Impact of AI medical scribes on physician productivity and satisfaction in medical oncology.
Bibliographic record
Abstract
11167 Background: AI Scribes are a leading example of AI implementation in clinical settings, with Oncology practices demonstrating exponential uptake since their introduction. Despite their ever-increasing usage, there are limited studies which directly interrogate the impact of AI Scribes on physician productivity metrics, and few which assess qualitative interpretations of the technology. Methods: This single-center, multi-site study enrolled 27 Medical Oncologists and 3 Primary Care Physicians randomly assigned in a 1:2 ratio to exposure to the Knowtex AI scribe in the initial phase (Phase 1) or control phase (Phase 2). Billing data was collected for 6 months prior to Phase 1 onboarding with Knowtex and for 16 weeks afterward—all within the 2024 fiscal year. During the same period, Phase 2 physicians billing data served as a non-exposed comparison group. Physicians completed opt-in surveys at Week 0 and Week 8 post-exposure assessing confidence and motivation to use the AI Scribe, documentation burden, documentation quality, and experience with the electronic medical record (EMR). Results: All providers adopted the Knowtex AI scribe during their study phase. 4 Phase 2 physicians were excluded from data analysis due to incomplete 2024 fiscal year data. Phase 1 physicians exhibited an increase in mean units (t(10) = 4.44, p < 0.01, d = 1.34, CI [0.90, 2.72]) and mean total billings per working day (t(10) = 4.30, p < 0.01, d = 1.28, CI [$377.55, $1206.75]), a pattern not observed in Phase 2 during the same period. There was no change in the number of diagnostic codes per unit amongst Phase 1 physicians. No learned effect emerged over time in Phase 1 billing metrics or diagnostic coding. Survey findings revealed a strong positive association between Week 0 self-assessed Knowtex understanding and increased units (r(13) = .579, p = 0.024). Physicians reported increased satisfaction with documentation workflow, a reduction in-clinic hours spent on documentation, and increased time spent with patients. Physicians' net impression of EMR challenges markedly decreased following the implementation of the AI scribe (U = 274.5, z=4.054, p < 0.0001). Conclusions: Adoption of an AI Scribe in oncology may enhance certain billing metrics and positively shift physician perceptions of EMR interactions, without affecting the quality of documentation. These findings highlight potential benefits of AI Scribes in improving physician productivity and satisfaction. As AI Scribes trend towards delivering multimodal clinical support tools, future research may focus on the adjunctive effects of AI scribes on procedural efficiencies, such as consistency in billing codes.
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 imitationNot 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.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.005 | 0.025 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.005 | 0.000 |
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.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
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".