Large language models and artificial intelligence for generating, simplifying, and enhancing outpatient clinic letters: A systematic review

@article{sheldon2026large,
  title = {{Large language models and artificial intelligence for generating, simplifying, and enhancing outpatient clinic letters: A systematic review}},
  author = {Rocco Sheldon and Morya Wadodkar and Andrew Gan and Rachael Yip and Yimeng Zhang and Jyoti Baharani},
  journal = {PLOS Digital Health},
  year = {2026},
  month = sep,
  doi = {10.1371/journal.pdig.0001745},
  url = {https://journals.plos.org/digitalhealth/article?id=10.1371/journal.pdig.0001745},
  note = {CC BY 4.0},
}

TY  - JOUR
TI  - Large language models and artificial intelligence for generating, simplifying, and enhancing outpatient clinic letters: A systematic review
AU  - Rocco Sheldon
AU  - Morya Wadodkar
AU  - Andrew Gan
AU  - Rachael Yip
AU  - Yimeng Zhang
AU  - Jyoti Baharani
PY  - 2026
DA  - 2026/09/22
JO  - PLOS Digital Health
DO  - 10.1371/journal.pdig.0001745
UR  - https://journals.plos.org/digitalhealth/article?id=10.1371/journal.pdig.0001745
AB  - Outpatient clinic letters are a cornerstone of clinical communication but frequently exceed recommended reading levels, limiting patient comprehension and engagement. Large Language Models (LLMs) and Artificial Intelligence (AI) have been proposed as scalable tools for generating and simplifying these documents, but the evidence regarding their clinical accuracy, safety, and effectiveness remains fragmented. We conducted a systematic review of studies evaluating AI or LLMs for generating, simplifying, or enhancing outpatient clinic letters. Five databases (PubMed, EMBASE, Web of Science, CENTRAL, and CINAHL) were searched from inception to 1 November 2025. Two independent reviewers screened studies, extracted data, assessed quality using the Mixed Methods Appraisal Tool (MMAT), and certainty using the GRADE appraisal tool. Seven studies were included, comprising two studies using real-world clinic data and five using synthetic or hypothetical scenarios. Findings regarding readability were mixed; while some AI models improved readability scores compared to human-authored letters, most AI-generated content still failed to meet the recommended US Grade 6 reading level. The two studies that measured patient-reported outcomes reported high satisfaction and comprehension with AI-simplified letters, but these studies relied on subjective measures of understanding or lacked a human-authored control group. Clinical accuracy varied substantially, with information fidelity ranging from 10% to 100%. Risks of ‘hallucinations’, inappropriate medical advice, and tactless patient descriptions were documented. One study reported a ten-fold reduction in drafting time compared to human dictation. Overall certainty of evidence was rated very low across all primary outcomes due to methodological limitations, small sample sizes, and reliance on hypothetical scenarios. The current evidence supporting AI and LLMs use for outpatient clinic letters is preliminary and limited, providing insufficient evidence to support routine clinical implementation beyond supervised experimental and quality-improvement settings. Concerns regarding accuracy, safety, and generalisability remain. Further real-world patient-centred research is required before widespread adoption. PROSPERO registration: CRD420251181303. Author summary: Outpatient clinic letters are a central part of how doctors communicate with patients, yet they are routinely written at a reading level far above what most people can understand. Despite longstanding guidance in the United Kingdom recommending that letters be written directly to patients, this remains inconsistently practised: partly due to time pressures and concerns about losing clinical detail. Artificial intelligence (AI) tools, particularly large language models such as ChatGPT, have been proposed as a scalable solution to this problem; however, the evidence underpinning their use has not previously been synthesised. We conducted a systematic review of seven studies evaluating AI for writing, simplifying, or enhancing clinic letters. We found that preservation of clinical information was highly variable (10%-100%), and that while two studies reported positive patient satisfaction and understanding, these studies had significant limitations in their design. No AI tool consistently achieved recommended reading levels using formal readability measures. Risks of inaccurate or inappropriate hallucinations were documented across studies. Overall, the quality of the evidence was very low. We conclude that whilst promising, the current evidence does not support the routine use of AI for clinic letters beyond research settings, and that larger, real-world, patient-centred studies are urgently needed.
SN  - 2767-3170
ER  -