AI promises relief for rural hospitals and adds risks they are not staffed to manage

In many ways, rural hospitals are primed to adopt AI. Hospitals constantly fighting budgetary constraints could leverage the technology to do more with less as they serve their high-risk patient populations. Their workforces, stretched thin and burned out, could use it to cut down on administrative burden and bolster clinical decision support.

“[Rural health systems] are natural incubators for innovation because they are working with limited resources,” Alycia Bayne, a public health researcher with NORC Walsh Center for Rural Health Analysis at the University of Chicago, said.

But rural health systems face a catch-22. Those limited resources that AI could augment also have the potential to fuel privacy and cybersecurity risks.

Thin data and thinner staffing complicate AI adoption for rural hospitals

AI is being marketed as a tool to support both administrative and clinical functions, but adopting these tools isn’t as simple as buying them and turning them loose.

“You have a data privacy paradox. If you’re trying to improve model efficiency, you need richer rural data,” Qian Huang, PhD, MA, MPA, a research faculty member with the Center for Rural Health and Research at East Tennessee State University College of Public Health and author of the paper, said. “However, rural data have fewer patients, and you need to suppress the data.”

Smaller rural populations and the need for strong privacy protection can impact the data quality and amplify the risks of underrepresentation and bias, according to Huang.

Whether rural hospitals plan to develop AI tools internally or turn to third-party vendors, they cannot expose or compromise protected health information.

“One of the greatest challenges or concerns that I’ve heard from rural health system leaders [is] ensuring that they have the right systems and checks and balances and guardrails in place for knowing: How are these data being used? Where do they live? How long are they on our system? Who has access to them?” Bayne said.

Rural hospitals also must manage the cybersecurity challenges inherent to AI adoption. The technology contributes to a growing attack surface and introduces more third-party risk.

Managing the data privacy and cybersecurity concerns associated with AI requires human expertise, but rural health systems are typically staffed by small teams working at capacity. AI could eventually help these teams do more, but before that happens, people need to learn how to use these tools.

Bayne — the lead author of the AI in Rural Healthcare Toolkit, developed by the NORC Walsh Center for Rural Health Analysis and Rural Health Information Hub (RHIhub) — highlighted the workforce challenges that complicate AI adoption in rural healthcare settings.

“They have limitations with respect to providers and cybersecurity staff and IT staff and informatics teams,” she said. “AI has the potential to make processes more efficient. But at the same time, it’s very difficult to implement AI without workforce capacity.”

Horizon Health, located in Paris, Ill., is a critical access hospital serving Edgar County, home to approximately 16,000 people, according to Kym Pfrank, vice president and COO of Horizon Health. Thus far, the rural healthcare provider is taking a slow approach to rolling out AI, beginning with ambient listening for physician practice charting and documentation. Pfrank is grappling with how to include user education with the rollout of AI.

“We don’t have a large staff. We just have normal routine business, and everyone is busy,” he said. “It’s just another thing that we have to try to do, and it is a limitation.”

Rural hospitals could face even more resource pressure with Medicaid cuts expected to squeeze already tight budgets, which could translate into greater cybersecurity risks.

“Any amount of money that is cut does impact what we can do operationally,” Pfrank said.

Rural hospitals are playing catch-up on AI governance

Governance is essential to safe and secure AI implementation and use, but getting a framework and ongoing oversight in place is a formidable task for resource-strapped rural health systems.

“AI is evolving so rapidly,” Bayne said. “What we’re finding is health systems are implementing these tools, and then, they’re actually doing what’s called a look-back process.”

Pfrank shared that Horizon Health is in that position right now. “At this point, we don’t have a formal governance policy. That’s one of the things we’re working on,” he said.

Rural hospitals can tap resources for responsible AI adoption

Rural hospitals have the potential to benefit enormously from AI, but achieving those benefits requires a disciplined path to embracing this technology.

“We cannot just have a plug and play [approach]. That’s the most dangerous thing,” Huang said.

While there is currently little in the way of AI regulation to guide how AI is adopted in healthcare, rural settings or otherwise, there are some resources to help guide rural hospitals.

The Centers for Medicare and Medicaid Services’ Rural Health Transformation Program (RHTP) is providing $50 billion to help states improve rural healthcare. Many states are planning to use funds to support AI adoption initiatives.

Toolkits such as the one developed by the NORC Walsh Center for Rural Health Analysis and RHIhub, offer rural healthcare leaders guidance on where to begin identifying opportunities for AI and moving forward with adoption. Industry associations like the American Medical Association have healthcare-specific resources, like the recently released governance framework for the use of agentic AI in clinical functions.

As enormous as the potential of AI seems, getting to that transformation will be an incremental process for rural hospitals.

“Start small,” Bayne said. “Small use cases, low-risk use cases are really where we’re seeing rural health systems begin.”

As these use cases scale, Huang hopes to see the development of metrics rural health systems can use “…for monitoring what’s going on and to identify the discrepancy between what we want to do and what we’re currently doing.”

Rural hospital leaders are under the familiar pressures of workforce and budget limitations, and AI could be part of the solution if healthcare providers can use it without inadvertently introducing even more risk to data privacy, cybersecurity and, critically, patient outcomes.

“I do see some possibilities. The question I have is how fast and how…well can we implement it?” Pfrank said.

Carrie Pallardy, a Chicago-based freelance writer and editor, began her career covering healthcare more than a decade ago. Her work has taken into many different industries, but covering healthcare delivery remains a constant focus. She can be reached at [email protected] or on LinkedIn.