# A randomised trial of AI scribes in oncology clinics measuring errors and time

Source: https://onco.cc/ideas/idea-data-llm-documentation-rct-oncology/  
OnCo record `idea-data-llm-documentation-rct-oncology` (Idea). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

AI tools that write clinic notes are spreading fast in cancer clinics. Test them properly: do they save time, do they make mistakes about drugs and doses, and do patients notice a difference?

## Summary

Ambient documentation tools built on large language models are being adopted across clinics without randomised evidence, and oncology notes carry high-stakes details (regimens, doses, trial eligibility, goals of care). The proposal is a multi-centre randomised trial of AI scribes versus usual documentation in oncology clinics, with primary outcomes of clinically significant documentation errors (blinded audit), clinician time and burnout, and patient-reported communication quality, plus a secondary analysis of structured data completeness (mCODE elements captured).

## Fields

- Kind: Idea
- Last checked: 2026-09-08
- Hypothesis: AI scribes will reduce documentation time and burnout but will introduce a non-trivial rate of clinically significant errors in oncology-specific content unless paired with structured verification, and the trial will quantify both.
- Rationale: Early observational reports show time savings and occasional hallucinated content; the trade-off in oncology, where a wrong dose or regimen in the note propagates, must be measured rather than assumed.
- Proposed test: Randomise 200 oncologists across ten centres for six months; audit 5,000 notes blinded for errors; measure time, burnout and patient experience.
- Maturity: early-clinical
- Actor: research

## Sources

- Bottleneck evidence (AI that is built but not validated or deployed): Wu et al., How medical AI devices are evaluated: limitations and recommendations from an analysis of FDA approvals (Nature Medicine 2021): https://doi.org/10.1038/s41591-021-01312-x

## Connected records

- fronts: [AI & Computation](https://onco.cc/fronts/ai-computation/)
- bottlenecks: [AI that is built but not validated or deployed](https://onco.cc/bottlenecks/b-ai-validation/), [Not enough oncologists, nurses, pathologists, physicists](https://onco.cc/bottlenecks/b-workforce/)
- key papers: [How medical AI devices are evaluated: limitations and recommendations from an analysis of FDA approvals](https://onco.cc/key-papers/paper-wu-nat-med/)

---
JSON: https://onco.cc/api/v1/entities/idea-data-llm-documentation-rct-oncology.json