Methods We examined cross-sectional and longitudinal relationships between self-reported CAM use and health outcomes in urban adolescents with asthma. Participants (Time 1: N=151; Time 2: N=132) completed questionnaires regarding the use of 10 CAM modalities following two clinic visits one year apart as part of a larger study. CAM use was dichotomized (high/low) due to its nonnormal distribution. T-tests examined between-group differences in outcomes at both time points. Multivariable regression analyses using backwards elimination examined relationships between CAM use at Time 1 and health outcomes at Time 1 and Time 2, when controlling for key covariates and, in longitudinal analyses, Time 1 functioning.
