JMIR Publications

ABSTRACT

Background:

Large language model (LLM) have demonstrated potential in automating the analysis of unstructured clinical data, yet their application in rehabilitation trigae for work injury cases remains underexplored.

Objective:

We aimed to evaluate the performance of an LLM-assisted approach for the rapid identification of anomalous rehabilitation cases of work injury, aiming to enhance scalability and precision in case management.

Methods:

We retrospectively analysed 110,346 de-identified work-injury cases between 2001 year, and 2024 year from a leading rehabilitation coordination company in Hong Kong, representing approximately 20% of all work injury incidents in the region. Large Language Models were used to estimate the expected duration of recovery based on free-text injury descriptions. Cases in which the actual number of medically certified sick leave days exceeded the LLM-predicted maximum were classified as anomalies.

Results:

The LLM-assisted method achieved high accuracy, with GPT-4o achieving over 73% accuracy in normality classification and 79% accuracy in all dataset detection, outperforming comparator models. The model maintained high accuracy across subgroups and demonstrated reliable extraction of information from free-text notes.

Conclusions:

The proposed method demonstrated robustness when evaluated on a large-scale dataset with a bimodal age distribution. This study highlights the potential of LLMs to transform rehabilitation workflows by automating anomaly detection at scale. The method also shows promise in tailoring rehabilitation strategies to age-specific needs and leveraging LLM tools for efficient case management. However, a key limitation is that the dataset includes only injury cases from a single geographic region, which may limit the generalizability of the findings to other populations or healthcare systems. Clinical Trial: NA