Related Experiment Video
Updated: May 3, 2026
E-Patient Counseling Trial E-PACO: Computer Based Education versus Nurse Counseling for Patients to Prepare for Colonoscopy
Published on: August 1, 2019
Enhancing patient education in cataract surgery using a conversational artificial intelligence chatbot: pilot
Paul Cardon de Lichtbuer1, William Declerck, Quinten Rosseel
1From the Faculty of Medicine and Pharmacy, Vrije Universiteit Brussel (VUB), Brussels, Belgium (Cardon de Lichtbuer, Oellerich, Ní Dhubhghaill); Department of Ophthalmology, Universitair Ziekenhuis Brussel (UZB), Jette, Belgium (Cardon de Lichtbuer, Oellerich, Ní Dhubhghaill); AI Developer at Venture Queue, Belgium (Declerck, Rosseel); Core Facility - Support for Quantitative and Qualitative Research (SQUARE), Vrije Universiteit Brussel (VUB), Brussels, Belgium (Savieri); Biostatistics and Medical Informatics Research Group, Vrije Universiteit Brussel (VUB), Brussels, Belgium (Savieri).
A new AI chatbot for cataract surgery education was highly usable but did not improve patient knowledge, anxiety, or satisfaction compared to brochures. Engagement was limited, especially in older adults, suggesting a hybrid approach is needed.
Area of Science:
- Ophthalmology
- Medical Informatics
- Patient Education
Background:
- Cataract surgery education traditionally relies on brochures.
- Conversational AI chatbots offer a potential alternative for patient information delivery.
- Evaluating AI effectiveness in healthcare requires rigorous comparison with standard methods.
Purpose of the Study:
- To assess the effectiveness and usability of a clinician-validated AI chatbot for cataract surgery education.
- To compare the chatbot's impact on patient knowledge, anxiety, and satisfaction against standard brochures.
- To identify barriers to chatbot engagement in a real-world clinical setting.
Main Methods:
- A prospective, single-center randomized controlled trial was conducted.
- Adults undergoing cataract surgery were randomized to receive brochures alone or brochures plus chatbot access.
- Outcomes included knowledge gain, anxiety change, satisfaction, chatbot usability (System Usability Scale), and engagement.
Main Results:
- No significant differences in knowledge gain, anxiety, or satisfaction were found between the chatbot and brochure groups.
- Knowledge increased in both groups post-information (p<0.001).
- 52% of patients in the chatbot group did not engage; users tended to be younger. The chatbot demonstrated high usability (SUS score 83.1).
Conclusions:
- A clinician-validated AI chatbot for cataract surgery education showed high usability but did not outperform standard brochures in improving patient outcomes.
- Engagement barriers, particularly for older adults, and limitations of validated-only content suggest a need for hybrid digital education models.
- Future strategies should balance AI safety with flexibility to enhance patient education effectively.
