Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
August 22, 2026BMC Health Services ResearchOpen Access

An approach to classifying and measuring stigmatizing and positive language at scale in electronic health records

View Full Paper
Ask AI
Bookmark
Share

Authors

SSSomnath SahaJohns Hopkins UniversityKHKeith HarrigianJohns Hopkins UniversityBCBrant CheeJohns Hopkins University

Discussion

Loading...

Member takes

Overview

Informatics study demonstrates widespread stigmatizing and templated positive language across millions of electronic health records, highlighting targets for reducing clinician bias.

Key Points

  • Develop and evaluate a natural language processing framework to identify, semantically classify, and quantify stigmatizing and positive language within clinical notes.
  • Created a categorized dictionary across four semantic domains (credibility, demeanor, conformity with care, physical appearance) and annotated 10,244 note examples from two academic health systems to train and validate natural language processing models.
  • Applied the models to 12,841,577 notes from 2016 to 2023 across Emergency Medicine, Internal Medicine, Obstetrics & Gynecology, and Surgery departments to evaluate language prevalence, copy-forward practices, and template usage.
  • Stigmatizing language regarding patient conformity (e.g., 'noncompliant') was the most frequent (4.23% of notes), followed by terms questioning credibility (0.86%) and describing a challenging demeanor (0.82%).
  • Stigmatizing terms were copied forward in up to 25% of notes, while positive descriptors (e.g., 'pleasant') were commonly embedded in documentation templates in up to 65% of notes.

Cite This Study

Saha et al. (2026) studied this question.

synapsesocial.com/papers/6a895f74ca7ade938187e176https://doi.org/10.1186/s12913-026-15358-5
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Detecting and understanding stigmatizing language in electronic health records using natural language processing2026
  2. 2Semantic Similarity Search Approach to Extract Exemplars of Stigmatizing and Positive Language in Obstetric Clinical Notes: Exploratory Study2026
  3. 3Words Matter: Content Analysis of Language Used When Documenting in the Medical Records of Patients in Mental Health2026
  4. 4Characterizing Stigmatizing and Biased Language in Clinical Pharmacist Documentation2025 · 2 citations
  5. 5CARE-SD: Classifier-based analysis for recognizing and eliminating stigmatizing and doubt marker labels in electronic health records: model development and validation2024