Resource-Conscious Modeling for Next- Day Discharge Prediction Using Clinical Notes
For clinical resource management, this work shows that interpretable, resource-efficient models can outperform compact LLMs in imbalanced prediction tasks, but the performance is modest and incremental.
This study evaluates lightweight models for predicting next-day discharge in elective spine surgery using clinical notes. TF-IDF with LGBM achieved the best performance with an F1-score of 0.47, recall of 0.51, and AUC-ROC of 0.80, outperforming compact LLMs.
Timely discharge prediction is essential for optimizing bed turnover and resource allocation in elective spine surgery units. This study evaluates the feasibility of lightweight, fine-tuned large language models (LLMs) and traditional text-based models for predicting next-day discharge using postoperative clinical notes. We compared 13 models, including TF-IDF with XGBoost and LGBM, and compact LLMs (DistilGPT-2, Bio_ClinicalBERT) fine-tuned via LoRA. TF-IDF with LGBM achieved the best balance, with an F1-score of 0.47 for the discharge class, a recall of 0.51, and the highest AUC-ROC (0.80). While LoRA improved recall in DistilGPT2, overall transformer-based and generative models underperformed. These findings suggest interpretable, resource-efficient models may outperform compact LLMs in real-world, imbalanced clinical prediction tasks.