Comparing Perceptions of ChatGPT Use in Health Attitude Contexts Among Users and Nonusers: Cross-Sectional Study

Clinical Snapshot

Key Findings

• P Value: P>.05 for sociodemographic differences between groups

• Effect Size: Not reported

• Primary Outcome: 76.1% of respondents used ChatGPT, with users finding it useful but expressing privacy and reliability concerns

• Nnt Or Sensitivity: 54.7% of users concerned about privacy, 44.9% about reliability, 63.9% of nonusers cited security risks as barrier

• Confidence Interval: Not reported

Clinical Application

Limited by privacy, security, and reliability concerns identified in study Relevant for TGA considerations of AI medical devices and RACGP digital health guidelines, though cultural differences may limit direct applicability Healthcare consumers and professionals considering AI chatbot integration

Abstract

BACKGROUND: In light of the growing use of artificial intelligence (AI) in health care, individuals' access to and use of health information are transforming. ChatGPT, an AI chatbot, provides immediate responses to health queries, with the potential to influence health-related attitudes, thereby raising concerns related to privacy, reliability, and security. OBJECTIVE: This study aimed to investigate the perceived usefulness, risks, anxiety, and social influence of ChatGPT on health attitudes among users and nonusers in Saudi Arabia. METHODS: A cross-sectional study was conducted using an online survey based on a validated tool. In total, 337 participants aged 18 years and older responded to questions assessing their perceptions of ChatGPT on health-related attitudes. RESULTS: Data showed that 76.1% (194/255) of the respondents used ChatGPT, with the majority being younger and more highly educated. The main uses for health-related purposes were health education (43/194, 22.2%) and physical activity guidance (31/194, 16%). The analysis showed that users considered ChatGPT useful for health-related decisions, with 45.9% (89/194) finding it easy to learn and use, but concerns about privacy (106/194, 54.7%) and reliability (87/194, 44.9%) remained. Among nonusers, security risks (39/61, 63.9%) were the major barrier to using AI-based tools for health purposes, and 68.9% (42/61) found such tools attractive and engaging. There were no statistically significant differences between users and nonusers across all examined sociodemographic characteristics (P>.05). CONCLUSIONS: The study established the potential of ChatGPT in improving health decision-making and revealed cultural, privacy, and trust issues that may affect its implementation. These findings underscore the importance of strengthening the security of AI-based applications to enhance public acceptability of related health policies and to support the safe integration of tools such as ChatGPT into the health care system.

References

1. 1.Al Suwaidan, H., Althumairi, A., Al-Rayes, S. A., & Alkhurayji, K. (2026). Comparing perceptions of ChatGPT use in health attitude contexts among users and nonusers: Cross-sectional study. JMIR Formative Research. https://doi.org/10.2196/79276