Key result
XGBoost predicts in-hospital heart failure mortality with ~0.86 AUC, outperforming logistic regression.
Why the study?
Utilizing administrative data may facilitate risk prediction in heart failure inpatients, prompting the evaluation of machine learning models to predict in-hospital mortality on an individual basis.
Do machine learning algorithms using administrative data improve the prediction of in-hospital mortality compared to logistic regression in hospitalized patients with heart failure?
Cohort (n=59,074)
Yes
Do machine learning algorithms using administrative data improve the prediction of in-hospital mortality compared to logistic regression in hospitalized patients with heart failure?
Machine learning models—particularly XGBoost and GBM—applied to routine administrative data accurately predict in-hospital mortality in heart failure patients (AUC ~0.855-0.856), performing comparably to or slightly better than tuned logistic regression (AUC 0.853).
Answers are grounded in this paper and cite their sources.
XGBoost may modestly improve in-hospital HF mortality prediction from administrative data; leaves open need for external validation before clinical use.
Aims: Utilizing administrative data may facilitate risk prediction in heart failure inpatients. In this short report, we present different machine learning models that predict in-hospital mortality on an individual basis utilizing this widely available data source. Methods and results: Inpatient cases with a main discharge diagnosis of heart failure hospitalized between 1 January 2016 and 31 December 2018 in one of 86 German Helios hospitals were examined. Comorbidities were defined by ICD-10 codes from administrative data. The data set was randomly split into 75/25% portions for model development and testing. Five algorithms were evaluated: logistic regression [generalized linear models (GLMs)], random forest (RF), gradient boosting machine (GBM), single-layer neural network (NNET), and extreme gradient boosting (XGBoost). After model tuning, the receiver operating characteristics area under the curves (ROC AUCs) were calculated and compared with DeLong's test. A total of 59 074 inpatient cases (mean age 77.6 ± 11.1 years, 51.9% female, 89.4% NYHA Class III/IV) were included and in-hospital mortality was 6.2%. In the test data set, calculated ROC AUCs were 0.853 [95% confidence interval (CI) 0.842-0.863] for GLM, 0.851 (95% CI 0.840-0.862) for RF, 0.855 (95% CI 0.844-0.865) for GBM, 0.836 (95% CI 0.823-0.849) for NNET, and 0.856 (95% CI 9.846-0.867) for XGBoost. XGBoost outperformed all models except GBM. Conclusion: Machine learning-based processing of administrative data enables the creation of well-performing prediction models for in-hospital mortality in heart failure patients.
No takes yet. Share an insight, caveat, or question.
König et al. (2022) conducted a cohort in Heart failure (n=59,074). Machine learning algorithms (RF, GBM, NNET, XGBoost) using administrative data vs. Logistic regression (generalized linear model [GLM]) was evaluated on In-hospital mortality prediction (ROC AUC) (95% CI 0.846-0.867, p=0.023). Extreme gradient boosting (XGBoost) predicted in-hospital heart failure mortality with an ROC AUC of 0.856 (95% CI 0.846-0.867), outperforming logistic regression (AUC 0.853, P=0.023).
Synapse has enriched 10 related papers. Consider them for comparative context: