Observational, MR, and longitudinal association: mediators and asthma
Both observational and MR studies supported the causal relationship between mediators and asthma (Figure 3). High FEV1/FVC and physical fitness reduced the risk of active asthma (OR, 0.98 and 0.93; 95% CI, 0.97–0.99 and 0.88–0.98, respectively), whereas SDB and early puberty increased the risk of active asthma (OR, 1.03 and 1.22; 95% CI, 1.01–1.05 and 1.05–1.42, respectively). The IVW results of the MR sensitivity analysis supported the findings of observational and MR analyses (Figure E3). However, the results of the causal effect of FeNO on asthma based on observational and MR analyses were in the opposite direction, indicating that FeNO is unlikely to mediate the association between obesity and asthma. The prospective cohort analyses also demonstrated that higher FEV1/FVC and physical fitness were associated with lower risk of active asthma (RR<1), whereas early puberty increased the risk of active asthma (RR>1) (Table 3).
The GRSs of the mediators served as an adequate indicator for the mediators, except for the GRS of FeNO. The F statistics for FEV1/FVC, FeNO, physical fitness, SDB, and early puberty were 16.25, 3.22, 10.27, 19.93, and 88.54, respectively (Figure 3). Furthermore, the GRSs of all the mediators, except for FeNO, were significantly associated with active asthma (Table E2). We did not observe strong associations of the individual SNPs of the mediators with active asthma (Tables E7–S11) and the confounders (Tables E12–E16).