JMIR Publications

ABSTRACT

Background:

To optimize the implementation of AI in healthcare, there is a strong need to better understand AI acceptance and its antecedents among (prospective) healthcare professionals. This requires adequate measurement instruments and scales tailored specifically to AI applications in healthcare. To the best of our knowledge, no such overview currently exists. To address this gap, this scoping review used the Ability, Motivation, and Opportunity (AMO) framework to provide researchers with a comprehensive overview of available measurement instruments and scales to assess predictors of AI acceptance in healthcare settings.

Objective:

This review aimed to synthesize quantitative instruments used to measure (prospective) healthcare professionals’ ability, motivation, and/or opportunity related to AI tools in the workplace. Second, we determined the extent to which these measurement instruments have been psychometrically validated and highlighted gaps for future research.

Methods:

We conducted a scoping review in accordance with the JBI Manual for Scoping Reviews and PRISMA-ScR guidelines, and preregistered the review protocol on OSF at https://osf.io/er2dg. First, we systematically searched PubMed, Web of Science, APA PsycINFO, APA PsycArticles, MEDLINE, and Scopus for English-language studies published from January 1, 2020, to June 2, 2025. Second, retrieved references were imported into EndNote v 21, a reference management platform that streamlines review management. Third, eligibility criteria were validated by two independent reviewers, after which the first reviewer performed title and abstract screening, followed by full-text screening. Fourth and last, the first reviewer, in collaboration with the other reviewers, extracted data using a predefined data extraction template and summarized the collected data using a narrative synthesis approach.

Results:

A total of 75 studies, comprising 49 measurement instruments and scales, were included in this review. Analysis revealed three distinct categories: validated instruments, scales based on general technology acceptance frameworks, and scales outside these frameworks (which are occasionally treated as extensions of existing scales). Furthermore, while domains like literacy and attitudes are represented by a multitude of standardized instruments, constructs such as trust and risk perception are operationalized inconsistently across studies. Regarding psychometric quality, most instruments and scales (n = 36) met only the baseline criteria of one reliability and one validity estimate.

Conclusions:

The findings of this scoping review demonstrate the need for a convergence of questionnaires to operationalize predictors of AI acceptance. The tension between standardized instruments and customized scales highlights a fundamental challenge in the field: while standardized tools offer theoretical validation, no single "gold standard" exists, forcing researchers to balance methodological rigor against context-specific needs. To this end, we recommend relying on an established set of scales for the different predictors of AI acceptance to ensure comparability while also allowing the flexible adaptation of items to fit the specific AI tool and healthcare context.