JMIR Publications

ABSTRACT

Background:

Transportation insecurity is a key barrier to accessing ophthalmic care, impacting health outcomes. Traditional methods to identify these barriers are limited. This study explores the use of natural language processing (NLP) to detect transportation insecurity from free-text electronic health record (EHR) data.

Objective:

To identify transportation insecurity in the free text of ophthalmology clinic notes using NLP.

Methods:

This retrospective, cross-sectional study examined ophthalmology clinic notes from adult patients with an encounter at a tertiary academic eye center from 2016-2023. Demographic information and free text from clinic notes were extracted from the EHR and deidentified for analysis. Free text was utilized to develop a rule-based NLP algorithm to identify transportation insecurity. The NLP algorithm was trained and validated using a gold standard of expert review, and precision, recall, and F1 score were used to evaluate the algorithm performance. Logistic regression evaluated associations between demographics and transportation insecurity.

Results:

A total of 1,801,572 encounter notes from 118,518 unique patients were examined, and the NLP algorithm identified 726 patients (0.6%) with transportation insecurity. The algorithm’s precision, recall, and F1 score were 0.860, 0.960, 0.778, respectively, indicating high agreement with gold-standard chart review. Patients with identified transportation insecurity were more likely to be older (OR=3.01, 95% CI: 2.38-3.78 for 80+ vs 18-60 years) and less likely to identify as Asian (OR=0.04, 95% CI: 0-0.18) compared to White. There was no difference by sex (OR=1.13, 95% CI: 0.97-1.31) or between Black and White race (OR=0.98, 95% CI: 0.79-1.22).

Conclusions:

NLP has the potential to identify patients experiencing transportation insecurity from ophthalmology clinic notes, which may help to facilitate referrals to transportation resources.