Social Determinants of Health Prediction for ICD-9 Code with Reasoning Models

TL;DR

This paper investigates the use of reasoning models and large language models to predict Social Determinants of Health ICD-9 codes from hospital admission notes, addressing challenges in long text dependencies and improving classification accuracy.

Contribution

It introduces a novel approach combining reasoning models with large language models for multi-label SDoH ICD-9 code prediction on MIMIC-III data, including analysis of missing codes and reproducible code.

Findings

Achieved 89% F1 score in ICD-9 code prediction

Identified missing SDoH codes in 139 admissions

Provided reproducible code for the methodology

Abstract

Social Determinants of Health correlate with patient outcomes but are rarely captured in structured data. Recent attention has been given to automatically extracting these markers from clinical text to supplement diagnostic systems with knowledge of patients' social circumstances. Large language models demonstrate strong performance in identifying Social Determinants of Health labels from sentences. However, prediction in large admissions or longitudinal notes is challenging given long distance dependencies. In this paper, we explore hospital admission multi-label Social Determinants of Health ICD-9 code classification on the MIMIC-III dataset using reasoning models and traditional large language models. We exploit existing ICD-9 codes for prediction on admissions, which achieved an 89% F1. Our contributions include our findings, missing SDoH codes in 139 admissions, and code to reproduce the results.