Evaluating the Potential of Machine Learning for Discharge Management on Routine Health Insurance Data

Clinical Snapshot

Key Findings

• P Value: Not reported in abstract

• Effect Size: AdaBoost achieved best overall performance, logistic regression performed comparably

• Primary Outcome: Prediction of mortality and unplanned readmissions

• Nnt Or Sensitivity: Significant performance gain with diagnostic/procedural information (M2) vs demographics alone (M1)

• Confidence Interval: Not reported in abstract

Clinical Application

Requires access to comprehensive health insurance claims data and ML infrastructure Limited applicability due to different healthcare system structure; Australian Medicare and private insurance data architecture differs significantly from German model Hospitalized patients requiring discharge planning and readmission risk assessment

Abstract

Machine learning (ML) has great potential in healthcare, especially with large structured data. Routine health insurance claims (HIC) data are a valuable resource, comprising standardized longitudinal patient information. However, to fully leverage ML in HIC, challenges such as large data volumes, variability, data gaps, sparsity, and high dimensionality must be addressed. To overcome these barriers, our publicly funded KI-THRUST project accessed a large-scale dataset of 2 million anonymized routine insurance records from four German providers. Subsequently, a thorough performance evaluation of various linear and nonlinear ML models (logistic regression, AdaBoost, Random Forest, Neural Network) for their potential to predict clinical outcomes such as mortality and unplanned readmissions was conducted, and the results were interpreted using explainable AI methods (Integrated Gradients, Shapley values, LIME). In particular, we investigated the predictive value of feature sets comprising base demographics (M1) and ICD-10 codes for previous conditions and procedures (M2, M3). While AdaBoost achieved the best overall performance, the less complex logistic regression performed comparably. The additional diagnostic and procedural information in M2 led to a significant performance gain, whereas additional time-related information did not. Overall, our results show the potential of large-scale HIC data when combined with ML methods and XAI. Ultimately, paving the way for future applications like discharge management systems and other healthcare areas.

References

1. 1.Ritter, Z., Maurer, M. C., Metsch, J. M., Weller, L., Pollmann, T., Kretzler, M., Grobe, T., & Hauschild, A. C. (2026). Evaluating the Potential of Machine Learning for Discharge Management on Routine Health Insurance Data. Studies in Health Technology and Informatics. https://doi.org/10.3233/SHTI260183