A national study of 6,173 U.S. hospitals published in the Journal of Medical Internet Research shows that hospitals with scaled telehealth operations are far more likely to have adopted AI tools, particularly for clinical decision-making. The finding suggests telehealth capacity could be a practical signal of an organization's overall digital readiness, but the cross-sectional design stops short of proving that telehealth expansion directly drives AI investment.
Telehealth volume tracks with AI maturity
Researchers stratified hospitals into three clinical AI adoption tiers using data from the 2024 American Hospital Association Annual Survey and Information Technology Supplement. Tier 0 hospitals (3,441 facilities) reported a median clinical AI score of zero; tier 2 hospitals (1,188 facilities) reached a median score of 22. Operational AI scores followed a similar pattern, rising from zero to 14.
Telehealth volume climbed in lockstep with those tiers, the data showed. "In this national cross-sectional study of US hospitals, telehealth scale was strongly associated with hospital AI adoption tiers, with a particularly strong relationship for clinical AI," the researchers concluded.
Missing telehealth data points to a digital divide
More than half of hospitals-57%-did not report any telehealth volume. Of those, 91.4% were concentrated in clinical AI adoption tier 0. Hospitals in tier 2 rarely lacked telehealth data. This pattern shows how digital gaps tend to cluster: organizations without telehealth infrastructure are far less likely to deploy AI in patient care.
The rural-urban split sharpens that picture. Rural hospitals with telehealth volumes comparable to metropolitan peers had weaker associations between telehealth scale and clinical AI patterns. "From a digital health and health informatics perspective, the results indicate that equitable AI adoption may depend not only on access to AI tools themselves but also on the organizational and infrastructural conditions that support digital care delivery," the study authors said. This adds complexity for policymakers aiming to close digital divides in healthcare.
Cross-sectional design leaves causality open
The study cannot establish whether scaling telehealth leads to AI adoption or vice versa. Because the data captures a single point in time, the relationship could be driven by a third factor, such as overall IT investment appetite or leadership priorities. The authors acknowledge this limitation directly, noting the design does not support causal claims.
AI adoption in healthcare keeps gaining ground
The findings arrive as AI use grows quickly in clinical settings. A separate survey of 3,151 physicians found AI utilization jumped from 47% in April 2025 to 63% in January 2026. Among those not using AI in early 2026, 31% expressed interest in adopting the technology. AI tools are increasingly woven into virtual care workflows, from automating intake assessments to supporting clinical decisions during telehealth visits.
For healthcare leaders, the study reinforces that telehealth investments and AI readiness are not siloed initiatives. The infrastructure built for remote care delivery-data pipelines, integrated workflows, clinician familiarity with digital tools-may lower the barrier for AI deployment. The connection is explored further in discussions around AI for Healthcare.
Why this matters for healthcare professionals
For hospital administrators, CIOs, and clinical informatics leads, telehealth volumes offer a tangible proxy for gauging whether their organization has the digital foundation AI requires. If telehealth metrics are flat or nonexistent, the gaps are unlikely to be limited to virtual visits-they probably extend to data integration, workflow standardization, and staff readiness for AI tools. Investing in telehealth is not just about remote care; it can act as a forcing function for the infrastructure that makes safe, effective AI adoption possible.
Your membership also unlocks:
