Talking technology: exploring chatbots as a tool for cataract patient education

Area of Science:

• Ophthalmology
• Health Informatics
• Artificial Intelligence

Background:

• Cataract education is vital for improving patient outcomes and treatment adherence.
• Barriers to effective cataract education include health literacy, language, culture, personal preferences, and resources.
• AI chatbots can enhance patient education through personalized, interactive, and accessible information delivery.

Purpose of the Study:

• To compare the efficacy of AI chatbots versus traditional resources (AAO website) for cataract patient education.
• To evaluate information accuracy, understandability, actionability, and readability of chatbots and the AAO website.

Main Methods:

• A descriptive comparative design was employed.
• Quantitative data from frequently asked questions about cataracts were analyzed from ChatGPT, Bard, Bing AI, and the AAO website.
• SOLO taxonomy, PEMAT, and Flesch-Kincaid readability scores were used for data analysis.

Main Results:

• Chatbots demonstrated higher accuracy in answering cataract-related questions compared to the AAO website (p < 0.001).
• The AAO website outperformed chatbots in understandability and actionability (p < 0.001).
• Bard offered the highest readability ease, followed by the AAO website, Bing AI, and ChatGPT (p < 0.001).

Conclusions:

• AI chatbots show potential for delivering more accurate and detailed cataract information.
• The AAO website provides more understandable and actionable information for cataract patients.
• Considering patient preferences is essential to avoid generalized or biased information and ensure reliability in cataract education.