Statistical Analyses
Logistic regression was used to examine the association of the SRBD
score as well as sub-questions with the presence of OSA. Screening
accuracies of the SRBD score and sub-questions associated with OSA were
examined by calculating the
positive predictive values (PPV),
negative predictive values (NPV), sensitivities, specificities, and the
area under the curve (AUC) for the receiver‐operating characteristic
(ROC). AUCs were compared using the Delong’s test for two ROC curves.
Since asthma may increase the susceptibility for upper airway
obstruction4,5, in post-hoc analyses, we examined
whether the odds of having OSA increased with asthma severity using
logistic regression in an unadjusted model and a model adjusted for age,
sex and BMI z-score.
All analyses were performed using R (www.r-project.org, with the
“stats”, “remotes”, “ggplot2”, and “lme4”
packages)18 and a two-sided P < 0.05 was
considered statistically significant. All values were reported as
means ± SD interquartile ranges and N (%) as appropriate.