Article

Half of alcohol, drug, and self-harm presentations cannot be identified in coded emergency department data: a diagnostic accuracy study of a large language model

2026-08-31

Abstract excerpt

<h4>Objective: </h4> To measure the accuracy of clinical coding, clinician review, and a locally deployed large language model (LLM) in identifying alcohol, drug, and self-harm involvement in emergency department (ED) attendances, and quantify prevalence. Design Two-phase diagnostic accuracy study. In a validation week, the identification strategies were assessed against a conflict-adjudicated reference standard (...

Identifiers and source

Literature Corpus work: cbf8c4de-52a6-5e9e-b8cf-bac88105f957
DOI: 10.64898/2026.08.26.26361443

Open publication

Related research

Semantic proximity does not establish scientific evidence. Graph snapshot: 2026-09-29.

Click a neighbor to travelStep 1 · 12 closest

Half of alcohol, drug, and self-harm presentations cannot be identified in coded emergency department data: a diagnostic accuracy study of a large language modelDOI 10.64898/2026.08.26.26361443Read

Comparing computable structured phenotype- versus large language model-identification of opioid use disorder using electronic health record dataSemantic proximity Read

Identifying the Presence and Timing of Self-harm in Electronic Mental Health Records Using Privacy-Preserving Local Language Models: Methodological StudySemantic proximity Read

Computational Phenotypes for Patients with Opioid-Related Disorders Presenting to the Emergency DepartmentSemantic proximity Read

Simulated Reasoning and Self-Verification in Generalist Large Language Models for Psychiatric Diagnostic Performance: Cross-Sectional StudySemantic proximity Read

Making Broad Evidence Synthesis Feasible: An LLM Screening Agent for Meta-Analyses Applied To Suicide PreventionSemantic proximity Read

INSIGHTFUL: Insight Generation through Clinical Annotation, Analysis, and Modeling of Suicide-Related Factors towards Understanding and LifesavingSemantic proximity Read

Detecting Stigmatizing Language in Clinical Notes with Large Language Models for Addiction CareSemantic proximity Read

Detection of Suicidality Through Privacy-Preserving Large Language ModelsSemantic proximity Read

Conversational trajectory degrades large language model detection of suicidal ideation relative to clinicians: a preregistered studySemantic proximity Read

Develop and validate a computable phenotype for identifying alcohol-use disorder patients using structure and unstructured EHR data.Semantic proximity Read

Computational phenotypes for patients with opioid-related disorders presenting to the emergency department.Semantic proximity Read

Robust Application of Supervised Machine Learning Techniques for Cause of Death Determination from Verbal AutopsiesSemantic proximity Read