ISSN: 2167-0277
Thibaut Gentina, Elodie Gentina, Catherine Lamblin, François Jouniaux, François Codron and Aiman Tulaimat
Objective: Early prediction of which patients with obstructive sleep apnea will not adhere with CPAP can trigger physicians to correct possible causes or offer alternative therapies. This study aims to determine the ability of early CPAP use to predict long-term adherence with CPAP independent from its definition and to test previously described predictors of adherence.
Method: This is a prospective, observational cohort study that was performed in nine sleep disorder centers in France. Patients were newly diagnosed with obstructive sleep apnea and a CPAP was prescribed for the first time. We collected socioeconomic, medical, sleep, and psychological variables to be associated with adherence with CPAP. The patients were evaluated 15 days after CPAP set up and monthly for 4 months until they were adherent. We assessed potential predictors of CPAP adherence at 4 months with emphasis on CPAP use at 15 days. CPAP adherence was defined as an average daily use of at least 3, 4, 5, and 6 hours.
Results: We enrolled 420 patients. Early CPAP use and CPAP use at 4 months were similar. At 4 months, 94% of patients used a CPAP ≥3 hours and 49% of patients used it ≥6 hours. The area under the ROC curve for early CPAP use predicting adherence at 4 months was ≥0.85 for all adherence definitions. The optimal threshold for early CPAP use to predict adherence increased from 3.2 to 6.4 hours as the definition of CPAP adherence increased from 3 to 6 hours. CPAP use at 4 months was higher in older patients.
Conclusions: Early CPAP use was the single best predictor of CPAP adherence and was independent of how adherence was defined. Physicians must assess CPAP use no later than 2 weeks to address its causes or prescribe an alternative therapy.