JMIR Publications

ABSTRACT

Background:

In the current digital era, eHealth literacy plays an indispensable role in healthcare and self-management among older adults with noncommunicable diseases (NCDs). Measuring eHealth literacy appropriately and accurately ensures the successful implementation and evaluation of pertinent research and interventions. However, existing eHealth literacy measures focus mainly on individuals’ abilities of accessing and comprehending eHealth information (Web1.0), whereas the capabilities for online interaction (Web2.0) and utilizing eHealth information (Web3.0) have not been adequately evaluated.

Objective:

The purpose of this study was to examine the reliability, validity, and measurement invariance of eHealth Literacy Scale-Web3.0 (eHLS-Web3.0) among older adults with NCDs.

Methods:

A total of 642 Chinese older adults with NCDs (65.78 ± 3.91 years; 55.8% females) were recruited in the baseline assessment, of which 134 (65.63 ± 3.99 years; 58.2% females) completed the 1-month follow-up assessment. Baseline measures included the Chinese version of the 24-item three-dimensional eHLS-Web3.0, the Chinese version of the 8-item unidimensional eHealth Literacy Scale (eHEALS), and demographic information. Follow-up measures included the 24-item eHLS-Web 3.0 and accelerometer-measured physical activity (PA) and sedentary behavior. A series of statistical analyses, e.g., Cronbach alpha, composite reliability coefficient (CR), confirmatory factor analysis (CFA), and multigroup CFA), were performed to examine the internal-consistency and test-retest reliabilities, as well as construct, concurrent, convergent, discriminant, and predictive validities, and measurement invariance of eHLS-Web3.0 across gender, education level and residence.

Results:

Cronbach alpha and CR were within acceptable ranges of 0.89-0.94 and 0.90-0.97, respectively, indicating adequate internal consistency of the eHLS-Web3.0 and its subscales. The eHLS-Web3.0 also demonstrated cross-time stability, with baseline and follow-up measures showing a significant correlation of 0.81-0.91. The construct validity of the three-dimensional structure model of eHLS-Web3.0 was supported by confirmatory factor analyses. The eHLS-Web3.0 exhibited convergent validity with an AVE value of 0.58 and CR value of 0.97. Discriminant validity was supported by CFA results for a proposed four-factor model integrating the three eHLS-Web3.0 subscales and eHEALS. The predictive validity of eHLS-Web3.0 for health behaviors was supported by significant associations of eHLS-Web3.0 with light PA (β = 0.36, P = 0.004), moderate-to-vigorous PA (β = 0.49, P < 0.001) and sedentary behavior (β = -0.26, P = 0.002). Finally, measurement invariance of eHLS-Web3.0 across gender, education level, and residence was supported by the establishment of configural, metric, strong and strict invariances.

Conclusions:

This study provides timely empirical evidence on the reliability, validity, and measurement invariance of eHLS-Web3.0, suggesting that the 24-item three-dimensional eHLS-Web3.0 is a appropriate and valid tool for measuring eHealth literacy among older adults with NCDs within the Web3.0 sphere. Clinical Trial: Not appliable.