Tipping the Balance: Impact of Class Imbalance Correction on the Performance of Clinical Risk Prediction Models

TL;DR

This study systematically evaluated the impact of class imbalance correction techniques on clinical risk prediction models, finding that these methods do not improve discrimination and often worsen calibration across diverse real-world datasets.

Contribution

It provides comprehensive evidence that common resampling strategies do not enhance model discrimination and impair calibration in clinical risk prediction tasks.

Findings

Resampling strategies showed no consistent improvement in ROC-AUC.

Imbalance correction degraded calibration and increased Brier scores.

Models with imbalance correction exhibited systematic risk prediction distortions.

Abstract

Objective: ML-based clinical risk prediction models are increasingly used to support decision-making in healthcare. While class-imbalance correction techniques are commonly applied to improve model performance in settings with rare outcomes, their impact on probabilistic calibration remains insufficiently understood. This study evaluated the effect of widely used resampling strategies on both discrimination and calibration across real-world clinical prediction tasks. Methods: Ten clinical datasets spanning diverse medical domains and including 605,842 patients were analyzed. Multiple machine-learning model families, including linear models and several non-linear approaches, were evaluated. Models were trained on the original data and under three commonly used 1:1 class-imbalance correction strategies (SMOTE, RUS, ROS). Performance was assessed on held-out data using discrimination and calibration metrics. Results: Across all datasets and model families, resampling had no positive impact on predictive performance. Changes in the Receiver Operating Characteristic Area Under Curve (ROC-AUC) relative to models trained on the original data were small and inconsistent (ROS: -0.002, p<0.05; RUS: -0.004, p>0.05; SMOTE: -0.01, p<0.05), with no resampling strategy demonstrating a systematic improvement. In contrast, resampling in general degraded the calibration performance. Models trained using imbalance correction exhibited higher Brier scores (0.029 to 0.080, p<0.05), reflecting poorer probabilistic accuracy, and marked deviations in calibration intercept and slope, indicating systematic distortions of predicted risk despite preserved rank-based performance. Conclusion: In a diverse set of real-world clinical prediction tasks, commonly used class-imbalance correction techniques did not provide generalizable improvements in discrimination and were associated with degraded calibration.