JMIR Publications

ABSTRACT

Background:

Musculoskeletal conditions are managed within primary care but patients can be referred to secondary care if a specialist opinion is required. The ever increasing demand of healthcare resources emphasizes the need to streamline care pathways with the ultimate aim of ensuring that patients receive timely and optimal care. Information contained in referral letters underpins the referral decision-making process but is yet to be explored systematically for the purposes of treatment prioritization for musculoskeletal conditions.

Objective:

This study aims to explore the feasibility of using natural language processing and machine learning to automate triage of patients with musculoskeletal conditions by analyzing information from referral letters. Specifically, we aim to determine whether referral letters can be automatically assorted into latent topics that are clinically relevant, i.e. considered relevant when prescribing treatments. Here, clinical relevance is assessed by posing two research questions. Can latent topics be used to automatically predict the treatment? Can clinicians interpret latent topics as cohorts of patients who share common characteristics or experience such as medical history, demographics and possible treatments?

Methods:

We used latent Dirichlet allocation to model each referral letter as a finite mixture over an underlying set of topics and model each topic as an infinite mixture over an underlying set of topic probabilities. The topic model was evaluated in the context of automating patient triage. Given a set of treatment outcomes, a binary classifier was trained for each outcome using previously extracted topics as the input features of the machine learning algorithm. In addition, qualitative evaluation was performed to assess human interpretability of topics.

Results:

The prediction accuracy of binary classifiers outperformed the stratified random classifier by a large margin giving an indication that topic modelling could be used to predict the treatment thus effectively supporting patient triage. Qualitative evaluation confirmed high clinical interpretability of the topic model.

Conclusions:

The results established the feasibility of using natural language processing and machine learning to automate triage of patients with knee and/or hip pain by analyzing information from their referral letters.