Can Passive Cough Monitoring Predict COPD Exacerbations?

Area of Science:

• Respiratory Medicine
• Digital Health
• Biomedical Engineering

Background:

• Chronic obstructive pulmonary disease (COPD) exacerbations significantly impact patient health and healthcare costs.
• Current methods for detecting exacerbations can be subjective and lack continuous monitoring.
• Objective, unobtrusive monitoring systems are needed to improve early detection of COPD exacerbations.

Purpose of the Study:

• To validate an alert mechanism for detecting COPD exacerbations using a stationary, unobtrusive nighttime cough monitor.
• To assess the feasibility and effectiveness of a personalized cough classifier for identifying deteriorating trends in COPD patients.
• To compare alerts generated by the cough monitoring system with patient-reported exacerbation onsets.

Main Methods:

• A prospective, double-blind, longitudinal study involving 40 COPD patients monitored for 12-24 weeks.
• Utilized a stationary, unobtrusive nighttime monitor to continuously record cough frequency.
• Employed an automated system with a personalized cough classifier to identify deteriorating cough trends and generate alerts.

Main Results:

• The study included 38 patients with 41 exacerbations over 3981 monitored days.
• The cough-based alert system flagged 59% of exacerbations, with alerts preceding onset by 4 or more days in some cases.
• An estimated 30% of exacerbations were not associated with increased coughing, indicating limitations of a single-variable approach.

Conclusions:

• The cough-based alert system demonstrates potential for capturing a significant proportion of COPD exacerbations in a real-world setting.
• The unobtrusive and hassle-free nature of the monitoring approach was well-received by patients, with no reported adherence issues.
• While effective, the system's reliance on a single variable (cough) suggests potential for integration with other monitoring parameters for enhanced sensitivity.