Predictors of Positive Airway Pressure Therapy Adherence Among U.S. Active Duty Military Personnel With Obstructive Sleep Apnea.

Introduction: Obstructive sleep apnea (OSA) is prevalent among U.S. military personnel, but adherence to positive airway pressure (PAP) treatment is suboptimal. This study sought to identify factors that predict the adherence to PAP therapy of active duty military patients newly diagnosed with OSA.

Materials And Methods: The study was a retrospective cross-sectional analysis of 239 active duty military patients (94% men, mean age 40.0 ± 7.8 years), who were admitted into PAP therapy and participated in a single 90-minute behavioral intervention session. The session included a self-assessment of sleep trouble, sleep distress, impact of sleep on functioning, emotional difficulty from sleep, stress appraisal, and beliefs about OSA and PAP. PAP usage data of the patients were analyzed at 30-, 60-, and 180 days post-intervention using descriptive statistics and multivariable linear regression analysis.

Results: Positive airway pressure adherence of at least 4 hours nightly usage decreased slightly over the course of the study; mean number of days used over a 30-day timeframe at the 30-, 60-, and 180-day mark were 15.6, 14.9, and 14.1 days, respectively. On nights used, PAP usage remained consistent at 4.8 to 5.0 hours over the 30-, 60-, and 180-day mark. Age, apnea-hypopnea index, readiness, self-efficacy or confidence in treatment implementation, and worries about sleep were statistically significant predictors of adherence. Furthermore, worries about sleep negatively affected PAP usage hours on nights used, regardless of meeting adherence criterion, in the early phase of treatment and in the long run at the 180-day mark.

Conclusions: Readiness for therapy, self-efficacy, confidence in putting into action the treatment plan, and worries about sleep are modifiable variables that may be targeted in programs to boost PAP adherence and usage among the military population. Future studies should explore the predictive aspects of each of these variables and identify interventions to improve them.