Keywords: Inverse propensity score, full matching, causal mediation, asthma, obesity, national longitudinal survey
This study examines whether asthma predicts obesity, physical activity, and self-rated health among middle-aged US adults. We used inverse-propensity score weighting (IPW) in a subsample from the National Longitudinal Survey of Youth 1979 (N=5077), and confirmed the results with full matching (FM) (N=4949). Subjects were 40 year olds at baseline and 50 at the follow up. Both methods balanced asthma versus non-asthma on 7 covariates. IPW method preserved full sample, but FM discarded 128 subjects. In IPW-weighted binary logistic model, adults with asthma had 16% greater odds of developing obesity than adults without asthma, OR (95% CI) =1.16 (1.04-1.29). In the IPW-weighted cumulative logistic model, those with asthma had 59% greater odds of reporting that their health limited physical activity compared to those without asthma; OR=1.59 (1.18-2.14). FM method confirmed these findings [OR=1.41(1.04-1.93) for obesity and OR=1.70(1.28-2.27) for physical activity. Our study establishes that baseline asthma is predictive of subsequent obesity and limited physical activity among middle-aged us adults. However, physical activity does not mediate asthma and obesity relationship.