Key result
Machine learning model predicts discharge destinations with ~82% accuracy, outperforming standard of care.
Why the study?
Comprehensive geriatric assessment improves discharge planning for older surgical patients but faces resource constraints, raising the question of whether machine learning can support individualized discharge recommendations.
Does a machine learning model improve the accuracy of discharge destination recommendations compared to standard-of-care in geriatric surgical inpatients?
Observational (n=169)
No
Does a machine learning model improve the accuracy of discharge destination recommendations compared to standard-of-care in geriatric surgical inpatients?
Absolute Event Rate: 82% vs 73.4%
p-value: p=0.041
A machine learning model using electronic health record and geriatric assessment data can accurately predict discharge destinations for older surgical patients, outperforming standard-of-care decisions.
ML-based discharge predictions may aid geriatric surgical planning; hypothesis-generating and should not yet change practice.
PURPOSE: Older adults undergoing surgery are at high risk for adverse outcomes due to age-related vulnerabilities. Comprehensive geriatric assessment (CGA) can improve discharge planning but is often limited by resource constraints. We evaluated whether machine learning (ML) can support individualized discharge destination recommendations for geriatric surgical inpatients. METHODS: We developed and internally validated an AdaBoost classifier using electronic health record and CGA data from 169 patients aged ≥ 70 years undergoing surgery at a single center. The model predicted discharge destinations (home, acute geriatric care unit, rehabilitation facility, nursing home) and was evaluated using accuracy, receiver operating characteristic (ROC) curves, and calibration within a fivefold cross-validation framework. Model performance was also compared with previously reported standard-of-care discharge decisions. RESULTS: The study population consisted of 169 older adults, with mean age of 80.5 years (SD ± 6.3). The ML model achieved a micro-averaged ROC area under the curve of 0.94 [95% CI 0.92-0.96] and an accuracy of 82% [95% CI 0.76-0.86], representing a significant improvement over standard-of-care decisions (138 vs 124 correct, p = 0.041). Mimicking a human-in-the-loop strategy, we excluded the 15% of predictions with the highest uncertainty, which increased accuracy for the remaining predictions to 85% (p = 0.006). Model inspection identified Barthel Index, Clinical Frailty Scale, age, and number of medications as key predictors. CONCLUSION: Our findings provide a proof-of-concept that an ML-based approach can effectively support individualized discharge planning for geriatric surgical patients. With 82% congruence with expert recommendations, the model significantly outperformed the standard of care. Its real-world clinical impact is currently being evaluated in an ongoing interventional trial. TRIAL REGISTRATION: German clinical trials registry (DRKS00030684), registered on 21st November 2022.
No takes yet. Share an insight, caveat, or question.
Kocar et al. (2026) conducted an observational in Geriatric surgical inpatients (n=169). Machine learning (AdaBoost classifier) recommendations vs. Standard-of-care discharge decisions was evaluated on Accuracy of discharge destination prediction (95% CI 0.76-0.86, p=0.041). An AdaBoost machine learning model predicted discharge destinations with 82% accuracy (95% CI 0.76-0.86), significantly outperforming standard-of-care decisions (138 vs 124 correct; p=0.041).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: