Artificial intelligence-generated pre-visit summaries improved communication outcomes in neurosurgical consultations

1. Patients who received artificial intelligence (AI)-generated language summaries in non-clinical language before consultations reported better interactions with physicians.

2. Pre-visit summaries resulted in shorter consultations and greater reported autonomy support, but effects on clinical outcomes remain unestablished.

Evidence Rating Level: 1 (Excellent)

Study Rundown: Limited patient health literacy and clinical jargon often impair patients’ understanding of medical findings during physician consultations. Benker-Hasani investigated whether providing AI-generated summaries of patients’ medical documents before consultations can improve communication. Researchers randomized adult patients to receive usual care plus AI-generated summaries or usual care alone. The summaries underwent automated checks before delivery to patients, as well as retrospective physician review. After the consultation, patients and physicians completed questionnaires to assess communication-based outcomes. The study found that patients who received the summaries reported better interaction quality and autonomy, but there were no perceived empathy differences between the two groups. However, physicians rated their own autonomy as lower, and the improvement in communication was below the pre-specified clinically significant difference. Other limitations included the unblinded and the single-center design, which limited internal validity and generalizability. Ultimately, these results support further evaluation of the use of pre-visit summaries to improve patient communication, but more studies are needed to determine whether there are significant improvements in comprehension and clinical outcomes.

Click here to read the study in NEJM AI

Relevant Reading: Impact of Large Language Model–Based AI Tools on Physician-Patient Communication: Systematic Review and Meta-Analysis

In-Depth [randomized controlled trial]: 300 adults were enrolled in this trial, with 150 participants randomized to each group. One group received usual care alone, and the other group received usual care plus a lay language summary. An AI model generated the summaries, which underwent automated quality checks before delivery. The study participants and physicians were not blinded. The primary outcome was the FAPI questionnaire, which measured patient-reported interaction quality on a 1-5 Likert scale. Secondary measures included physician- and patient-rated autonomy scores, as well as staff-recorded consultation duration. Mean FAPI scores were 4.42 in the intervention group versus 4.25 in the control group (difference, 0.17; 95% confidence interval [CI], 0.05-0.29). This difference was below the pre-specified 0.3-0.5 point clinically relevant cutoff. Additionally, shared decision-making scores increased in the intervention group (difference, 0.29; 95% CI, 0.15 to 0.42), while perceived empathy showed no clear difference. Median visit durations were 25 minutes in the intervention group and 33 minutes in the control group. Patient-rated autonomy scores increased by 0.14 points (difference, 0.14; 95% CI, 0.06 to 0.23), whereas physician ratings of autonomy decreased by 0.12 points (difference, −0.12; 95% CI, −0.21 to −0.04). Encounter difficulty differences were statistically insignificant. Study limitations included an unblinded design, reliance on self-report, and no assessment of post-consultation clinical outcome.

Citation: Benker-Hasani E, Richter S, Buszello C, Willkommen S, Prem MG, Hijazi M, et al. Effect of Previsit AI-Generated Document Summaries on Consultation Time and Patient-Physician Communication - The AI-INFOCARE Randomized Controlled Trial. NEJM AI. 2026 Nov;3(11):AIoa2600136. doi: 10.1056/aioa2600136. Epub 2026 Oct 15. PMID: 42932468.

© 2026 2 Minute Medicine, Inc. All rights reserved. No part of this article may be reproduced, distributed or transmitted in any form or by any means without the prior written permission of 2 Minute Medicine, Inc. To license this content, contact 2 Minute Medicine, Inc. This article is for informational and educational purposes only; it is not medical advice and does not replace the judgment of a qualified healthcare professional.
Image: © 2 Minute Medicine