Wearable Sleep Data Predict Adherence to Pulmonary Rehabilitation

Chronic obstructive pulmonary disease (COPD) is a long-term lung disorder that makes breathing difficult and often disturbs sleep, reducing energy for daily activities. Limited engagement in pulmonary rehabilitation remains a major barrier to delivering effective remote care for these patients. A new study shows that sleep data from wearable devices may help predict which patients will stay engaged in home-based rehabilitation. The approach aims to support earlier, targeted interventions that keep patients active in therapy.

Mayo Clinic researchers in the Kern Center for the Science of Health Care Delivery (Rochester, MN, USA) evaluated whether objective sleep measures could forecast participation in remote pulmonary rehabilitation. The team captured one week of baseline sleep data from a wrist activity monitor and generated a Composite Sleep Health Score before program start. They combined that score with traditional clinical indicators using machine learning to estimate how consistently patients would participate in a 12‑week home rehabilitation program.

The method translated everyday sleep quality into a practical engagement signal that clinicians could use to tailor support. Investigators framed the work as a proof‑of‑concept and focused on home-based care to reflect real-world conditions. The findings were published in Mayo Clinic Proceedings: Digital Health.

At the end of 12 weeks, analysis showed that adding the Composite Sleep Health Score improved prediction of patient engagement over the study period. This information can help clinicians personalize rehabilitation plans and identify patients who may need additional assistance to complete remote therapy. The results may also guide the design of future remote‑care programs that integrate objective behavioral data with clinical assessments and patient-reported information.

Researchers noted that additional investigation is needed to validate and refine the predictive model in broader patient populations before wider clinical application.

“As a scientist and engineer, I wanted to explore how wearable data could improve the drop-out rates of remote pulmonary rehabilitation programs. By better understanding a patient's day-to-day life, we can make more personalized and potentially more effective care plan recommendations,” said Stephanie Zawada, Ph.D., M.S., a Mayo Clinic research associate and first author of the study.

Related Links
Kern Center for the Science of Health Care Delivery