Artificial intelligence (AI) is making significant strides in healthcare, and KLAS has been diving deep into this topic, conducting research both independently and with industry partners. Our goal is to understand and share how healthcare organizations perceive and adopt AI and what outcomes they’ve achieved. This blog post reviews some high-level findings about AI adoption trends, challenges, and where AI is headed in the future.
For more AI research from KLAS, see our 2024 Healthcare AI report, How Health Systems Are Navigating the Complexities of AI (a collaboration with the Center for Connected Medicine [CCM]), and the 2023 Healthcare Provider IT Report (a collaboration with Bain & Company).
Points to Know
- AI adoption is increasing, despite organizations’ mixed feelings about the technology
- Organizations have encountered barriers to successfully deploying AI
- AI use cases are expected to expand in the future
How KLAS Defines AI & the Technologies It Encompasses
- Artificial intelligence (AI): Provides machine learning or natural language processing capabilities for healthcare-related areas
- Machine learning (ML): Studies and learns computer systems’ algorithms and statistical models to effectively perform tasks without requiring explicit instructions, relying instead on patterns and inference to determine results
- Natural language processing (NLP): Enables software solutions to understand, process, and analyze natural language (speech or text)
- Generative AI (Gen AI): Generates new content or data patterns to assist in medical diagnoses, treatment planning, documentation, research, etc.
- Robotic process automation (RPA): Enables software to automate repetitive and rule-based tasks typically performed by humans, boosting an organization’s overall efficiency
AI Adoption Is Increasing, despite Organizations’ Mixed Feelings about the Technology
Recent KLAS data shows that 67% of large healthcare organizations (500+ beds) are currently using a data science platform (often from their EHR vendor), most commonly for clinical and population health use cases. Unsurprisingly, large organizations have more widespread adoption than smaller organizations, since the former tend to have the resources needed to maintain, utilize, and operationalize AI solutions.
Generally, AI adoption is on the rise, and more organizations report they already have—or are in the process of creating—an AI strategy. Interest and confidence in AI have grown as available HIT solutions have evolved; many now come with built-in AI functionalities, making AI more accessible. Because of the buzz around AI, some organizations want to adopt the technology to ensure they don’t fall behind competitors.
In particular, we expect Gen AI will see greater adoption in the coming years. In a recent KLAS study, roughly 25% of the surveyed organizations said they had deployed Gen AI, and 59% said they were likely or highly likely to adopt it in the next two years. Still, many organizations are waiting to see how the market evolves before they commit to adopting.
Amid this rise in AI adoption, providers’ attitudes toward AI are mixed; many are approaching adoption with caution largely due to AI’s cost-prohibitive nature. Organizations that have more-advanced AI strategies tend to have a more positive outlook. Overall though, sentiments about AI have leveled out since three years ago, when most respondents were either very enthusiastic about AI or deeply skeptical.
Organizations Have Encountered Barriers to Successfully Deploying AI
Unsurprisingly, healthcare organizations face numerous barriers as they try to achieve their digital transformation goals. For example, healthcare leaders may be unsure of what strategies to pursue, what goals to prioritize, and what technology to use to drive results. There can also be a lack of change management, resulting in slow adoption rates.
Regarding AI, leaders will need to address a variety of concerns related to data privacy and security, especially as AI advances in healthcare and continues to be adopted. Governance policies are an important part of AI usage and data access, yet in KLAS’ study with the CCM, only 16% of the interviewed organizations reported having a governance policy. It’s possible this percentage is so low because many organizations are still in the early stages of evaluating and implementing AI solutions.
Along the same lines, organizations want to protect their data after implementing AI. They often see established/trusted vendors as safer choices when it comes to deploying AI. Security vetting processes are commonly used for protecting data, and some organizations use security and data use agreements to comply with HIPAA.
Other mentioned AI challenges include accuracy and reliability, ethical and legal concerns, and the cost of AI.
AI Use Cases Are Expected to Expand in the Future
As organizations move forward with adoption, the use cases for AI will expand. According to KLAS data, operationally focused use cases are where organizations plan to invest first, due to the burden of staff shortages, burnout, and financial pressures. Value-based care is another top-of-mind area that respondents want to address. Also, some organizations have seen early promising results with ambient speech solutions, as they have been easy to deploy and have given organizations a quick ROI.
It’s important to note that organizations most often find success when they focus on the problems they want to solve, not the shiny, new solutions that they want to use. We expect organizations will become even more thoughtful in their approach to AI as time goes on and as they evaluate what outcomes they want to achieve.
We at KLAS plan to keep a pulse on how AI usage changes over time—in fact, we are collaborating with Bain & Company later in 2024 to see where healthcare providers are investing their money and why. Keep a lookout for future reports from us and our partners as we all continue to learn what AI looks like in healthcare.
© sirichai / Adobe Stock
