Separating Technology and Trust: A Survey Analysis of Patients’ Attitudes toward AI-Assisted Healthcare Decision-Making

Abstract

The integration of artificial intelligence (AI) and Internet of Things (IoT) technologies is fundamentally transforming healthcare delivery from hospital-centric to people-centered models. While AI demonstrates significant potential in improving diagnostic accuracy and enabling predictive disease management, important challenges persist regarding ethical considerations, social trust, data privacy, and equitable access across diverse populations. This research significance despite technological advances, comprehensive frameworks addressing AI transparency, social trust, and ethical implications in healthcare remain underdeveloped. This study addresses critical gaps by examining consumer perspectives on AI-assisted clinical decision support (AI-CDS) systems, focusing on perceived benefits, risks, and factors influencing acceptance across demographic segments. These statistical methods and measures were implemented using IBM SPSS 27.0 software for analyzing the survey data on attitudes toward AI-CDS systems. Data were collected through structured online surveys from 442 participants across diverse demographic backgrounds in the United States. The questionnaire assessed nine key dimensions including AI-CDS knowledge, trust, and ease of use, bias concerns, and willingness to follow recommendations using validated five-point Likert scales. Statistical analysis employed descriptive statistics, chi-square tests, and one-sample t-tests. The results respondents demonstrated moderate, neutral attitudes toward AI-CDS across all dimensions (means 3.02-3.18). Education level significantly influenced understanding and comfort with AI systems, while income affected approval needs and bias concerns. Healthcare provider status and clinical documentation experience emerged as crucial factors shaping trust and acceptance. Successful AI-CDS implementation requires addressing trust deficits through experiential learning, robust regulatory frameworks, and maintaining human-AI collaboration in healthcare decision-making.