Evaluation of the perception of information from ChatGPT in myopia education: Perspectives of students and professionals

Abstract

Purpose: The purpose of this study was to compare the perception and understanding of the information provided by ChatGPT regarding myopia among optometry students, optometrists undertaking a Master degree and practicing optometrists.

Methods: This was a cross-sectional descriptive study using a structured questionnaire distributed via Wooclap to 225 participants (125 optometry students, 21 Masters students and 79 practicing optometrists). All participants evaluated the responses generated by ChatGPT Version 4.0 using a five-point scale: very poor, poor, acceptable, good and very good. Data were analysed using SPSS, applying descriptive analysis and chi-squared tests to assess the significance of differences observed between the groups.

Results: Practicing optometrists, with an average age of 39.7 ± 11.0 and 15.2 ± 7.5 years of professional experience, rated the information provided by ChatGPT on myopia complications more positively than the optometry and Masters students, highlighting the importance of clinical experience (p < 0.001). Alternatively, this may indicate a difference in awareness, knowledge and training, with younger students potentially being more speculative in their evaluations. The perceptions about myopia prevention among the groups were similar (p = 0.28). At the same time, the perceptions of the effectiveness of treatments such as contact lenses and pharmaceutical agents showed differences, with optometrists rating these treatments more positively (p < 0.001 and p = 0.004, respectively).

Conclusions: This study showed differences in the perception and interpretation of the information provided by ChatGPT about myopia depending on the level of education. While AI plays an important role in education, practicing optometrists trust the information extracted by ChatGPT more than optometry students.