SDOH-NLI: a Dataset for Inferring Social Determinants of Health from Clinical Notes

Alvin Rajkomar
Eric Loreaux
Ming-Jun Chen
Findings of EMNLP(2023)


Social and behavioral determinants of health (SDOH) play a significant role in shaping health outcomes, and extracting these determinants from clinical notes is a first step to help healthcare providers systematically identify opportunities to provide appropriate care and address disparities. Progress on using NLP methods for this task has been hindered by the lack of high-quality public datasets, largely due to the privacy and regulatory constraints on the use of real patient data. This paper introduces a new dataset, SDOH-NLI, that is based on fully public data. We formulate SDOH extraction as a natural language inference task, and provide binary textual entailment labels obtained from human raters for a cross product of a set of social history snippets as premises and SDOH factors as hypotheses. Our dataset differs from standard NLI benchmark in that our premises and hypotheses are obtained independently. We evaluate both "off-the-shelf" entailment models as well as models fine-tuned on our data, and highlight the ways in which our dataset appears more challenging than commonly used NLI datasets.