Body Mass Index, Walking Pace, and Risk of Coronary Heart Disease: a Mendelian Randomization and Mediation Analysis

Tao Zhou,Dianjianyi Sun,Xiang Li,Mengyu Fan,Yoriko Heianza,JoAnn Manson,Lu Qi
DOI: https://doi.org/10.1161/circ.139.suppl_1.P324
IF: 37.8
2019-01-01
Circulation
Abstract:Introduction: Being obese has been related to changes in physical fitness, such as pace-speed declination, which was recently associated with increased coronary heart disease (CHD) risk. The causal link between obesity, walking pace declination, and CHD remains to be established. Hypothesis: High BMI is causally related to slower walking pace, and the relationship between high BMI and CHD that is mediated by altered walking pace. Methods: Data form UK biobank including 408,836 white British participants (enrolled from 2006 to 2010) was used in this study. The usual walking pace was self-reported as slow, steady/average, and brisk, among 352,663 participants with 3,634 CHD cases (enrolled from 2006 to 2010 and followed up to 2016). The causal relation between BMI and walking pace was assessed by Mendelian randomization analysis using 97 BMI related single-nucleotide polymorphisms. A further mediation analysis was performed to assess the association between BMI, walking pace, and incident CHD. Results and Conclusions: In Mendelian randomization analysis performed among 408,836 individuals in the UK Biobank (baseline mean [SD] age, 56.8 [8.0] years; 54.1% (221,196 of 408,836) women), we found a significant causal relationship between genetically instrumented higher BMI and slower walking pace (odds ratio: 1.58, 95% CI: 1.45-1.71, per 1 SD increase in BMI), independent of age, sex, socioeconomic status, alcohol intake, and smoking. In mediation analysis, we found that slow and average walking pace was estimated to mediate 29% and 19% of the association between BMI and incident CHD, respectively, compared with brisk walking pace (p<.001). In conclusion, our results indicate that higher BMI may be causally related to a slower walking pace, which mediates a significant proportion of the association between high BMI and CHD risk.
What problem does this paper attempt to address?