JMIR Publications

ABSTRACT

Background:

As artificial intelligence (AI) becomes increasingly embedded in healthcare practice and organisational systems, healthcare workers require a level of AI literacy that extends beyond basic awareness or technical familiarity. However, existing measures of AI literacy have largely been developed for general or student populations, limiting their relevance to healthcare settings.

Objective:

To address this gap, this study reports the development and validation of the Scale for Artificial Intelligence Literacy in Healthcare Workers (SAIL-HCW).

Methods:

Scale development followed a three-phase process. First, a review of the literature informed the identification of seven conceptual domains and the generation of an initial item pool. Second, content and face validity were established through expert review and pilot testing. Third, SAIL-HCW psychometric properties were examined using data from 425 healthcare workers from a major healthcare provider in Singapore. The dataset was randomly split to allow exploratory factor analysis for factor extraction and confirmatory factor analysis for tests of dimensionality.

Results:

Analyses supported a 14-item bifactor model comprising a general AI literacy factor and seven domain-specific dimensions: AI Concept, Data Fluency, AI Evaluation, AI in Practice, Ethics and Regulation, AI in Systems, and Continuous Learning. The scale demonstrated good internal consistency, satisfactory item–total correlations, and evidence of known-group validity.

Conclusions:

This suggests that SAIL-HCW is a concise and contextually grounded measure of AI literacy for healthcare workers. It is suitable for assessing baseline literacy, identifying training needs, and supporting research on workforce readiness as AI technologies continue to shape healthcare delivery. Further validation in diverse healthcare settings is recommended. Clinical Trial: Not applicable