A Dataset and Resources for Identifying Patient Health Literacy Information from Clinical Notes

TL;DR

This paper introduces HEALIX, the first annotated dataset of clinical notes for health literacy, enabling automated detection and benchmarking of LLMs in assessing patient health literacy levels.

Contribution

It provides a novel, publicly available dataset of clinical notes annotated for health literacy, facilitating research and development of automated detection methods.

Findings

HEALIX contains 589 annotated clinical notes across 9 types.

Benchmarking shows LLMs can classify health literacy with promising results.

Zero-shot and few-shot prompting strategies are effective for health literacy detection.

Abstract

Health literacy is a critical determinant of patient outcomes, yet current screening tools are not always feasible and differ considerably in the number of items, question format, and dimensions of health literacy they capture, making documentation in structured electronic health records difficult to achieve. Automated detection from unstructured clinical notes offers a promising alternative, as these notes often contain richer, more contextual health literacy information, but progress has been limited by the lack of annotated resources. We introduce HEALIX, the first publicly available annotated health literacy dataset derived from real clinical notes, curated through a combination of social worker note sampling, keyword-based filtering, and LLM-based active learning. HEALIX contains 589 notes across 9 note types, annotated with three health literacy labels: low, normal, and high. To demonstrate its utility, we benchmarked zero-shot and few-shot prompting strategies across four open source large language models (LLMs).