Hailey R Banack1, Jay S Kaufman2. 1. McGill University, Department of Epidemiology, Biostatistics, and Occupational Health, Canada. Electronic address: hailey.banack@mail.mcgill.ca. 2. McGill University, Department of Epidemiology, Biostatistics, and Occupational Health, Canada.
Abstract
OBJECTIVE: To discuss possible explanations for the obesity paradox and explore whether the paradox can be attributed to a form of selection bias known as collider stratification bias. METHOD: The paper is divided into three parts. First, possible explanations for the obesity paradox are reviewed. Second, a simulated example is provided to describe collider stratification bias and how it could generate the obesity paradox. Finally, an example is provided using data from 17,636 participants in the US National and Nutrition Examination Survey (NHANES III). Generalized linear models were fit to assess the effect of obesity on mortality both in the general population and among individuals with diagnosed cardiovascular disease (CVD). Additionally, results from a bias analysis are presented. RESULTS: In the general population, the adjusted risk ratio relating obesity and all-cause mortality was 1.24 (95% CI 1.11, 1.39). Adjusted risk ratios comparing obese and non-obese among individuals with and without CVD were 0.79 (95% CI 0.68, 0.91) and 1.30 (95% CI=1.12, 1.50), indicating that obesity has a protective association among individuals with CVD. CONCLUSION: Results demonstrate that collider stratification bias is one plausible explanation for the obesity paradox. After conditioning on CVD status in the design or analysis, obesity can appear protective among individuals with CVD.
OBJECTIVE: To discuss possible explanations for the obesity paradox and explore whether the paradox can be attributed to a form of selection bias known as collider stratification bias. METHOD: The paper is divided into three parts. First, possible explanations for the obesity paradox are reviewed. Second, a simulated example is provided to describe collider stratification bias and how it could generate the obesity paradox. Finally, an example is provided using data from 17,636 participants in the US National and Nutrition Examination Survey (NHANES III). Generalized linear models were fit to assess the effect of obesity on mortality both in the general population and among individuals with diagnosed cardiovascular disease (CVD). Additionally, results from a bias analysis are presented. RESULTS: In the general population, the adjusted risk ratio relating obesity and all-cause mortality was 1.24 (95% CI 1.11, 1.39). Adjusted risk ratios comparing obese and non-obese among individuals with and without CVD were 0.79 (95% CI 0.68, 0.91) and 1.30 (95% CI=1.12, 1.50), indicating that obesity has a protective association among individuals with CVD. CONCLUSION: Results demonstrate that collider stratification bias is one plausible explanation for the obesity paradox. After conditioning on CVD status in the design or analysis, obesity can appear protective among individuals with CVD.
Authors: Alan C Kinlaw; Jessie P Buckley; Stephanie M Engel; Charles Poole; M Alan Brookhart; Alexander P Keil Journal: Epidemiology Date: 2017-05 Impact factor: 4.822
Authors: Mani Arsalan; Giovanni Filardo; Won-Keun Kim; John J Squiers; Benjamin Pollock; Christoph Liebetrau; Johannes Blumenstein; Jörg Kempfert; Arnaud Van Linden; Annika Arsalan-Werner; Christian Hamm; Michael J Mack; Helge Moellmann; Thomas Walther Journal: Clin Res Cardiol Date: 2016-08-17 Impact factor: 5.460
Authors: Jonathan M Kocarnik; Andrew T Chan; Martha L Slattery; John D Potter; Jeffrey Meyerhardt; Amanda Phipps; Hongmei Nan; Tabitha Harrison; Thomas E Rohan; Lihong Qi; Lifang Hou; Bette Caan; Candyce H Kroenke; Howard Strickler; Richard B Hayes; Robert E Schoen; Dawn Q Chong; Emily White; Sonja I Berndt; Ulrike Peters; Polly A Newcomb Journal: Int J Cancer Date: 2016-05-14 Impact factor: 7.396