Evaluation of chatgpt's answers to questions in the field of women's health and child health

Artificial intelligence is effective in many areas of our lives and performs various tasks. ChatGPT, an AI-powered natural language processing model developed by OpenAI, is widely used in the healthcare sector. One of the main uses of ChatGPT in healthcare is to help the community search for medical information. However, ChatGPT has some limitations and the medical information it provides should be used with caution. Inaccurate or incomplete medical information can have a negative impact on public health. Therefore, the aim of this study was to evaluate the reliability, quality and comprehensibility of ChatGPT responses to questions about women's and children's health. In this cross-sectional study, 100 women's health and 100 children's health questions were posed to ChatGPT between 10.07.2024 and 10.09.2024. The quality of ChatGPT responses was assessed using the CLEAR, EQIP and mDISCERN scales. The readability of the responses was analysed using Ateşman and Flesch Kincaid readability formulas. The results were analysed with SPSS and descriptive statistics and Student T-test were used. According to the CLEAR scale, 83 % of the answers to women's health questions were of very good content quality and 79 % of the answers to children's health questions were of very good content quality. According to the EQIP scale, 68 % of the responses for women's health and 59 % of the responses for children's health were minor problems. According to the Ateşman readability score, 64 % of the answers related to women's health were considered to be of medium difficulty and 66 % of the answers related to children's health were considered to be of medium difficulty. According to the Flesch-Kincaid readability score, 68 % of the answers related to women's health were found to be difficult and 42 % of the answers related to children's health were found to be moderately difficult. ChatGPT responses on women's and children's health were generally of average or better quality, but many responses were difficult to read. ChatGPT is a valuable tool for providing medical information, but should be used with caution due to its limitations and difficult readability.