The Association Between eHealth Literacy and Health Behaviors During and Since the COVID-19 Pandemic: Systematic Review and Meta-Analysis

Clinical Snapshot

Key Findings

• P Value: Statistically significant for correlation and grouped OR syntheses; non-significant for continuous OR synthesis (CI crosses 1.0)

• Effect Size: Correlation-based synthesis: pooled r=0.43 (moderate positive association); Grouped OR synthesis: pooled OR=2.12 (higher eHealth literacy associated with approximately twice the odds of favorable health behavior); Continuous OR synthesis: pooled OR=1.07 (non-significant)

• Primary Outcome: Association between eHealth literacy and health behaviors (decision-making, health-promoting, and health management) during and after the COVID-19 pandemic

• Nnt Or Sensitivity: No NNT calculable from observational data. Prediction intervals are wide and clinically important: correlation PI 0.22–0.61 (all positive but variable magnitude); grouped OR PI 1.11–4.06 (all above 1.0 but wide); continuous OR PI 0.84–1.37 (crosses 1.0, indicating that in some future populations the association may be null or negative). GRADE certainty: Low (grouped OR) to Very Low (correlation and continuous OR syntheses).

• Confidence Interval: Correlation: 95% CI 0.36–0.51; Grouped OR: 95% CI 1.47–3.05; Continuous OR: 95% CI 0.89–1.30

Clinical Application

eHealth literacy is a modifiable determinant that can be targeted through digital health literacy programs, patient education initiatives, and co-design of user-friendly digital health services. However, the low-to-very-low GRADE certainty means that investment in eHealth literacy interventions should be accompanied by rigorous evaluation rather than assumed to produce behavior change. The authors appropriately recommend pairing eHealth literacy improvement with trustworthy digital services and clinician support. In Australia, eHealth literacy is increasingly relevant given the national My Health Record system, telehealth expansion post-COVID-19 (Medicare Benefits Schedule telehealth items), and the Australian Digital Health Agency's national strategy. The RACGP supports digital health integration in general practice, and the TGA regulates digital health tools. However, this review's evidence base is likely dominated by non-Australian studies, and Australian populations differ in digital infrastructure access, health literacy baselines (particularly among Aboriginal and Torres Strait Islander communities and culturally and linguistically diverse populations), and healthcare system navigation. PBS-listed medications and Australian preventive health programs (e.g., National Bowel Cancer Screening Program, immunisation schedules) represent specific behavioral domains where eHealth literacy may be relevant but where direct evidence from this review is lacking. Australian clinicians should treat these findings as hypothesis-generating rather than practice-changing. Adults using digital health platforms for health information seeking, particularly those navigating pandemic-era or post-pandemic healthcare. Subgroup evidence suggests the association may be stronger in older (geriatric) populations, though this finding requires replication.

Abstract

BACKGROUND: Since COVID-19, health information seeking, service navigation, and routine care have become increasingly digitally mediated. It remains unclear whether the association between eHealth literacy and health behaviors is consistent across behavioral domains, populations, and analytic frameworks. OBJECTIVE: This systematic review and meta-analysis synthesized COVID-19 and post-COVID-19 evidence on the association between eHealth literacy and health behaviors and examined variation across study contexts. METHODS: We conducted a PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 2020-compliant, PROSPERO-registered review (CRD420251009048). PubMed, Embase, Web of Science Core Collection, CINAHL Ultimate, and Scopus were searched from January 1, 2020, through March 27, 2026. Eligible observational studies assessed eHealth literacy and reported analyzable associations with health behavior outcomes collected in 2020 or later. Health behaviors were classified as health decision-making, health-promoting, or health management behaviors. Correlation coefficients, grouped odds ratios (ORs), and continuous ORs were synthesized separately using random-effects models with Knapp-Hartung adjustment. Certainty was assessed using the Grading of Recommendations Assessment, Development, and Evaluation framework. RESULTS: In total, 19 studies were included: 10 contributed correlation coefficients, 6 grouped ORs, and 3 continuous ORs. In the correlation-based synthesis, higher eHealth literacy was associated with more favorable health behaviors (pooled r=0.43, 95% CI 0.36-0.51; 95% prediction interval 0.22-0.61), with substantial heterogeneity (I2=80.50%). In the grouped OR synthesis, higher eHealth literacy was also associated with more favorable health behaviors (pooled OR 2.12, 95% CI 1.47-3.05; 95% prediction interval 1.11-4.06), with moderate heterogeneity (I2=46.33%). In the continuous OR synthesis, all 3 studies showed positive associations, but the pooled effect was not statistically significant (pooled OR 1.07, 95% CI 0.89-1.30; 95% prediction interval 0.84-1.37), with very high heterogeneity (I2=97.98%). Subgroup analyses showed a significant difference only by geriatric status in the grouped OR synthesis. Certainty was low for the grouped OR synthesis and very low for the other 2 syntheses. CONCLUSIONS: Higher eHealth literacy was generally associated with more favorable health behaviors in the correlation-based and grouped OR syntheses, whereas evidence from the continuous OR synthesis was inconclusive. Given the predominantly cross-sectional evidence base, heterogeneity, risk-of-bias concerns, and low to very low certainty, the association should be interpreted as contextual and associative rather than causal or uniform. This review is innovative in synthesizing COVID-19 and post-COVID-19 evidence, applying a functional classification of health behaviors, and analyzing distinct effect measures separately. Unlike previous reviews that summarized the association more broadly, it avoids a single mixed pooled effect and provides a cautious, context-specific interpretation. In practice, interventions should pair eHealth literacy improvement with trustworthy digital services, clinician support, and behavior-specific conditions that help translate digital information into sustained health-related action.

References

1. 1.Ruan, B., Tian, J., Wang, X., & Su, D. (2026). The association between eHealth literacy and health behaviors during and since the COVID-19 pandemic: Systematic review and meta-analysis. Journal of Medical Internet Research. https://doi.org/10.2196/94233