LLM tool reduced patient anxiety before prostate cancer surgery
A randomized trial showed that AI-generated answers to patient questions improved psychological outcomes and saved physician time.
A clinical trial published in npj Digital Medicine found that using large language models (LLM) for preoperative communication helps prostate cancer patients feel more at ease. Researchers tested whether AI-generated responses could replace standard information sessions for those preparing for a radical prostatectomy.
In this study, 268 patients were divided into two groups. One group received standard care, while the other group interacted with an LLM that provided personalized, locally generated answers to their concerns before meeting with their doctors.
AI-assisted preoperative communication significantly reduces patient anxiety and improves psychological and cognitive outcomes while decreasing physician workload.
Patients using the AI assistant reported lower anxiety levels and higher satisfaction compared to the control group. The study tracked 3,507 total questions, suggesting that patients had a wide range of specific concerns that the AI helped address effectively.
Doctors also benefited from the system. The time spent on preoperative communication dropped from about 20 minutes to roughly 11 minutes per patient. Additionally, physicians reported lower workloads, as indicated by their scores on the NASA-TLX task load index.
Could this type of automated assistant eventually become a standard part of every hospital visit?
It is important to note that this was a phase II trial focused specifically on prostate cancer surgery. The results might not apply to other medical fields or types of surgeries. Furthermore, the AI acted as an assistant to facilitate communication rather than replacing the clinical judgment of the surgical team.
Source: npj Digital Medicine