AI Can Predict Who Will Stop Using Their CPAP
Oxford Academic — Sleep Journal (April 6, 2026)
Artificial intelligence can now analyze CPAP usage patterns to predict non-adherence before it becomes chronic, giving clinicians a window to intervene early.
Published in Sleep (Oxford Academic) on April 6, 2026, this study is one of the first comprehensive applications of machine learning to CPAP therapy adherence. The AI model learns from real-world usage data — night-by-night mask-on time, pressure changes, leak events — to identify patterns that historically precede a patient giving up.
The core finding: adherence problems don't appear suddenly. They follow predictable trajectories that standard clinic follow-ups miss. By the time a patient reports struggling at their three-month appointment, the data had been signaling trouble for weeks.
With AI monitoring, a sleep coach or respiratory therapist could receive an alert and reach out proactively — before the patient has already decided CPAP isn't for them.
CPAP remains the gold-standard treatment for obstructive sleep apnea, but adherence rates hover around 50% — meaning half of patients prescribed CPAP don't benefit from it. If AI can close even part of that gap, the downstream health benefits are enormous: lower cardiovascular risk, better blood pressure control, fewer accidents, and better daytime function.
