Introduction: Adherence is important in many aspects of healthcare, including pulmonary rehabilitation (PR), as it is related to clinical outcomes, and to the (economic) burden for healthcare providers.

Aim: Understanding who are adherent and who are non-adherent could be helpful to differentiate between patients who need more or less support. Therefore, the aim of this study was to develop and validate a model to predict exercise adherence to PR in patients with COPD.

Methods: A multivariable logistic regression model for exercise adherence was developed. Eight candidate predictors were obtained in a prospective cohort study from 196 patients with COPD following PR in 53 primary physiotherapy practices in the Netherlands and Belgium. To create a parsimonious model, variable selection using backward selection was performed with a p-value of > 0.05 for elimination. Model performance was assessed by discrimination, calibration and clinical utility. Internal validation was assessed by bootstrapping.

Results: The final model included four predictors; intention, depression, MRC-score and alliance. The optimism-corrected AUC, after bootstrap internal validation was 0.79 (95% CI, 0.72-0.85). Calibration plots suggested good calibration and decision curve analysis showed great net benefit in a wide range of risk thresholds.

Conclusion: The exercise adherence prediction model has potential for clinical utility to predict adherence in patients with COPD. Information from this model can be used to facilitate discussions regarding clinical care and target services to better manage COPD and make more efficient use of health care.