Development of a Digital Health Literacy Assessment Framework for Older Adults: Delphi-Based Study.

Background: Digital health literacy (DHL) is crucial for improving health outcomes. As a digitally marginalized group, older adults face significant barriers in using digital technologies for health information access and eHealth services, which hinders their ability to benefit from digital dividends. A comprehensive and contextually relevant assessment tool is essential to understand their DHL and to inform effective interventions. Objective: This study aimed to develop a theoretically grounded and content-valid assessment framework for DHL among older adults in China, establishing a foundation for future measurement and research. Methods: A multiphase, theory-informed approach was adopted. The construct domains and an initial item pool were derived from a literature review and qualitative focus group discussions. A 2-round Delphi consultation was conducted to refine and reach consensus on the indicator system. Subsequently, a pilot cross-sectional survey was conducted to validate the reliability and validity of the scale. Results: Two rounds of expert consultation were conducted with 15 experts, and 322 older adult participants were included in the cross-sectional survey. The expert authority coefficients (Cr) were 0.748 and 0.768, respectively, and the Kendall harmony coefficient was statistically significant. The resulting "individual competency-environmental support" dual-pathway framework comprises 6 primary domains (confidence and trust in application, operational skills, critical evaluation skills, application skills, security and privacy, and external support), 13 subdomains, and 27 specific indicators. In the pilot sample (N=332), the overall framework showed excellent internal consistency, and the Cronbach α coefficients ranged from 0.913 to 0.987. Criterion validity was supported by a strong correlation with the established eHealth Literacy Scale tool (r=0.891; P<.001). Known-groups validity was confirmed, as the framework successfully discriminated between older adults with different educational levels (t330=-4.37; P<.001). Conclusions: This study developed a dual-dimensional "individual competency-environmental support" framework to assess DHL among older adults. The framework systematically identifies competency variations and contextual barriers, offering a standardized tool for needs assessment, intervention design, and evaluation of digital inclusion efforts. It also supports the development of targeted, multisectoral strategies to enhance digital health engagement in aging populations. Further validation across diverse settings is recommended to strengthen its generalizability and practical applicability.