ArticlePLOS digital health2026
Implementation outcomes of AI documentation support in routine clinical practice: A non-randomized controlled trial.
Article in PLOS digital health, 2026. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Not yet cited in PubMed.
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.
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.
Who cites it
0 citing papers in PubMed.
No citing paper in PubMed yet.
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
5 authors.
Funding
No grant is acknowledged in the PubMed record.
Abstract
While large language models (LLMs) show promise for reducing clinical documentation burden, existing studies rely on simulated environments or require supplemental audio inputs. This study evaluates an AI documentation copilot integrated into real-world electronic health record (EHR) workflows, measuring its impact on physician efficiency and documentation review burden. We conducted a non-randomized controlled trial with 27 ophthalmology clinicians over 12 months (January 2024-February 2025). Participants self-selected into intervention (n = 11) or control (n = 17) groups after a 4-month baseline. The AI system generated inline suggestions for discharge letter conclusions using only the patient's EHR record as input. The system was based on a Quantized Low-Rank Adaptation (QLoRA) tuned LLM trained on 80,000 institutional records. Primary outcomes included words-per-minute (WPM) rates and the count of required document revisions after checks by the supervising senior physician across 15,615 documents. Intervention group clinicians achieved greater WPM improvements than controls (Δ+7.2 vs. Δ+2.6), while revision frequency decreased within the intervention group (from 1.36 ± 0.14 to 1.26 ± 0.08 per document), with no significant change observed in the control group. Linear mixed-effects modeling confirmed a significant group × exposure interaction for documentation efficiency (p = 0.045), whereas no statistically significant interaction was observed for revision count. This real-world evaluation suggests that AI documentation assistance is associated with improved documentation efficiency under routine clinical conditions, while maintaining documentation standards under established supervision processes. These findings support the cooperative "copilot" design as a viable approach to enhance documentation efficiency without compromising patient safety protocols and indicate its potential to help reduce documentation burden in clinical practice.
Identifiers
What OpenQuestion holds
Registered trials
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.