Reliability and Validity of the Telephone-Based eHealth Literacy Scale Among Older Adults: Cross-Sectional Survey

Abstract

BackgroundOnly a handful of studies have examined reliability and validity evidence of scores produced by the 8-item eHealth literacy Scale (eHEALS) among older adults. Older adults are generally more comfortable responding to survey items when asked by a real person rather than by completing self-administered paper-and-pencil or online questionnaires. However, no studies have explored the psychometrics of this scale when administered to older adults over the telephone.ObjectiveThe objective of our study was to examine the reliability and internal structure of eHEALS data collected from older adults aged 50 years or older responding to items over the telephone.MethodsRespondents (N=283) completed eHEALS as part of a cross-sectional landline telephone survey. Exploratory structural equation modeling (E-SEM) analyses examined model fit of eHEALS scores with 1-, 2-, and 3-factor structures. Subsequent analyses based on the partial credit model explored the internal structure of eHEALS data.ResultsCompared with 1- and 2-factor models, the 3-factor eHEALS structure showed the best global E-SEM model fit indices (root mean square error of approximation=.07; comparative fit index=1.0; Tucker-Lewis index=1.0). Nonetheless, the 3 factors were highly correlated (r range .36 to .65). Item analyses revealed that eHEALS items 2 through 5 were overfit to a minor degree (mean square infit/outfit values <1.0; t statistics less than –2.0), but the internal structure of Likert scale response options functioned as expected. Overfitting eHEALS items (2-5) displayed a similar degree of information for respondents at similar points on the latent continuum. Test information curves suggested that eHEALS may capture more information about older adults at the higher end of the latent continuum (ie, those with high eHealth literacy) than at the lower end of the continuum (ie, those with low eHealth literacy). Item reliability (value=.92) and item separation (value=11.31) estimates indicated that eHEALS responses were reliable and stable.ConclusionsResults support administering eHEALS over the telephone when surveying older adults regarding their use of the Internet for health information. eHEALS scores best captured 3 factors (or subscales) to measure eHealth literacy in older adults; however, statistically significant correlations between these 3 factors suggest an overarching unidimensional structure with 3 underlying dimensions. As older adults continue to use the Internet more frequently to find and evaluate health information, it will be important to consider modifying the original eHEALS to adequately measure societal shifts in online health information seeking among aging populations.

Electronic Health Literacy Across the Lifespan: Measurement Invariance Study (Preprint)

“…Sudbury-Riley and colleagues [39] also found that a 3-factor model of information awareness, seeking, and appraisal skills among baby boomers and older adults from European and North American nations. Consistent with these studies, a 3-factor model has also been confirmed among baby boomers and older adults [36], including chronic disease patients [34] and adults, with an average age of 53, in a magnetic resonance imaging and computed tomography medical imaging outpatient clinic [40]. Despite evidence for measurement invariance among middle-to-older age adults in multiple nations [39], empirical attention is needed to confirm the 3-factor structure among a general population consisting of younger and older adults.…”

“…The Mdn age value of the current sample was 49, and an inspection of a histogram showed a bimodal age distribution. Consistent with principles of SEM and the average age value in existing exploratory eHEALS measurement studies [ [34][35][36]39], using age 49 as a cut-off value was appropriate to create two age groups: young and old . Frequency and descriptive statistics were computed to describe the sample and eHealth literacy scores.…”

“…There is sufficient reliability and validity evidence of eHEALS delivered via telephone among middle-to-older age adults [36], a population most likely to respond differently to webbased versus telephone administered surveys [50]. Future research could explore the degree of measurement invariance across the 3-dimensional eHEALS model by mode of survey administration.…”

“…The eHEALS is a brief, 8-item rating scale grounded in self-efficacy, or the confidence in one's capability to engage in a behavior to result in a desired outcome [32]. Despite its initial development with high school adolescents, evidence has been established for the validity of eHEALS scores across diverse populations, including college-aged students [33], the general adult population [33], patients with chronic disease [34], and older adults recruited to surveys conducted online [35] and through the telephone [36]. These studies provide evidence for using eHEALS in diverse populations, but they do not provide empirical evidence for valid age-group comparisons.…”

<p>Testing Measurement Properties of the Norwegian Version of Electronic Health Literacy Scale (eHEALS) in a Group of Day Surgery Patients</p>

“…However, the Internet use among older people is steadily increasing, as they search and visit websites for on-line health information and discussion. 34 There has been a tremendous development on the Internet front since the original eHEALS was designed 15 years ago, and, as Stellefson et al 34 point out, it is important to consider modifying the scale to better fit the relevance for use among an older population.…”