Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.19082 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908901434195968 |
|---|---|
| author | Bittner, Madeline Demner-Fushman, Dina Shabazz, Yasmeen Bartels, Davis Yoon, Dukyong Quitadamo, Brad Menghrajani, Rajiv Celi, Leo Soni, Sarvesh |
| author_facet | Bittner, Madeline Demner-Fushman, Dina Shabazz, Yasmeen Bartels, Davis Yoon, Dukyong Quitadamo, Brad Menghrajani, Rajiv Celi, Leo Soni, Sarvesh |
| contents | 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). |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_19082 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | A Dataset and Resources for Identifying Patient Health Literacy Information from Clinical Notes Bittner, Madeline Demner-Fushman, Dina Shabazz, Yasmeen Bartels, Davis Yoon, Dukyong Quitadamo, Brad Menghrajani, Rajiv Celi, Leo Soni, Sarvesh Computation and Language 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). |
| title | A Dataset and Resources for Identifying Patient Health Literacy Information from Clinical Notes |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2603.19082 |