MétaCan
Menu
← Back to cohort
Record W4416055696 · doi:10.1101/2025.11.07.25339620

Ambient Only vs. Longitudinal Data-Enhanced AI Documentation: A Pilot Study Quantifying the Value of Historical Clinical Context in Primary Care

2025· preprint· W4416055696 on OpenAlexaff
Michael Zuckerman, Gal Eyal, Roei Magen, Omer Harnof, Shiri Shifman, Zach Avraham, Kevin Gallagher, Yair E. Lewis

Bibliographic record

VenuemedRxiv · 2025
Typepreprint
Language
FieldHealth Professions
TopicElectronic Health Records Systems
Canadian institutionsHeadwaters Health Care Centre
Fundersnot available
KeywordsDocumentationWorkflowContext (archaeology)ConversationCompleteness (order theory)WorkaroundObservational studyScale (ratio)

Abstract

fetched live from OpenAlex

Abstract Background Ambient artificial intelligence (AI) clinical documentation tools have gained rapid adoption in healthcare to address physician burnout from documentation burden. However, current implementations primarily rely on real-time audio capture without systematically incorporating longitudinal patient data, potentially limiting documentation completeness for chronic disease management. Objective To compare documentation completeness between ambient audio-only workflows and those augmented with historical clinical data from electronic health records (EHRs) for type 2 diabetes and hypertension encounters in primary care. Methods We conducted a retrospective, paired, cross-sectional study of 354 primary care encounters in which diabetes mellitus (DM, n=119) and/or hypertension (HTN, n=281) were treated. Each condition instance was analysed twice to compare two methods of automated documentation: using only physician-patient conversation transcripts (termed “ambient only”) compared with consolidated automated documentation that includes historical clinical data in addition to the ambient conversation (ambient + history; termed “consolidated”). Documentation completeness was assessed using the “assessment” subset of the QNOTE clinical documentation quality measurement instrument, evaluating four domains: completeness, clinical coherence, clarity, conciseness. Scoring was automated using an LLM pipeline with physician validation on a 20% sample. Results Consolidated documentation achieved significantly higher mean total assessment composite score compared to ambient-only (94.8 vs. 80.1 on a scale of 0-100; difference 14.6 points; 95% CI 13.4-15.8; P<0.001). The largest improvements was observed in the completeness domain (difference 42.5 points; P<0.001). DM and HTN both showed similar performance of consolidated documentation vs. ambient only. Conclusions Augmenting ambient AI documentation with historical EHR data significantly improves documentation completeness for chronic disease management in primary care. These preliminary findings challenge the prevailing audio-first implementation paradigm and suggest that bidirectional EHR integration may be essential for comprehensive AI-assisted documentation, particularly for conditions requiring synthesis of longitudinal clinical data.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.030
metaresearch head score (Gemma)0.092
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: Observational
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.030
Threshold uncertainty score0.157

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0300.092
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0010.001
Bibliometrics0.0010.001
Science and technology studies0.0010.001
Scholarly communication0.0020.001
Open science0.0010.002
Research integrity0.0010.001
Insufficient payload (model declined to judge)0.0020.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.282
GPT teacher head0.511
Teacher spread0.229 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designObservational
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2025
Admission routes1
Has abstractyes

Explore more

Same venuemedRxiv→Same topicElectronic Health Records Systems→French-language works237,207→