Healthcare leaders are undoubtedly facing an onslaught of questions from their boards and other executives on the topic of, what else, AI. The enthusiasm that I and many others feel for the opportunity to leverage AI in healthcare stems from a shared feeling that we need to do better and that now we can. Electronic health records (EHRs) have been foundational advances, moving us from analog to digital healthcare data. Yet our current American healthcare system is unsustainable, unable to deliver the cost, quality, experience, equity, and access that patients deserve and that healthcare organizations aspire to.
Having deployed EHRs, we now have a generational opportunity to unlock previously intractable solutions, reducing burdensome work queues and tedious tasks, democratizing access to care and achieving what Dr. Atul Butte calls “scalable privilege,” augmenting the physician-patient relationship, personalizing treatment and engagement, and so much more.
2023 was a year of rapid and significant advances in artificial intelligence (AI) technology. We saw the rise of generative AI, large language models (LLMs), conversational AI, ChatGPT, and more. Headlines and conversations related to these innovations are seemingly everywhere as the industry grapples with their potential impacts.
KLAS Research found that 64% of healthcare executives see improving operational efficiency as the biggest opportunity for generative AI in the industry. Those same executives are focused on leveraging AI to address the documentation burden, to be able to personalize patient communications at scale and automate tedious manual workflows.
I recently spoke with leaders from dozens of health systems of all shapes and sizes. The consensus, when it came to AI, was that healthcare organizations have an opportunity right now to take an enterprise-wide approach that encompasses everything from workforce to infrastructure to governance and beyond. Those conversations, combined with the experience of deploying AI to thousands of sites of care, have led to these ten essential elements of an enterprise-wide AI strategy for healthcare organizations.
Have a vision and communicate it
While it is valuable to have a broad strategy that enables AI across your organization, it is also an opportunity to have a focused vision. Is there a specific opportunity for your organization to advance your unique brand and mission that is now unlocked by AI capabilities that might have previously felt unattainable? Seek something bold, aspirational, that inspires the institution, and that can provide institutional focus.
Develop guiding principles
Because AI has such massive potential implications, for both good and potential harm, any AI strategy must include core Guiding Principles that underlie your work and inform stakeholders across your organization.
Stay laser-focused on business objectives
AI should be viewed through the lens of how it can support and accelerate your existing business strategy. Whether you are primarily focused on trying to build a population health capability, working to grow through a consumer focus, delivering the highest quality complex care, or reducing administrative burden, your AI projects should focus on your core strategy and on solving your hardest, most important challenges. Alternatively, some organizations may want to utilize AI to test the development of new business models aligned with their strategy.
Go deep on process improvement, adoption, and change management
McKinsey found that the biggest predictor of organizational success with AI came down to those who excel in design thinking. That’s because AI cannot be successful in a vacuum, but must be thoughtfully deployed into the right workflows. But, it doesn’t stop there. Even the most intuitively designed solutions can fail due to a lack of related change management. Your organization needs to build institutional capability in managing through change, and in creating business cases that capture value in innovative ways.
Reimagine the workforce
Assess the current state versus the future state of your workforce and kick off a transition plan that includes training and skill development, as well as updating and shifting certain roles and responsibilities. McKinsey found that the highest performing organizations invest the time to teach users how their AI models work and have a dedicated training center that develops nontechnical personnel’s AI skills through hands-on learning.
Make risk, governance, and safety a priority
Most likely, it is not necessary to start from scratch. AI governance can typically build from your existing governance and risk policies, and a robust AI and data governance program can pave the way for rapid deployment and adoption.
Get technical
More than half (56%) of US healthcare provider executives surveyed by Bain & Company cited software and technology as one of their top three strategic priorities moving forward, an indication of the growing strategic importance of IT. As technology grows in importance, so too does the most senior IT executive role in a healthcare organization, typically either a Chief Information Officer or Chief Digital Officer. As such, AI strategies and AI initiatives must be a partnership between operational leaders and IT leaders. To take advantage of AI, organizations must solidify their core. That means a focus on building and integrating data from different core systems, implementing data governance, and creating data-sharing frameworks.
Partner with organizations who share your values and complement your expertise
Because AI technology is rapidly changing, ensure that your partners are aligned with your core values, and share a vision of the problem space you are tackling together, so that you can make continued progress over time. Choose partners with expertise that complements your own. As you identify your core platforms and technology partners, carefully select solutions that offer multiple, synergistic, integrated offerings because they provide economies of scale over time.
Make agility and rapid iteration the norm
The world is moving faster and faster, with complexity and uncertainty. Traditional approaches that rely on long-term planning can leave organizations flat-footed and slow to adapt. Developing institutional agility, through processes, structures, culture, and mindset to enable rapid experimentation and learning, is a critical enabler to success.
Focus on and celebrate the early wins as the strategy is deployed
Any large strategy brings the risk of boiling the ocean, with lots of time, resources, and energy burned and little to show for it several years down the line. Be mindful to place a few early bets on areas of high importance, to deliver early impact that will build institutional confidence as well as drive enterprise-wide learnings and the development of key capabilities.
The bottom line: learn and build capabilities through doing, and resist the common institutional temptation to plan, plan, plan, before ever trying to do.
Conclusion
Many out there are wondering how we, as an industry, can afford the risks posed by AI. I think that’s the wrong question. Instead, we should ask ourselves how we can afford to maintain the status quo. I hope these ten tips help you launch into the new year prepared with an AI strategy for your organization.
Photo: chombosan, Getty Images
