Determining medical necessity is a key component of the utilization review (UR) workflow.
Medicare is the largest payer in the U.S., and as such, the Centers for Medicare and Medicaid Services (CMS) defines medical necessity as:
“Services or supplies that: are proper and needed for the diagnosis or treatment of your medical condition, are provided for the diagnosis, direct care, and treatment of your medical condition, meet the standards of good medical practice in the local area, and aren’t mainly for the convenience of you or your doctor.”
Ensuring that plans of care, treatments, procedures, and other needs are medically necessary – and will be paid by Medicare or other payer organization – also ultimately supports the patient and payer, by aligning the right care at the right time, reducing administrative waste and surprise costs back to the patient.
However, when using traditional, criteria-based technology solutions for documenting medical necessity, UR processes may seem subjective, inefficient, and relegated to little more than checking boxes. This is ironic since UR was originally designed to reduce waste and overutilization of resources. When using traditional UR tools, highly educated and experienced staff may not be able to fully utilize their knowledge and abilities. That’s a huge loss for their organizations and the patients they care for.
Incorporating AI into medical necessity determinations can effectively automate large portions of the UR process. In turn, that empowers clinical staff to leverage their critical thinking skills and develop a deeper understanding of patient needs.
Medical Necessity and AI: Reducing the Administrative Burden on Skilled Clinical Staff
Digging through an EMR in search of sufficient documentation to indicate medical necessity for admission can be time-consuming and frustrating – and such administrative tasks have also been closely linked with burnout. More valuable ways for clinical staff to spend their time include: analyzing more complex cases that require deeper clinical validation, providing care coordination and transitions of care services, and developing knowledge and context about managing patient status.
Determining if patients meet medical necessity is a higher-level administrative task which better utilizes valuable clinical expertise and critical thinking skills. Nurses and other clinical staff should especially be involved to help determine and document complex cases.
AI and machine learning (ML) can be leveraged to automatically and objectively review documentation from within the EMR and make predictive patient status determinations, saving time and elevating the role of the clinician. XSOLIS’ CORTEX® assesses pivotal EMR data through the power of AI, including:
- Lab work
- Vital signs
- General wellness
- And more
CORTEX utilizes this information to generate a Care Level Score (CLS) for each patient. The score is gradient versus a traditional binary output from a point in time. It is continuous, objective, and works when you aren’t. Higher scores strongly suggest an inpatient status. Conversely, patients with a lower score suggest a status of outpatient with observation services.
Healthcare AI That Complements and Empowers Clinicians
Providing a ranked list of patients, updated real-time throughout the stay, allows nurses and other clinical professionals to prioritize their time where it’s needed most – to focus on the revenue sensitive cases first. Minutes and hours previously spent reviewing the chart, looking for criteria, and checking boxes can be directed toward:
- More focused interactions with a smaller group of higher-risk patients
- Working to resolve more complex patient issues and concerns
- Developing more effective and personalized discharge plans
CORTEX, as a framework for collaboration between providers and payers, also makes patient data from within the EMR more readily available between both parties, providing for shared and more focused views. That means it’s easier to identify and communicate the specifics that drive effective defenses of medical necessity decisions – which leads to more first-touch determinations, collaboration between providers and payers, and elevated discussions about “what’s missing” versus previous back-and-forth communication, especially with the more complicated cases.
Nurses, physicians, and other skilled healthcare professionals are most effective when they can leverage analytics and AI to streamline their workflow, while also applying their uniquely human abilities, like critical thinking in conjunction with their specialized knowledge and experience, to UR processes.
Incorporating AI into medical necessity decision-making and the broader utilization review process can benefit all stakeholders, elevating the role and job satisfaction for case managers, creating clinical efficiencies, saving both payer and provider organizations time and money, while improving the visibility of patient status and appropriateness of care.
Patty Dietz BSN, RN, CPHQ, ACM-RN, is Senior Clinical Consultant at XSOLIS and has over 20 years’ experience in healthcare technology consulting, organizational performance improvement, and case management.