JMIR Publications

ABSTRACT

Background:

The digital divide has evolved from a simple binary concept of connectivity into a three-level framework encompassing access, usage, and outcomes. In the context of aging populations, this divide may significantly impact cognitive health, yet empirical validation of this framework remains limited.

Objective:

This study aimed to document the cascade of influences from digital access to internet use and then to cognitive outcomes and examine how sociodemographic factors affect each element of this pathway over time.

Methods:

A longitudinal analysis was conducted using nationally representative data from middle-aged and older adults in China. Parallel Process Latent Growth Curve Modeling (PP-LGCM) was employed to validate the three-level digital divide construct and track its impact on cognitive trajectories.

Results:

The β coefficients of potential influence showed that digital access more strongly influenced internet use (β=0.53) than internet use affected cognition (β=0.19) at baseline. However, this pattern shifted over time, with the impact of usage growth on cognitive trajectories (β=0.31) becoming more pronounced than access growth on usage development (β=0.25). Age had a stronger negative effect on the growth of digital usage (β=-0.41) than on initial access and cognition, while education was positively associated with baseline digital engagement and showed strong benefits for initial cognition (β=0.48). Rural residence was associated with lower initial digital access (β=-0.30) and continued to impede the development of digital usage despite positive growth in access.

Conclusions:

To address cognitive health disparities in aging populations, interventions must target both immediate and sustained digital engagement following the digital divide pathway over time, with a particular focus on improving digital access and use among older, less educated, and rural populations.