Medical reports can be difficult to understand. They often use unfamiliar words, and patients may have little time to ask questions during an appointment. The AI-INFOCARE study tested whether a plain-language summary sent before a visit could help patients and doctors communicate more clearly.
The study included 300 adults attending an outpatient neurosurgery clinic. Half received usual care, while the other half received a plain-language summary of their medical information before the appointment. AI helped create the summaries, and people checked them before sharing them with patients. The study took place at one clinic, so the results may differ in other health services (Benker-Hasani et al., 2026).
Patients who received a summary gave slightly higher ratings to their conversations with doctors. They especially reported better shared decision-making, that is, feeling more involved in choices about their care. The study found no clear change in how much empathy patients felt from their doctors.
Patients also felt slightly more supported in making their own healthcare decisions. A simple summary may help patients prepare questions and follow the discussion. But the study did not show whether patients understood their health information better over time or made different treatment choices.
Appointments were shorter for patients who received summaries. The middle appointment length was 25 minutes, compared with 33 minutes for patients receiving usual care. This may be useful in busy clinics. However, a shorter visit is not always a better visit, and the study did not show whether shorter appointments reduced doctors’ workload or affected later care.
Patients and doctors did not view every part of the experience in the same way. Patients felt slightly more supported in making decisions, while doctors rated their own support for patient choice a little lower. This difference is worth studying. It may show that patients and clinicians experience the same conversation differently.
The findings are promising, but they do not prove that AI summaries are right for every patient or clinic. The study was done at one neurosurgery clinic. The benefits may be different in therapy, primary care, or other medical settings. The study also did not test whether the summaries improved health outcomes.
AI can make mistakes or leave out important details. For this reason, a trained person should check each summary before a patient receives it. Patients and clinicians should know that AI was used. The summary should help start a conversation, not replace the medical record or the clinician’s explanation. The care team remains responsible for checking the information and answering patients’ questions.
For therapists, this study offers a useful idea: people may feel more prepared for an appointment when complex information is explained in everyday language beforehand. But this trial studied neurosurgery patients, not therapy clients. More research is needed to learn whether similar summaries help in mental health care and how to protect privacy and avoid errors.
Overall, AI-generated summaries may help patients take part in conversations and may make some appointments shorter. These early results are encouraging, but they are not enough to recommend routine use everywhere. Future studies should include different clinics and patient groups, and should measure understanding, decision-making, workload, and safety.
Reference
Benker-Hasani, E., Richter, S., Buszello, C., Willkommen, S., Prem, M. G., Hijazi, M., Martin, K. D., Juratli, T., Eyüpoglu, I. Y., & Polanski, W. H. (2026). Effect of previsit AI-generated document summaries on consultation time and patient–physician communication: The AI-INFOCARE randomized controlled trial. NEJM AI, 3(11). https://doi.org/10.1056/AIoa2600136
