Hailey R Banack1, Jay S Kaufman2. 1. Department of Epidemiology, Biostatistics, and Occupational Health, McGill University, Montreal, QC, Canada. Electronic address: hailey.banack@mail.mcgill.ca. 2. Department of Epidemiology, Biostatistics, and Occupational Health, McGill University, Montreal, QC, Canada.
Abstract
OBJECTIVES: The objectives of this article are to demonstrate that the obesity paradox may be explained by collider stratification bias and to estimate the biasing effects of unmeasured common causes of cardiovascular disease (CVD) and mortality on the observed obesity-mortality relationship. METHODS: We use directed acyclic graphs, regression modeling, and sensitivity analyses to explore whether the observed protective effect of obesity among individuals with CVD can be plausibly attributed to selection bias. Data from the third National Health and Examination Survey was used for the analyses. RESULTS: The adjusted total effect of obesity on mortality was a risk difference (RD) of 0.03 (95% confidence interval [CI]: 0.02, 0.05). However, the controlled direct effect of obesity on mortality among individuals without CVD was RD = 0.03 (95% CI: 0.01, 0.05) and RD = -0.12 (95% CI: -0.20, -0.04) among individuals with CVD. The adjusted total effect estimate demonstrates an increased number of deaths among obese individuals relative to nonobese counterparts, whereas the controlled direct effect shows a paradoxical decrease in morality among obese individuals with CVD. CONCLUSIONS: Sensitivity analysis demonstrates unmeasured confounding of the mediator-outcome relationship provides a sufficient explanation for the observed protective effect of obesity on mortality among individuals with CVD.
OBJECTIVES: The objectives of this article are to demonstrate that the obesity paradox may be explained by collider stratification bias and to estimate the biasing effects of unmeasured common causes of cardiovascular disease (CVD) and mortality on the observed obesity-mortality relationship. METHODS: We use directed acyclic graphs, regression modeling, and sensitivity analyses to explore whether the observed protective effect of obesity among individuals with CVD can be plausibly attributed to selection bias. Data from the third National Health and Examination Survey was used for the analyses. RESULTS: The adjusted total effect of obesity on mortality was a risk difference (RD) of 0.03 (95% confidence interval [CI]: 0.02, 0.05). However, the controlled direct effect of obesity on mortality among individuals without CVD was RD = 0.03 (95% CI: 0.01, 0.05) and RD = -0.12 (95% CI: -0.20, -0.04) among individuals with CVD. The adjusted total effect estimate demonstrates an increased number of deaths among obese individuals relative to nonobese counterparts, whereas the controlled direct effect shows a paradoxical decrease in morality among obese individuals with CVD. CONCLUSIONS: Sensitivity analysis demonstrates unmeasured confounding of the mediator-outcome relationship provides a sufficient explanation for the observed protective effect of obesity on mortality among individuals with CVD.
Authors: Elizabeth Rose Mayeda; Hailey R Banack; Kirsten Bibbins-Domingo; Adina Zeki Al Hazzouri; Jessica R Marden; Rachel A Whitmer; M Maria Glymour Journal: Epidemiology Date: 2018-07 Impact factor: 4.822
Authors: Bryant R England; Joshua F Baker; Harlan Sayles; Kaleb Michaud; Liron Caplan; Lisa A Davis; Grant W Cannon; Brian C Sauer; E Blair Solow; Andreas M Reimold; Gail S Kerr; Ted R Mikuls Journal: Arthritis Care Res (Hoboken) Date: 2017-12-06 Impact factor: 4.794
Authors: Bette J Caan; Jeffrey A Meyerhardt; Candyce H Kroenke; Stacey Alexeeff; Jingjie Xiao; Erin Weltzien; Elizabeth Cespedes Feliciano; Adrienne L Castillo; Charles P Quesenberry; Marilyn L Kwan; Carla M Prado Journal: Cancer Epidemiol Biomarkers Prev Date: 2017-05-15 Impact factor: 4.254
Authors: Chesney D Castleberry; John L Jefferies; Ling Shi; James D Wilkinson; Jeffrey A Towbin; Ryan W Harrison; Joseph W Rossano; Elfriede Pahl; Teresa M Lee; Linda J Addonizio; Melanie D Everitt; Justin Godown; Joseph Mahgerefteh; Paolo Rusconi; Charles E Canter; Steven D Colan; Paul F Kantor; Hiedy Razoky; Steven E Lipshultz; Tracie L Miller Journal: JACC Heart Fail Date: 2018-02-07 Impact factor: 12.035