A new telemonitoring feature for detection of long-term CPAP adherence

Rationale. Suboptimal adherence limits the efficacy of Continuous Positive Airway Pressure (CPAP) in Obstructive Sleep Apnea (OSA). Objective. To determine whether the Monthly Adherence Standard Deviation (MASD), that quantifies the variability in CPAP use during the baseline month (January 2021, at least the fourth month of treatment) provides predictive information about adherence, 6 and 12 months after baseline that would not be captured by the Monthly Adherence Mean (MAM) value alone. Methods. This retrospective analysis includes CPAP telemonitoring data from a population of 1612 patients. The overall population was randomly assigned to a construction (80 %) and test cohort (20 %) for internal validation. A threshold on baseline MASD was defined using a Receiver Operating Characteristic (ROC) curve. Results. A MASD threshold of 1.76 h was identified. Based on this threshold and the standard 4 h/day criterion applied to the MAM, patients were classified into four groups: high MAM/low MASD, high MAM/high MASD, low MAM/low MASD, and low MAM/high MASD. Significant differences were observed among the groups 6 and 12 months after baseline data. Six months after baseline, average MAM for each patient group in the test population were 6.84 ± 1.58, 5.66 ± 1.97, 1.27 ± 2.09, and 3.04 ± 1.90 h/day, respectively (p < 0.001); percentages of adherent patients were 91.4 %, 69.9 %, 6.25 %, and 13.9 % (p < 0.001). Similar patterns were found 12 months after baseline. Conclusions. MASD in CPAP adherence can distinguish between patients with different adherence behaviors 6 and 12 months after, capturing patterns not evident from MAM alone.