Observational and MR association
By using observational and MR 2SLS analysis, we calculated the causal associations of multiple obesity-related factors with asthma and pulmonary function (Table 2). The F statistics for obesity, physical activity, cardiopulmonary physical fitness, sleep-disordered breathing, and sleep quality were 80.60, 18.62, 11.68, 10.27, and 7.09, respectively. Three of the five obesity-related factors exhibited robust associations with asthma in both the observational and MR analysis. In order of decreasing effect sizes in MR analysis, these were sleep quality, SDB, and obesity, with the highest effect sizes per IQR increase, observed for sleep quality (OR = 1.42; 95% CI, 1.06 to 1.92) and the lowest for obesity (OR = 1.08; 95% CI, 1.00 to 1.16). Moreover, a one-IQR increase in z-BMI was also associated with a decrease in FEV1/FVC (β=−2.44, 95%CI, −4.24 to −0.65). Most of the MR sensitivity analysis results for the associations between obesity and respiratory outcomes were consistent with the main result (Table E2).
The GRSs of four tested factors (except for sleep quality) were robustly correlated with each factor and were associated with asthma (Table E3). Tables E4–8 reveal the associations of SNPs with the factors and asthma. The data listed in Tables E9–13 verify that most SNPs were independent of confounders.