Abdominal obesity and myocardial infarction risk-we demonstrate the anthropometric and mathematical reasons that justify the association bias of the waist-to-hip ratio

Abstract

Background: The waist-to-hip ratio (WHR) is widely used to evaluate the association of abdominal obesity with myocardial infarction (MI). Objective: Our aim was to determine whether WHR-associated risk provides a bias. Methods: A case-control study in 252 men. Stratification was used as an approach for removing bias effects. We created a baseline covariate (WHR0.95-0.99) from a new matched sample in the stratum between 0.95 and 0.99. This stratum coincides with the overlap area of the distribution, where all subjects have a similar propensity score. We considered other covariate (WHRS), conditioned on WHR<1 and waist circumference (WC) being assigned a spurious risk. We hypothesized that subtracting hip circumference from WC (WHD) can be essential to observe the confounding effect provided by WHR. Results: BMI: AUC: 0.694, 95% CI (0.628-0.760); OR: 3.8. WC: AUC: 0.743, 95% CI (0.681-0.805); OR: 5.7. WHR: AUC: 0.798, 95% CI (0.740-0.855); OR: 8.6. Waist-height ratio (WHtR): AUC: 0.782, 95% CI (0.724-0.840); OR: 8.5. WHD: AUC: 0.204, 95% CI (0.146-0.261); OR: 0.36. Prevalence in cases: WHR≥0.95 (84.1% vs. 38%; OR: 8.6); WHR<1 (36.3% vs. 85.7%; OR: 2.3); WHR≥1 (63.4% vs. 14.2%; OR: 4.4); WC≥94.4 (71.4% vs. 30.1%; OR: 5.7); WHD≥2.2 (27.7% vs. 75.3%; OR: 7.9); WHRs (50% vs. 25%; OR: 2). Conclusions: WHR provides an association bias in MI cases. This can be extrapolated to other study populations. The bias is explained by a mathematical misconception where the protective effect of HC is overestimated concerning WC and height. The risk associated with WHR as higher than that associated with WC and WHtR entails anthropometric inconsistency and bias, to the extent of becoming epidemiologically false.