The term artificial intelligence has become a buzzword, not just in industries such as manufacturing and cybersecurity, but also in healthcare. Yet many clinicians have legitimate questions and concerns about how to apply this concept in practice. Part of that concern is the common misperception that opening the door to artificial intelligence (AI) means replacing clinicians with robots, but quite the opposite is true—when AI is leveraged strategically, it can offer support for clinical decision making and help providers spend their time more efficiently and effectively.
What is AI, and why is it relevant to healthcare?
Far more complex than robots performing human tasks, AI describes learning systems that are so sophisticated that they not only take in data and external information and analyze it, but then also use those learnings to adapt and refine future responses. Given the data-heavy environment that providers practice in today it’s easy to see why there’s so much interest in applying AI to healthcare. The implications for managing patient populations with AI that synthesizes and recognizes patterns in patient data are profound, particularly given the rise of value-based care models that require a data-driven approach and prioritize coordinated care across the continuum.
Leveraging AI to efficiently allocate resources in settings across the continuum
Ensuring that patients receive the appropriate level of care – at both admission and discharge – is key for organizations whether they be health systems, ACOs, or even payers, but can be a time-consuming, administrative process. AI reduces the burden of identifying which patients require a manual intervention, usually the middle layer of patients that are neither extremely sick nor extraordinarily healthy. With its ability to manage massive amounts of data, it gives providers the information they need to manage by exception, freeing up time to focus their clinical judgment where it’s needed most. Here are three examples where AI supports decision making during care transitions:
- Admission to the hospital for observation vs. inpatient– It’s not always immediately clear whether patients should be admitted to the hospital for observation or for an inpatient stay. The wrong decision has significant implications not just in this care setting but across an entire health episode. Putting the patient on the right path at the outset, when a health problem first emerges, sets them up to receive appropriate treatment as they transition to different levels of care across the entire patient journey. Not to mention the impact on reimbursement for hospitals and coverage of services for patients.
- Discharge to a post-acute setting vs. the community – When a patient with multiple chronic conditions and no home support is ready to leave the hospital, it’s clear they should be referred to a skilled nursing facility. But what about a patient experiencing an isolated acute illness who does have support in their community? Is it safe to discharge that patient home if they’re referred to a home health agency? AI gives providers back the time they need to focus on the patients requiring extra support to select the most clinically appropriate services. As they weigh these decisions, AI also gives them the confidence to predict when patients can achieve a successful outcome in a lower-cost post-acute setting.
- Readmissions – It’s not enough for providers to be aware of recurrent hospitalizations. With almost every value-based program including a readmissions metric, today’s providers need to understand which patients are at risk of rehospitalization before they even show up at the ED. AI’s predictive power identifies these high-risk patients and frees up clinicians to focus on transitioning these patients to an appropriate level of care.
While it is still early days for AI in healthcare, the application of adaptive learning technology to the care transitions process shows great promise.
Interested in learning more on how CarePort is integrating AI into our suite of solutions? Reach out.
