Driving Patient Access and Decreasing Tech Debt in Healthcare with AI – with Aaron Chamberlain of Intermountain Health

Sharon Moran

Sharon is a former Senior Functional Analyst at a major global consulting firm. She now focuses on the data pre-processing stage of the machine learning pipeline for LLMs. She also has prior experience as a machine learning engineer customizing OCR models for a learning platform in the EdTech space.

Driving Patient Access and Decreasing Tech Debt in Healthcare with AI-1-min

Headquartered in Salt Lake City, Utah, Intermountain Health is a not-for-profit healthcare system comprised of 385 clinics and 33 hospitals dedicated to creating healthier communities and helping patients thrive. Intermountain Health merged with SCL Health in 2022 and now employs more than 58,000 people. They serve patients in Kansas, Colorado, Utah, Nevada, Wyoming, Idaho, and Montana.

Technical debt is an unsurprisingly common problem in healthcare organizations, and Intermountain Health is no exception. One key driver of technical debt across all organizations is the need to comply with regulations, which are complex and continually evolve. 

To promptly comply with changing regulations, healthcare organizations sometimes will implement temporary and sub-optimal or less efficient technical fixes that eventually need to be refactored at a later point.

Predicting the performance of EHR systems presents a challenge for engineers and administrators. EHR systems cannot adequately accommodate the changing state of knowledge present in human-entry data. Any time someone enters or deletes data, temporal inconsistencies result. Over time, this leads to technical debt.

Emerj Senior Editor Matthew DeMello recently sat down with Aaron Chamberlain, Senior Medical Director, Intermountain Health, on the ”AI in Business’Business’ podcast to talk about significant challenges healthcare leaders encounter when driving patient access and improved outcomes. In addition to discussing the approach Intermountain Health takes to reducing technical debt, the following article will focus on two key takeaways from their conversation: 

  • Creating an enhanced, seamless patient experience: Using decision intelligence at the point of initial patient contact to facilitate clinically informed discussions.
  • Integrating new technologies into existing healthcare systems: Reducing technical debt through careful measurement of related issues, and moving forward with innovative pilot programs in phases and at specific thresholds before scaling.

Listen to the full episode below:

Guest: Aaron Chamberlain, Senior Medical Director, Musculoskeletal Clinical Program, Intermountain Health 

Expertise: Value-based care, innovative care delivery models

Brief Recognition: Chamberlain was previously the Director of Quality Improvement and Efficiency in the Department of Orthopedic Surgery at Washington University School of Medicine in St. Louis.

Creating an Enhanced and Seamless Patient Experience

Chamberlain begins his podcast appearance by describing what Intermountain has learned about driving quality in patient experiences from surveying its patients. “We have a large, complicated U.S. healthcare system that’s complicated for our patients to navigate,” he says.  

Chamberlain mentions the difficulties patients have trying to understand where to seek help for the particular challenges they face. This difficulty persists even after seeing a primary care physician. One of Intermountain’s areas of focus is making the experience seamless for patients.  

There has been a greater recognition in recent years about how important it is to understand patient expectations of healthcare. To such an extent, it’s been recommended that every clinical encounter should begin with a determination of the patient’s expectations.

Chamberlain acknowledges that patients have come to accept that they shouldn’t have high expectations for their experience navigating healthcare systems. He explains that healthcare leaders have a clear incentive to drive functions and efficiencies that provide value to the patient, as they usually also bring value to the healthcare system, the payer, and other healthcare stakeholders. 

The dynamic also works in reverse: pain points that are related to decreased patient satisfaction are also more costly for healthcare systems. Chamberlain provides the example of patient expectations that are accompanied by a high-cost aspect when not met. 

He explains that when patients struggle to navigate the healthcare system when seeking care, they end up with additional visits that could be unnecessary, including more expensive ER visits.

Chamberlain thinks there is a real opportunity to utilize AI during the patient’s initial contact with the healthcare system. He explains that a clinically informed discussion at the outset facilitates connecting patients not only with the right provider but also at the right time, rather than the more common scenario involving a telephone operator routing different calls on a switchboard-style interface. 

When a nurse answers a patient’s call, the nurse can determine within a few questions whether or not the patient needs to see a primary care physician or another type of specialist. Chamberlain explains how that approach requires a large group of humans with specific expertise. He sees AI as being needed to help his organization grow and scale that process to reach more patients.

Integrating New Technologies into Existing Healthcare Systems

Every medical provider handles technical debt differently. It’s evident from the discussion with Chamberlain that Intermountain Health’s approach to technical debt is focused on using pilot programs for innovation. The ability to understand the possible impact of new technology on patient care is paramount for healthcare organizations, and the pilot programs Chamberlain describes all make significant strides in achieving that goal. 

Chamberlain confirms how technical debt in healthcare creates more work for practitioners. This additional burden reduces their willingness to embrace new technological developments readily. When asked about Intermountain’s approach to using the latest AI healthcare, Chamberlain offered valuable insight. “I think a thoughtful approach is useful,” he says. “Our approach within Intermountain is that we have a large area of focus and a tradition around healthcare innovation.” 

Aaron goes on to describe Intermountain’s approach at length, explaining how they try to identify areas within their system that are good candidates for innovative pilots. They innovate, learn from the pilots, and then scale as needed. According to Chamberlain, the work required for these pilots is all done internally. 

“Granted, we’re a large organization, but we do a lot of that work internally, through those selective pilots, where we try to identify a problem, make sure we understand the problem we’re trying to solve, and then develop, utilize, or pilot that tech in that particular area,” he explains to the executive podcast audience. 

It’s evident from the discussion with Chamberlain that Intermountain’s approach identifies issues caused by technical debt first, then minimizes them by allowing improvements to be made before a new technology is scaled up and rolled out into current workflows.

When podcast host and Emerj Senior Editor Matthew DeMello draws comparisons to financial services call center workflows, Chamberlain acknowledges that the stakes are lower for comparatively less regulated industries.

His point emphasizes that many aspects of workflows in financial services are recoverable by contrast. An incorrectly routed payment can be reversed. In healthcare, mistakes are not recoverable because healthcare providers are dealing with patients’ lives and well-being, leading to requisitely elevated compliance and regulatory and ethical concerns.

“We have a long history of watching, observing, and responding to data,” he says. Aaron then goes on to explain how Intermountain has long-standing and mature data systems and reporting systems around data such as complication rates, patient satisfaction rates, and various types of outcomes. 

During pilots, they are particularly observant of whether or not there is any deterioration in outcomes or adverse effects. He thinks the first step is to have data to measure. Organizations need to ensure they have the systems ready to measure the outcomes of pilot projects. This emphasis on data analysis supports an evidence-based approach to introducing and integrating new technologies, ensuring that any new technology introduced in workflows contributes positively to patient outcomes.

Subscribe
subscribe-image
Stay Ahead of the Machine Learning Curve

Join over 20,000 AI-focused business leaders and receive our latest AI research and trends delivered weekly.

Thanks for subscribing to the Emerj "AI Advantage" newsletter, check your email inbox for confirmation.