Spurious Correlations and Beyond: Understanding and Mitigating Shortcut Learning in SDOH Extraction with Large Language Models

• URL: http://arxiv.org/abs/2506.00134v1
• Date: Fri, 30 May 2025 18:11:33 GMT
• Title: Spurious Correlations and Beyond: Understanding and Mitigating Shortcut Learning in SDOH Extraction with Large Language Models
• Authors: Fardin Ahsan Sakib, Ziwei Zhu, Karen Trister Grace, Meliha Yetisgen, Ozlem Uzuner,
• Abstract summary: Large language models (LLMs) have shown promise, but they may rely on superficial cues leading to spurious predictions.<n>We demonstrate that mentions of alcohol or smoking can falsely induce models to predict current/past drug use where none is present.<n>We evaluate mitigation strategies - such as prompt engineering and chain-of-thought reasoning - to reduce these false positives.
• Score: 3.3408746880885003
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Social determinants of health (SDOH) extraction from clinical text is critical for downstream healthcare analytics. Although large language models (LLMs) have shown promise, they may rely on superficial cues leading to spurious predictions. Using the MIMIC portion of the SHAC (Social History Annotation Corpus) dataset and focusing on drug status extraction as a case study, we demonstrate that mentions of alcohol or smoking can falsely induce models to predict current/past drug use where none is present, while also uncovering concerning gender disparities in model performance. We further evaluate mitigation strategies - such as prompt engineering and chain-of-thought reasoning - to reduce these false positives, providing insights into enhancing LLM reliability in health domains.

Related papers

• LLM-Augmented Symptom Analysis for Cardiovascular Disease Risk Prediction: A Clinical NLP [2.2615384250361004]
  This study introduces a novel LLM-augmented clinical NLP pipeline that employs domain-adapted large language models for symptom extraction, contextual reasoning, and correlation from free-text reports.<n> Evaluations on MIMIC-III and CARDIO-NLP datasets demonstrate improved performance in precision, recall, F1-score, and AUROC, with high clinical relevance.
  arXiv  Detail & Related papers  (2025-07-15T07:32:16Z)
• LlaMADRS: Prompting Large Language Models for Interview-Based Depression Assessment [75.44934940580112]
  This study introduces LlaMADRS, a novel framework leveraging open-source Large Language Models (LLMs) to automate depression severity assessment.<n>We employ a zero-shot prompting strategy with carefully designed cues to guide the model in interpreting and scoring transcribed clinical interviews.<n>Our approach, tested on 236 real-world interviews, demonstrates strong correlations with clinician assessments.
  arXiv  Detail & Related papers  (2025-01-07T08:49:04Z)
• Reasoning-Enhanced Healthcare Predictions with Knowledge Graph Community Retrieval [61.70489848327436]
  KARE is a novel framework that integrates knowledge graph (KG) community-level retrieval with large language models (LLMs) reasoning.<n>Extensive experiments demonstrate that KARE outperforms leading models by up to 10.8-15.0% on MIMIC-III and 12.6-12.7% on MIMIC-IV for mortality and readmission predictions.
  arXiv  Detail & Related papers  (2024-10-06T18:46:28Z)
• Explainable Biomedical Hypothesis Generation via Retrieval Augmented Generation enabled Large Language Models [46.05020842978823]
  Large Language Models (LLMs) have emerged as powerful tools to navigate this complex data landscape. RAGGED is a comprehensive workflow designed to support investigators with knowledge integration and hypothesis generation.
  arXiv  Detail & Related papers  (2024-07-17T07:44:18Z)
• LLMs-based Few-Shot Disease Predictions using EHR: A Novel Approach Combining Predictive Agent Reasoning and Critical Agent Instruction [38.11497959553319]
  We investigate the feasibility of applying Large Language Models to convert structured patient visit data into natural language narratives. We evaluate the zero-shot and few-shot performance of LLMs using various EHR-prediction-oriented prompting strategies. Our results demonstrate that with the proposed approach, LLMs can achieve decent few-shot performance compared to traditional supervised learning methods in EHR-based disease predictions.
  arXiv  Detail & Related papers  (2024-03-19T18:10:13Z)
• Large Language Model Distilling Medication Recommendation Model [58.94186280631342]
  We harness the powerful semantic comprehension and input-agnostic characteristics of Large Language Models (LLMs)<n>Our research aims to transform existing medication recommendation methodologies using LLMs.<n>To mitigate this, we have developed a feature-level knowledge distillation technique, which transfers the LLM's proficiency to a more compact model.
  arXiv  Detail & Related papers  (2024-02-05T08:25:22Z)
• Clinical Risk Prediction Using Language Models: Benefits And Considerations [23.781690889237794]
  This study focuses on using structured descriptions within vocabularies to make predictions exclusively based on that information. We find that employing LMs to represent structured EHRs leads to improved or at least comparable performance in diverse risk prediction tasks.
  arXiv  Detail & Related papers  (2023-11-29T04:32:19Z)
• MedDiffusion: Boosting Health Risk Prediction via Diffusion-based Data Augmentation [58.93221876843639]
  This paper introduces a novel, end-to-end diffusion-based risk prediction model, named MedDiffusion. It enhances risk prediction performance by creating synthetic patient data during training to enlarge sample space. It discerns hidden relationships between patient visits using a step-wise attention mechanism, enabling the model to automatically retain the most vital information for generating high-quality data.
  arXiv  Detail & Related papers  (2023-10-04T01:36:30Z)
• Interpretable Medical Diagnostics with Structured Data Extraction by Large Language Models [59.89454513692417]
  Tabular data is often hidden in text, particularly in medical diagnostic reports. We propose a novel, simple, and effective methodology for extracting structured tabular data from textual medical reports, called TEMED-LLM. We demonstrate that our approach significantly outperforms state-of-the-art text classification models in medical diagnostics.
  arXiv  Detail & Related papers  (2023-06-08T09:12:28Z)
• Self-Verification Improves Few-Shot Clinical Information Extraction [73.6905567014859]
  Large language models (LLMs) have shown the potential to accelerate clinical curation via few-shot in-context learning. They still struggle with issues regarding accuracy and interpretability, especially in mission-critical domains such as health. Here, we explore a general mitigation framework using self-verification, which leverages the LLM to provide provenance for its own extraction and check its own outputs.
  arXiv  Detail & Related papers  (2023-05-30T22:05:11Z)
• Deep Stable Representation Learning on Electronic Health Records [8.256340233221112]
  Causal Healthcare Embedding (CHE) aims at eliminating the spurious statistical relationship by removing the dependencies between diagnoses and procedures. Our proposed CHE method can be used as a flexible plug-and-play module that can enhance existing deep learning models on EHR.
  arXiv  Detail & Related papers  (2022-09-03T04:10:45Z)
• Label Dependent Attention Model for Disease Risk Prediction Using Multimodal Electronic Health Records [8.854691034104071]
  Disease risk prediction has attracted increasing attention in the field of modern healthcare. One challenge of applying AI models for risk prediction lies in generating interpretable evidence. We propose the method of jointly embedding words and labels.
  arXiv  Detail & Related papers  (2022-01-18T07:21:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.