Obstructive sleep apnea and mental health disorders in the pediatric population: a retrospective population-based cohort study

Rationale — Information is limited about the association between obstructive sleep apnea (OSA) and mental disorders in children.

Objectives — In children, (1) to evaluate the association between OSA and new mental healthcare encounters; (2) to compare mental healthcare encounters two years post- to pre-OSA treatment initiation.

Methods — We conducted a retrospective longitudinal cohort study using Ontario health administrative data (Canada). Children (0-18 years) who underwent diagnostic polysomnography (PSG) 2009-2016 and met criteria for definition of moderate-severe OSA (PSG-OSA) were propensity score weighted by baseline characteristics and compared to children who underwent a PSG in the same period but did not meet the OSA definition (PSG-No-OSA). Children were followed until March 2021. Weighted cause-specific Cox Proportional Hazards and Modified Poisson regression models were used to compare time from PSG to first mental healthcare encounter and frequency of new mental healthcare encounters per person time, respectively. Among those who underwent adenotonsillectomy (AT) or were prescribed and claimed positive airway pressure therapy (PAP), we used age-adjusted conditional logistic regression models to compare two years post- to pre-treatment odds of mental healthcare encounters.

Results — Of 32,791 children analyzed, 7,724 (23.6%) children met criteria for moderate-severe OSA. In PSG-OSA group, 7,080 (91.7%) were treated (AT or PAP). Compared to PSG-No-OSA, the PSG-OSA group had a shorter time from PSG to first mental healthcare encounter (HR: 1.08; 95%CI: 1.05-1.12), but less frequent mental healthcare encounters in follow-up (RR: 0.92; 95% CI: 0.87-0.97). OSA treatment (AT or PAP) was associated with lower odds of mental healthcare encounters two years post-treatment initiation compared to two years prior (OR: 0.69; 95% CI: 0.65-0.74).

Conclusion — In this large population-based study of children who underwent PSG for sleep disorder assessment, OSA diagnosis/treatment was associated with an improvement in some mental health indicators, such as fewer new mental healthcare encounters compared to no OSA, and lower odds of mental healthcare encounters compared to pre-OSA treatment.