 780 women from the Nurses' Health Study and 8242 men from the Health Professionals Follow‐Up Study, all of European ancestry and free of major chronic disease at baseline. The lifestyle score comprised American Heart Association Essential 8 components (nonsmoking, physical activity, healthy eating, adequate sleep) plus moderate alcohol intake, modeled as a time‐varying variable. A genome‐wide polygenic score for BMI was derived from genome‐wide association study. Adjusted linear mixed‐effects models estimated associations and interactions on biennial BMI measures over up to 26 years.

Results

Each SD increase in the polygenic score was associated with 1.80 kg/m 2 (95% CI, 1.72–1.87) and 1.12 kg/m 2 (95% CI, 1.06–1.19) higher BMI in women and men, respectively. Significant interactions between the polygenic score and healthy lifestyle score (both P <0.05) showed a dose–response attenuation of the genetic effects with healthier lifestyles. Comparing the healthiest with the least healthy lifestyle groups, genetic effects on BMI were 35% lower in women and 28% lower in men. In sensitivity analyses, higher diet quality and physical activity consistently attenuated genetic associations in both cohorts, whereas current smoking showed similar effects in women only.

Conclusions

Adherence to a healthier lifestyle attenuated the association between obesity polygenic risk and BMI in a dose–response manner.

More from our Archive