Prediction of 30-Day Readmission for COPD Patients Using Accelerometer-Based Activity Monitoring.

Area of Science:

• Pulmonary Medicine
• Biomedical Engineering
• Data Science

Background:

• Chronic Obstructive Pulmonary Disease (COPD) is a leading cause of death globally.
• COPD patients have the highest 30-day hospital readmission risk among chronic diseases.
• Current clinical assessments fail to accurately predict this readmission risk.

Purpose of the Study:

• To develop a statistical model for predicting 30-day readmission risk in COPD patients.
• To utilize physical activity (PA) data from wearable devices for risk prediction.
• To introduce a novel 'quality of activity' (QoA) parameter for enhanced prediction.

Main Methods:

• A statistical model was developed based on previously established activity index (AI) and regularity index (RI) models.
• The model incorporates a new parameter, quality of activity (QoA), combining AI, RI, and other activity-based indices.
• Continuous PA monitoring using accelerometer-based wrist-worn devices was conducted on 16 COPD patients post-discharge.

Main Results:

• The proposed model demonstrated a sensitivity of 63% and a positive prediction rate of 37.78% in predicting 30-day readmissions.
• The model showed significant improvement compared to existing clinical assessment methods.
• Physical activity monitoring proved effective in identifying high-risk COPD patients.

Conclusions:

• Wearable device-based PA monitoring offers a promising approach for predicting COPD readmission risk.
• The novel QoA parameter enhances the accuracy of readmission risk prediction.
• This method has the potential to improve patient management and reduce healthcare costs associated with COPD readmissions.