Perception, concerns, and practice of ChatGPT among Egyptian pharmacists: a cross-sectional study in Egypt

Abstract Background The emergence of large language models (LLMs) like ChatGPT attracted significant attention for their potential to revolutionize pharmacy practice. While artificial intelligence (AI) offers promising benefits, its integration also presents unique challenges. Objectives This cross-sectional study aimed to explore the current Egyptian pharmacists’ perceptions, practices, and concerns regarding ChatGPT in pharmacy practice. Methods The study questionnaire was shared with pharmacists during March and April 2024. We included pharmacists licensed by the Egyptian Ministry of Health and Population. We adapted a convenient sampling technique by sending the research questionnaire via emails, student networks, social media (Facebook and WhatsApp), and student organizations. Any pharmacist interested in participating followed a link to review the study description and was asked to provide electronic consent before continuing with the study. Data were analyzed using SPSS software, employing Chi-square tests for categorical variables and Spearman’s correlation for continuous variables. Statistical significance was set at p < 0.05. Results The study sample size included 428 pharmacists from the main economic regions of Egypt. The results revealed a strong recognition (73.6%) among participants of ChatGPT’s anticipated benefits within pharmacy practice. Around two-thirds of the participants (65.9%) expressed disagreement or neutrality regarding the application of ChatGPT for analyzing patients’ medical inputs and providing individualized medical advice. Regarding factors affecting perception, we found that the region is the only factor that significantly contributed to the level of perception among pharmacists (P = 0.011) with Greater cairo region showing the highest perception level. We found that 73.6% of participants who have heard about ChatGPT reported high levels of concern. One-third of participants never use ChatGPT in their pharmacy work, and 20% rarely use it. Using Spearman’s correlation test, there was no significant correlation between anticipated advantages, concerns and practice level (P > 0.05). Conclusion This study reveals a generally positive perception of ChatGPT’s potential benefits among Egyptian pharmacists, despite existing concerns regarding accuracy, data privacy, and bias. Notably, no significant associations were found between demographic factors and pharmacists’ perceptions, practices, or concerns. This underscores the need for comprehensive educational initiatives to promote informed and responsible ChatGPT utilization within pharmacy practice. Future research should explore the development and implementation of tailored training programs and guidelines to ensure the safe and effective integration of ChatGPT into pharmacy workflows for optimal patient care.