Evidence map›Paper›PMID 40470194›Full record

ArticleResearch square2025

Enhancing Substance Use Detection in Clinical Notes with Large Language Models.

Fabrice Harel-Canada, Anabel Salimian, Brandon Moghanian, Sarah Clingan, Allan Nguyen, Tucker Avra, Michelle Poimboeuf, Ruby Romero, Arthur Funnell, Panayiotis Petousis and 4 more

Abstract readPreprint
PubMed Full text at PMC Publisher
In one paragraph

Article in Research square, 2025. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Not yet cited in PubMed.

0numbers the graph read from it
0cells of the map it votes in
0citing papers in PubMed
–field-weighted citation impact
1 · What the graph read from it

What it found

Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.

The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.

2 · The registry

The trial behind it

Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.

Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.

3 · Its place in the literature

Who cites it

0 citing papers in PubMed.

No citing paper in PubMed yet.

4 · The record

Corrections and comments

5 · Who and what money

Authors and funding

14 authors.

Fabrice Harel-CanadaComputer Science Department, University of California, Los Angeles, 404 Westwood Plaza Suite 277, Los Angeles, 90095, CA, USA.
Anabel SalimianSemel Institute for Neuroscience and Human Behavior at University of California, Los Angeles, 760 Westwood Plaza, Los Angeles, 90024, CA, USA.
Brandon MoghanianUniversity of California, Los Angeles, 200 Medical Plaza Suite 365C, Los Angeles, 90024, CA, USA.
Sarah ClinganIntegrated Substance Abuse Programs at University of California, Los Angeles, 10911 Weyburn Ave, Ste. 200, Los Angeles, 90024, CA, USA.
Allan NguyenUniversity of California, Los Angeles, 200 Medical Plaza Suite 365C, Los Angeles, 90024, CA, USA.
Tucker AvraDavid Geffen School of Medicine at University of California, Los Angeles, 10833 Le Conte Ave, Los Angeles, 90095, CA, USA.
Michelle PoimboeufDivision of General Internal Medicine and Health Services Research, University of California, Los Angeles, 1100 Glendon Ave STE 850, Los Angeles, 90024, CA, USA.
Ruby RomeroDivision of General Internal Medicine and Health Services Research, University of California, Los Angeles, 1100 Glendon Ave STE 850, Los Angeles, 90024, CA, USA.
Arthur FunnellClinical and Translational Science Institute, University of California, Los Angeles, 924 Westwood Blvd Suite 420, Los Angeles, 90024, CA, USA.
Panayiotis PetousisClinical and Translational Science Institute, University of California, Los Angeles, 924 Westwood Blvd Suite 420, Los Angeles, 90024, CA, USA.
Michael ShinDepartment of Geography, University of California, Los Angeles, 1255 Bunche Hall, Los Angeles, 90095, CA, USA.
Nanyun PengComputer Science Department, University of California, Los Angeles, 404 Westwood Plaza Suite 277, Los Angeles, 90095, CA, USA.
Chelsea L ShoverDivision of General Internal Medicine and Health Services Research, University of California, Los Angeles, 1100 Glendon Ave STE 850, Los Angeles, 90024, CA, USA.
David Goodman-MezaKirby Institute, University of New South Wales, Wallace Wurth Building (C27), Cnr High St & Botany St, UNSW, Sydney, 2052, NSW, Australia.

Funding

Predicting fatal and non-fatal overdose in Los Angeles County with Rapid Overdose Surveillance Dashboard to target street-based addiction treatment and harm reduction servicesR01DA057630 · NIDA · UNIVERSITY OF CALIFORNIA LOS ANGELES · PI David Goodman, Chelsea Leigh Shover · 2022 to 2026
$2.9M
Using data science to measure the impact of opioid agonist therapy in patients admitted with Staphylococcus aureus bloodstream infectionsK08DA048163 · NIDA · UNIVERSITY OF CALIFORNIA LOS ANGELES · PI GOODMAN, DAVID · 2019 to 2023
$1.0M
Adoption and scale-up of long-acting medications for opioid use disorder by U.S. clinicians.K01DA050771 · NIDA · UNIVERSITY OF CALIFORNIA LOS ANGELES · PI SHOVER, CHELSEA LEIGH · 2021 to 2025
$893k
NIDA NIH HHS K01 DA050771NIDA NIH HHS K08 DA048163NIDA NIH HHS R01 DA057630
6 · The paper itself

Abstract

Identifying substance use behaviors in electronic health records (EHRs) is challenging because critical details are often buried in unstructured notes that use varied terminology and negation, requiring careful contextual interpretation to distinguish relevant use from historical mentions or denials. Using MIMIC-III/IV discharge summaries, we created a large, annotated drug detection dataset to tackle this problem and support future systemic substance use surveillance. We then investigated the performance of multiple large language models (LLMs) for detecting eight substance use categories within this data. Evaluating models in zero-shot, few-shot, and fine-tuning configurations, we found that a fine-tuned model, Llama-DrugDetector-70B, outperformed others. It achieved near-perfect F1-scores (≥ 0.95) for most individual substances and strong scores for more complex tasks like prescription opioid misuse (F1=0.815) and polysubstance use (F1=0.917). These findings demonstrate that LLMs significantly enhance detection, showing promise for clinical decision support and research, although further work on scalability is warranted.

Indexed as

drug usenatural language processingNLPpeople who inject drugssubstance use

Identifiers

PMID40470194
PMCPMC12136207

What OpenQuestion holds

Textmetadata
LicenceCC BY
Read underepoch 390

Registered trials

None linked

Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.