The Digital Health Paradox: What Actually Predicts Internet-Based Provider Communication In Rural Black Belt Alabama

Background: As healthcare delivery increasingly moves online, disparities in digital health engagement risk deepening longstanding inequities — particularly where place-based structural conditions shape who gains meaningful access to digital care. Alabama's Black Belt, a predominantly African American region whose poverty and health inequities are rooted in the legacy of slavery and sharecropping, shows how neighborhood-level disinvestment produces compounded barriers, including hospital closures, provider shortages, and limited broadband infrastructure. Far less is known about what drives meaningful use in place-based contexts where structural barriers intersect with historical medical mistrust. This study examined how structural conditions in two rural Black Belt communities shape correlates of internet-based provider communication, applying Andersen's Behavioral Model to identify predisposing, enabling, and need factors.

Methods: A cross-sectional study was conducted with 160 adults from two community-embedded settings in rural Black Belt Alabama — an AIDS service organization serving people living with HIV and a community center serving residents with diverse healthcare engagement histories. Participants completed surveys organized within Andersen's framework: predisposing factors (age, gender, education, marital status), enabling factors (internet access, mobile phone access, telehealth use, health insurance, transportation), and need factors (HIV status, PTSD symptomology). Technology-related behaviors — telehealth use and internet use for health information — were examined as enabling factors with particular relevance to digital health contexts. The primary outcome was self-reported internet-based provider communication. Bivariate analyses and multivariable logistic regression identified independent correlates.

Results: Of 160 participants, 109 (68.1%) reported internet-based provider communication. Technology-related behaviors — not internet access itself — distinguished those who communicated with providers online from those who did not. Counterintuitively, higher education and internet access were inversely associated with digital provider communication in bivariate analyses, suggesting more resourced individuals rely on in-person or specialist care — consistent with the tiered care landscape produced by neighborhood-level disinvestment. In adjusted logistic regression, traditional predisposing and socioeconomic enabling factors lost significance entirely. Only technology-related enabling factors remained significant: telehealth users were five times more likely to engage in internet-based provider communication (OR = 0.199, 95% CI: 0.066–0.557, p = .003), and internet health information seekers seven times more likely (OR = 0.136, 95% CI: 0.033–0.451, p = .002). Prior digital health engagement — not demographics or income — predicts continued digital engagement with providers.

Conclusions and Implications: These findings reveal that place-based structural conditions shape digital health behaviors of predominantly African American rural residents in ways traditional applications of Andersen's Behavioral Model do not capture — and that technology-specific behaviors must be explicitly incorporated as enabling factors in digital health contexts. For social workers and community organizations, digital health literacy programs, supported first telehealth experiences, and community-based technology navigation services are essential — especially in communities where structural racism has driven both poor health outcomes and limited digital access. For policymakers, connectivity investment without community-level capacity building will not close the digital health gap. Treating technology engagement — not merely access — as a social determinant of health is an urgent priority for a field committed to racial justice and equitable health.