The Acute COPD Exacerbation Prediction Tool (ACCEPT): a modelling study.

Area of Science:

• Pulmonary Medicine
• Clinical Epidemiology
• Health Informatics

Background:

• Accurate prediction of exacerbation risk is crucial for personalized care in chronic obstructive pulmonary disease (COPD).
• Existing methods may not fully capture individual patient risk for exacerbations.
• Developing a generalizable model can improve patient outcomes and resource allocation.

Purpose of the Study:

• To develop and validate a generalizable model for predicting the individualised rate and severity of COPD exacerbations.
• To identify key predictors of COPD exacerbations.
• To provide a decision-making tool for clinicians to personalize COPD management.

Main Methods:

• A mixed-effect model was developed using pooled data from three COPD trials.
• Predictors included exacerbation history, demographics, lung function, symptoms, and medication use.
• External validation was performed using data from the Evaluation of COPD Longitudinally to Identify Predictive Surrogate End-points (ECLIPSE) study.

Main Results:

• The model achieved an Area Under the Curve (AUC) of 0.81 for predicting at least two exacerbations and 0.77 for at least one severe exacerbation in the validation cohort.
• Predicted and observed exacerbation rates were similar, demonstrating accurate forecasting.
• The model showed good performance even in patients with a history of exacerbations (AUC 0.73-0.74).

Conclusions:

• The developed model is a reliable tool for predicting individualised COPD exacerbation risk.
• This predictive model can guide personalized treatment strategies to prevent exacerbations.
• The findings support the use of this model in clinical practice for enhanced COPD management.