Academic Journal
View in EDS

Semantic Similarity Search Approach to Extract Exemplars of Stigmatizing and Positive Language in Obstetric Clinical Notes: Exploratory Study.

Bibliographic Details
Title: Semantic Similarity Search Approach to Extract Exemplars of Stigmatizing and Positive Language in Obstetric Clinical Notes: Exploratory Study.
Authors: Scroggins JK; School of Nursing, University of North Carolina at Chapel Hill, 120 Medical Drive, Chapel Hill, NC, 27514, United States, 1 (919) 966-4260., Hulchafo II; School of Nursing, Columbia University, New York, NY, United States., Barcelona V; School of Nursing, Columbia University, New York, NY, United States., Davoudi A; VNS Health, New York, NY, United States., Moen H; Department of Computer Science, Aalto University, Espoo, Finland., Harkins S; School of Nursing, Columbia University, New York, NY, United States., Scharp D; Icahn School of Medicine at Mount Sinai, New York, NY, United States., Cato K; School of Nursing, University of Pennsylvania, Philadelphia, PA, United States., Tadiello M; Center for Community-Engaged Health Informatics and Data Science, Columbia University Irving Medical Center, New York, NY, United States., Topaz M; School of Nursing, Columbia University, New York, NY, United States.
Source: JMIR medical informatics [JMIR Med Inform] 2026 Sep 22; Vol. 14, pp. e85088. Date of Electronic Publication: 2026 Sep 22.
Publication Type: Journal Article
Language: English
Journal Info: Publisher: JMIR Publications Country of Publication: Canada NLM ID: 101645109 Publication Model: Electronic Cited Medium: Internet ISSN: 2291-9694 (Electronic) Linking ISSN: 22919694 NLM ISO Abbreviation: JMIR Med Inform Subsets: MEDLINE
Imprint Name(s): Original Publication: Toronto : JMIR Publications, [2013]-
MeSH Terms: Electronic Health Records*/statistics & numerical data , Obstetrics*/methods , Natural Language Processing* , Semantics* , Stereotyping*, Humans ; Female ; Pregnancy ; United States ; Language
Abstract: Background: Natural language processing can extract meaningful information from clinical notes. However, human annotation is time-consuming and costly, and scarce data poses a challenge.
Objective: This study aimed to explore a semantic similarity search approach to extract exemplars of stigmatizing and positive language in obstetric clinical notes.
Methods: We used electronic health record data from labor and birth admissions at 2 hospitals in the United States from 2017 to 2019. We used a semantic similarity search approach, which used 200 randomly selected true exemplars, stratified by language categories, as queries to search for similar exemplar candidates. We extracted the top 5 candidates with the highest cosine similarities, which were assessed for accuracy.
Results: We retrieved 1000 candidates. An average precision of 0.69 was achieved when candidates with cosine similarity thresholds of 0.75 or higher were included, at which point 68.8% (64/93) of exemplar candidates accurately represented true cases. At the 0.75 threshold, the proportion of true cases was higher for preferred language (41/56, 73.2%) and unilateral/authoritarian decisions (5/7, 71.4%). The proportion of true cases was lower for difficult patients (2/7, 28.6%) and marginalized identities (3/9, 33.3%).
Conclusions: The semantic similarity search approach shows promise in efficiently extracting exemplars while reducing the annotation burden, laying the groundwork for future applications in other domains.
(© Jihye Kim Scroggins, Ismael Ibrahim Hulchafo, Veronica Barcelona, Anahita Davoudi, Hans Moen, Sarah Harkins, Danielle Scharp, Kenrick Cato, Michele Tadiello, Maxim Topaz. Originally published in JMIR Medical Informatics (https://medinform.jmir.org).)
References: Artif Intell Med. 2023 Dec;146:102701. (PMID: 38042599)
J Med Internet Res. 2026 Mar 25;28:e82997. (PMID: 41880603)
JAMA Netw Open. 2021 Jul 1;4(7):e2117052. (PMID: 34259849)
Matern Child Health J. 2024 Mar;28(3):578-586. (PMID: 38147277)
PLoS One. 2023 Dec 15;18(12):e0292356. (PMID: 38100453)
Obstet Gynecol. 2023 Oct 1;142(4):795-803. (PMID: 37678895)
Health Aff (Millwood). 2022 Feb;41(2):203-211. (PMID: 35044842)
J Am Med Inform Assoc. 2019 Nov 1;26(11):1314-1322. (PMID: 31294792)
J Obstet Gynecol Neonatal Nurs. 2025 Jan;54(1):112-122.e4. (PMID: 39577837)
Issues Ment Health Nurs. 2009 Mar;30(3):199. (PMID: 19291498)
Int J Med Inform. 2017 Oct;106:25-31. (PMID: 28870380)
Nurs Inq. 2023 Jul;30(3):e12557. (PMID: 37073504)
Nurs Outlook. 2026 May-Jun;74(3):102757. (PMID: 41962499)
JAMA Netw Open. 2022 Jan 4;5(1):e2144967. (PMID: 35084481)
JMIR Med Inform. 2020 Nov 27;8(11):e22508. (PMID: 33245284)
Behav Med. 2019 Apr-Jun;45(2):86-101. (PMID: 31343961)
AMIA Annu Symp Proc. 2018 Dec 05;2018:1552-1560. (PMID: 30815201)
JMIR Med Inform. 2021 May 26;9(5):e23099. (PMID: 34037527)
NPJ Digit Med. 2025 Jan 19;8(1):45. (PMID: 39828800)
J Biomed Inform. 2014 Aug;50:162-72. (PMID: 24859155)
J Am Med Inform Assoc. 2025 Feb 01;32(2):308-317. (PMID: 39569431)
J Am Med Inform Assoc. 2014 May-Jun;21(3):406-13. (PMID: 24001514)
Contributed Indexing: Keywords: bias; electronic health records; health communication; natural language processing; stigmatizing language
Entry Date(s): Date Created: 20260922 Date Completed: 20260922 Latest Revision: 20260924
Update Code: 20260924
PubMed Central ID: PMC13597023
DOI: 10.2196/85088
PMID: 42771877
Database: MEDLINE
Description
ISSN:2291-9694
DOI:10.2196/85088