Evaluation of the quality and usability of responses generated by ChatGPT artificial intelligence in response to common patient questions regarding self-care of ischemic heart patients

Coronary artery disease is increasing in Iran with complications such as psychological, disability, loss of work capacity and imposing heavy costs. This disease is one of the most common heart diseases in all countries; in Iran, ۳۳ to ۴۰ percent of deaths are due to cardiovascular diseases, ۷۹.۵ percent of which are related to heart attacks. One of the determining and effective factors in maintaining health and preventing exacerbation of signs and symptoms of the disease, especially in chronic diseases, is performing self-care behaviors. Therefore, this study aimed to determine the answers to common questions of ischemic heart patients in the field of self-care. Methods: This descriptive survey study was conducted in February ۲۰۲۴ through an interview with ChatGPT version ۴ to answer ۲۰ questions of the Miller Self-Care Questionnaire, which has five dimensions of dietary care, smoking cessation, physical care, recommended medications, and stress management. To examine the data quality indicators in this article, we used the ECHO and DUAL criteria, which collected a set of measurable criteria for metadata quality. To better understand the criteria, we examined several dimensions of comprehensiveness, accuracy, information richness, and inherent accuracy. The questions were asked in full to the artificial intelligence, and the results of this study were analyzed by considering a negative score for each incorrect answer. Results: ChatGPT version ۴ provided answers. After matching the responses generated regarding the self-care of ischemic heart patients with the Echeva and Duval tool, it was shown that out of a total of ۲۰ questions, ۱۶ were answered correctly and four were answered incompletely. The average percentage of confidence in answering the questions of the Miller questionnaire was ۸۰%. Conclusion: The findings of this study show that the artificial intelligence Chat GPT effectively answers the self-care questions of patients. The results of the study showed that consistent self-care at home can lead to changes in the self-care behaviors of patients with myocardial infarction. it is suggested that this model be used as a community-based approach in the health system to improve the self-care behaviors of patients with myocardial infarction.