Predicting Continuity of Asthma Care Using a Machine Learning Model: Retrospective Cohort Study.

Area of Science:

• Health Informatics
• Machine Learning in Healthcare
• Chronic Disease Management

Background:

• Continuity of care (COC) offers significant health benefits for chronic conditions.
• Predictive methodologies for determining patient COC are underdeveloped.
• Establishing COC levels aids clinical decisions and resource allocation.

Purpose of the Study:

• To develop a machine learning model for predicting future COC in asthma patients.
• To identify factors associated with continuity of care in asthma.
• To address the research gap in generalizable predictive methods for COC.

Main Methods:

• Utilized data from 31,724 adult asthma outpatients (2011-2018).
• Engineered a machine learning model using 138 features.
• Validated the model with 10-fold cross-validations.

Main Results:

• The predictive model achieved 88.20% accuracy.
• The model demonstrated an average area under the ROC curve of 0.96 and an F1 score of 0.86.
• Asthma severity, comorbidities, insurance, and age were significant predictors of COC.

Conclusions:

• A high-performing machine learning model for predicting asthma patient COC was developed.
• The model's findings can inform clinical decisions and healthcare resource management.
• Further optimization could enhance clinical practice and patient outcomes.