Previously submitted to: Journal of Medical Internet Research (no longer under consideration since Apr 30, 2023)
Date Submitted: Feb 19, 2023
Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Analyzing Doctors’ Acceptance of Artificial Intelligence-Enabled Clinical Decision Support Systems in Well-Known Tertiary Hospitals in China Based on an Extended Unified Theory of Acceptance and Use of Technology Model
ABSTRACT
Background:
Clinical decision support systems (CDSSs) have great potential to improve healthcare performance and efficiency, increase safety, and reduce costs. Artificial intelligence-enabled CDSSs (AI-CDSSs) enhance decision-making ability by introducing new AI technologies such as deep neural networks and knowledge graphs. However, factors influencing user acceptance of AI-CDSSs by doctors, especially those from well-known tertiary hospitals in China, are unclear.
Objective:
This study aimed to analyze the factors influencing doctors’ acceptance of AI-CDSSs in well-known tertiary hospitals in China.
Methods:
The unified theory of acceptance and use of technology (UTAUT) model was extended to propose a hypothesized model based on the perspective of doctors’ acceptance of AI-CDSSs. We conducted a web-based survey of doctors from four well-known tertiary hospitals in Liaoning, Zhejiang, Sichuan, and Guangdong provinces. We developed and evaluated a 25-item measurement scale. We used partial least squares structural equation modeling to analyze the data.
Results:
A total of 187 doctors completed the survey, and 137 questionnaires were considered effective based on a quality control strategy. Psychometric evaluation of the scale gave a Cronbach α value of 0.932 and corrected item to total correlation values ranging from 0.467 to 0.744. The average variance of extracted values ranged from 0.628 to 0.782, composite reliability values ranged from 0.871 to 0.931, and heterotrait-monotrait ratio values ranged from 0.254 to 0.845. In model testing, the variance inflation factor ranged from 1 to 2.21, the standardized root mean square residual value was 0.055, the squared Euclidean distance (0.995) and the geodesic distance (0.566) were less than the corresponding upper boundary of the 95% confidence interval (0.997 and 0.647). The final model had an explanatory power of 73.1% for user acceptance.
Conclusions:
Doctors’ user acceptance of AI-CDSSs is strongly influenced by effort expectancy, is moderately influenced by trust in (medical) AI and is weakly influenced by social factors. Performance expectancy does not directly influence doctors' acceptance. However, it could indirectly affect doctors' acceptance by trust in (medical) AI as an intermediary. Moreover, trust in (medical) AI as an important new factor in our extended UTAUT model is moderately influenced by social factors and is weakly influenced by effort expectancy and performance expectancy. Finally, effort expectancy is influenced by personal innovativeness, while performance expectancy is influenced by task-technology fit.
Citation
The author of this paper has made a PDF available, but requires the user to login, or create an account.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.