Key result
XGBoost model outperforms RISQ-PATH score for predicting drug-induced long QTc, achieving ~0.86 AUROC.
Why the study?
Current predictors of drug-induced long QTc have limited generalizability and modest performance, necessitating improved risk assessment tools.
Do machine learning-based models improve the prediction of drug-induced QTc prolongation compared to current clinical risk calculators in patients starting QTc-prolonging medications?
Observational (n=345,371)
Do machine learning-based models improve the prediction of drug-induced QTc prolongation compared to current clinical risk calculators in patients starting QTc-prolonging medications?
Machine learning models using EHR and ECG data significantly outperform traditional risk scores like Tisdale and RISQ-PATH in predicting drug-induced QTc prolongation.
ML models may enhance QTc risk stratification in EHRs; hypothesis-generating and requires prospective validation before adoption.
BACKGROUND: Many medications are associated with long QTc. Current long QTc predictors have limited generalizability and/or modest performance. OBJECTIVE: This study aimed to compare the performance of machine learning-enabled approaches to drug-induced long QTc prediction with current risk scores, including Tisdale and RISQ-PATH. METHODS: We identified patients who began QTc-prolonging medications (QTdrugs) and had follow-up 12-lead electrocardiograms (ECGs) within 1 year. Using 5-fold cross-validation, we trained XGBoost (XGB), an ECG-based deep neural network, and combined models using electronic health record data and ECG traces to predict QTc of ≥500 ms within 1 year. We assessed performance using the area under the receiver operating characteristic (AUROC) curves and compared them with corresponding Tisdale and RISQ-PATH scores. RESULTS: QTdrug records were identified for 345,371 patients, 5.7% of whom had long QTc within 1 year. In the subset with baseline ECGs available (N = 182,448; 7.7% events), both the XGB and deep neural network models demonstrated high performance (AUROC 0.869 and 0.864, respectively) but their combination yielded no significant improvement (AUROC 0.874). Therefore, focusing on the XGB model, we observed superior performance vs RISQ-PATH in the overall population (AUROC 0.859 vs 0.701), as well as Tisdale in predominantly inpatients (N = 110,558; 8.8% events; AUROC 0.855 vs 0.770). Positive predictive value was 61.5% vs 28.35% and 54.0% vs 28.3% at equivalent operating points for the XGB model vs RISQ-PATH and Tisdale, respectively. CONCLUSION: Development and retrospective validation of 3 machine learning-based models for predicting drug-induced long QTc at the time of new QTdrug starts demonstrated superior performance compared with current clinical risk calculators and may be useful tools to support medical decision making when initiating new therapies.
No takes yet. Share an insight, caveat, or question.
Jing et al. (2025) conducted an observational in Drug-induced long QTc (n=345,371). XGBoost machine learning model vs. RISQ-PATH and Tisdale risk scores was evaluated on QTc of ≥500 ms within 1 year. An XGBoost machine learning model demonstrated superior performance for predicting drug-induced long QTc within 1 year compared with the RISQ-PATH score (AUROC 0.859 vs 0.701).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: