Setting the policies, processes and principles that guide the development, deployment and use of augmented intelligence (AI) tools is a two-part process at Atlantic Health System, a large nonprofit health system with more than 500 locations in New Jersey and Pennsylvania.
Integrated security systems initially step in to ensure that AI processes are safe from a data privacy and legal perspective. Separately, there’s a team of physicians and other health professionals, data analysts, statisticians and ethicists to examine the clinical impact of AI on Atlantic Health’s patients.
Leaders at health care organizations seeking to make use of AI’s potential can be easily dazed by the governance challenges involved in making sure it is used wisely. This is why it's so important to set governance priorities “because these projects really come from everywhere,” noted Suja Mathew, MD. She is executive vice president and chief clinical officer at Atlantic Health System, which is part of the AMA Health System Member Program that provides enterprise solutions to equip leadership, physicians and care teams with resources to help drive the future of medicine.
AI is near top of mind for many physicians and health care leaders. Most health systems have already incorporated AI into their protocols, piloting new tools, said Dr. Mathew. A recent AMA survey (PDF) found that doctors’ use of AI rose 74% in just a year. Nearly two-thirds of physicians surveyed reported using health care AI in 2024, compared with 38% of physicians who said they used it in 2023.
That development is exciting, but AI—often called artificial intelligence—presents some challenges, said Dr. Mathew, who joined Margaret Lozovatsky, MD, AMA’s vice president of digital health innovations for an AMA Insight Network webinar on health AI that is available on demand. Among other benefits, the AMA Health System Member Program provides access to the AMA Insight Network’s Quality, Safety and Equity community. This virtual forum provides an opportunity for like-minded leaders from across the country to hear more examples of how leading systems are finding innovative ways to address health care inequities in their communities.
“Having a clear process for prioritization is going to be really important as we think about the speed at which these tools are being developed and brought to the clinical spaces,” said Dr. Lozovatsky.
From AI implementation to EHR adoption and usability, the AMA is fighting to make technology work for physicians, ensuring that it is an asset to doctors.
First step: engage your executive leaders
Dr. Lozovatsky fielded questions from Dr. Mathew on AI governance, highlighting some of the steps
for establishing effective governance to integrate AI into health care systems. These include setting priorities, developing policies and establishing processes for planning and implementation.
In its AI survey, the AMA found that 82% of physicians recognized the importance of thinking about the safety and the efficacy of AI tools. “We all are in medicine to take care of patients, and we are all very motivated to make sure we're providing the best care,” said Dr. Lozovatsky.
But AI governance can feel overwhelming. “It feels like there's a lot to do, but if you had some steps that were outlined for you, it makes it easier. And there are a few components that are critical to effective governance,” she said.
First off, it’s important to establish that strategic arm. ... Executive leaders should be engaged in the AI process, understanding its goals to guide the rest of the system. Underneath this is a working group that helps think through all the steps of AI implementation, from intake to adoption, monitoring and oversight.
In developing an AI strategy, health systems should assess how AI would support the strategic goals of the organization.
“Having that connection to the executive leaders and aligning these tools to the strategic goals of the organization is really critical for them to be successful,” said Dr. Lozovatsky.
Learn more with the AMA about the emerging landscape of augmented intelligence in health care (PDF).
Principles help set AI policies
It is also important for physicians and health care leaders to set guiding principles that align with strategic initiatives. A health system must decide: Does it want to be an early adopter of these AI tools? Which of these AI tools are most critical and best aligned with organizational imperatives?
Other things to consider:
- What ways do you plan to support this vision and what resources will you need?
- Who will ensure that there is compliance and thoughtful decision-making pathways that align with the needs of the organization?
- How will this be incorporated into existing structures?
“Setting that stage by the executive leaders will empower the teams to go out there to do the work, to be able to set up these structures and then ultimately to have successful use of the tools,” said Dr. Lozovatsky.
Health systems should also establish a process for evaluating AI tools, and ensuring they’re safe for end users. “How are you going to engage stakeholders and think about communicating that out? And then, what is going to be your feedback mechanism?” she said.
In terms of governance, there is a need to set policies at the organizational level for use, monitoring, training and more, as well as develop a way to engage physicians and other health-professional end-users across the health care organization.
Establishing an AI leader
At health systems, chief medical information officers have a very important role in the leadership chain of AI governance. They are the lynchpin between the technology team and clinicians and often set the strategy on how technology can support clinicians.
“Whatever the structure you choose—and whether you choose to have an AI officer or not—make sure that it is well integrated into the current clinical technology environment, especially because of the speed that we know that this is moving at,” said Dr. Lozovatsky. “If we don't do that and we follow some of the pathways that we had done in the past, it can create challenges with implementation, adoption and utilization of these tools.”
Governance is a partnership between all the leaders. All the leaders have a role in this space. This is truly a multidisciplinary approach because these AI tools will support both the clinical enterprise and the administrative spaces, she noted. There are tools that are going to be used in supply chain. There are tools that are going to be used in the financial groups in the organization.
“It’s going to be important for all of the leaders to be engaged in ensuring that the right representatives are at the table,” she said.
Explore further and earn CME with the “AMA ChangeMedEd® Artificial Intelligence in Health Care Series,” available on the AMA Ed Hub™.
Stay up to date on AI
Follow the latest news on AI and its applications and effects for health care—delivered to your inbox.
