Optimizing elderly care: A data-driven AI model for predicting polypharmacy risk in the elderly using SHARE data

Background: Aging is frequently accompanied by multimorbidity, the presence of multiple chronic conditions, which contributes to declines in both cognitive and physical function and presents complex health challenges. One such challenge is Polypharmacy (PP), defined as the concurrent use of more than five medications. Methods: We used data from participants older than 50 years who were present in wave 6 and at least one of the subsequent three waves of the SHARE study, aiming to predict PP risk at 2, 4, and 6-year intervals. We selected the predictor variables using LASSO regression and evaluated eight ML models using a rigorous cross-validation strategy to ensure robustness and reliability. Findings: Our analysis reveals an upward trend in PP prevalence across the surveyed countries, with aggregate figures rising from 34.03% (95% CI 33.1-34.9) in wave 7 to 36.75% (95% CI 35.6-37.9) in wave 8, reaching 39.91% (95% CI 38.9-40.9) in wave 9. LASSO regression identified 17 key predictors of PP risk, which were related to socio-demographic factors, lifestyle factors, physical and mental health, and disease history. Among the models evaluated, the Categorical Boosting ML model performed best, yielding overall accuracies of 75.08%, 73.7%, and 71.65% and recall rates of 72.83%, 70.48%, and 67.96% for the 2, 4, and 6-year intervals, respectively. Interpretation: This study uncovers a rising trend of PP. It demonstrated the potential of using longitudinal data and ML to predict PP. Moreover, our findings suggest that mental health is an important factor to consider when addressing PP.