Digital health literacy: comparing factor structures of the portuguese eHEALS using confirmatory factor analysis

Silva, Cláudia; Alves Lopes, António; Azevedo, Vitor Hugo; Lopes, Francisco; Ribeiro, Leonardo; Dias, Tomás

http://hdl.handle.net/10400.26/57615

Use this identifier to reference this record.

Send Feedback

Authors

Abstract(s)

Background and Objectives: Digital health literacy is essential for navigating digital health environments safely, particularly for future health professionals. Validated assessment tools are crucial for understanding competency levels and guiding educational interventions. This study evaluated the construct validity and internal consistency of the Portuguese version of the eHealth Literacy Scale (eHEALS) among health sciences students. Two models were tested using Confirmatory Factor Analysis (CFA): a unidimensional model and a second-order bidimensional model. Materials and Methods: A total of 375 undergraduate and postgraduate health sciences students (75.2% female, 23.2% male, 1.6% undisclosed) participated. CFA was performed to assess model fit using Chi-square divided by degrees of freedom (CMIN/DF), Root Mean Square Error of Approximation (RMSEA), Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), Goodness-of-Fit Index (GFI), Root Mean Square Residual (RMR), and Akaike Information Criterion (AIC). Internal consistency was assessed via Cronbach’s alpha. Results: Both models demonstrated very good fit. The bidimensional model showed slightly better fit (CMIN/DF = 1.704, RMSEA = 0.043, CFI = 0.991, TLI = 0.985, RMR = 0.015, GFI = 0.983, AIC = 67.261) than the unidimensional model (CMIN/DF = 1.767, RMSEA = 0.045, CFI = 0.990, TLI = 0.983, RMR = 0.017, GFI = 0.980, AIC = 68.047). The chi-square difference test (Δχ² = 2.786, Δdf = 1, p = 0.095) indicated that both models were statistically comparable. Internal consistency was high (α = 0.868 total, 0.850 Factor 1, 0.743 Factor 2). Conclusion: The Portuguese version of eHEALS demonstrates good construct validity and internal consistency for assessing digital health literacy in higher education students. The bidimensional model, distinguishing between "Ability to Search for Online Health Information" and "Ability to Evaluate and Apply Health Information," provides a more detailed understanding of competency areas, helping to identify specific aspects for improvement and inform targeted educational interventions.