Estimating Effect of Obesity on Stroke Using G-Estimation: The ARIC study

Obesity (Silver Spring). 2019 Feb;27(2):304-308. doi: 10.1002/oby.22365.

Abstract

Objective: This study quantified the obesity-stroke relationship by appropriately adjusting for time-varying confounders using G-estimation.

Methods: A total of 13,975 participants in the Atherosclerosis Risk in Communities (ARIC) study were included. General obesity (GOB) was defined as BMI ≥ 30 kg/m2 ; abdominal obesity (AOB) was defined as waist circumference ≥ 102 cm in men and ≥ 88 cm in women and waist to hip ratio ≥ 0.9 in men and ≥ 0.85 in women. The effects of obesity on stroke were estimated using G-estimation and compared with accelerated failure time models using three modeling strategies.

Results: The first accelerated failure time model adjusted for baseline covariates excluding metabolic mediators of obesity showed increased risk of stroke for all measures of obesity. Further adjustment for hypertension, diabetes mellitus, and lipid profiles resulted in decreasing hazard ratios (HRs) with intervals that included the null value for all measures of obesity. G-estimated HRs were 1.60 (95% CI: 1.08-2.40), 1.43 (95% CI: 1.14-1.99), and 1.99 (95% CI: 1.50-2.91) for GOB and AOB based on waist circumference and waist to hip ratio.

Conclusions: Both GOB and AOB affected the risk of stroke. The magnitude of the estimates was larger when modeled by G-estimation than when using standard models, suggesting that bias from mishandling of time-varying confounding was toward the null.

MeSH terms

  • Female
  • Humans
  • Male
  • Middle Aged
  • Obesity / complications*
  • Stroke / etiology*
  • Stroke / pathology