Strengths and Limitations
Our cohort is the largest sample of
children with asthma with
gold-standard in-laboratory polysomnography in a prospective study
design. Objective characterization of sleep and breathing allowed for
accurate assessment of obstructive disordered breathing events leading
to precise classification of participants with OSA, thus strengthening
our findings. A few limitations should be considered. First, our sample
was 84% African American; therefore, our findings may not be
generalizable to all children with asthma. Nonetheless, it is worth
noting that the study by Ehsan et al2 also
demonstrated a high prevalence of obesity and OSA in an asthma cohort
that comprised 36% African American, therefore, the results described
in this study may be applicable to asthma populations of other
ethnicities. Second, our sample was not large enough to be split into
adequate training and validation sets, which would have allowed for
internal performance testing of our OSA screening model. Nevertheless,
given the simplicity of obtaining anthropometry and calculating BMI
z-scores, our findings may be easily replicated in other asthma cohorts,
which would enhance external validity and generalizability.