The Centers for Medicare & Medicaid Services (CMS) will restrict how Medicare Advantage plans use predictive technology tools, including predictive AI tools, to make coverage decisions, starting in January 2024.

With the advancement of predictive technologies, some health plans are using predictive technology tools to sift through massive amounts of clinical data, from claims, electronic health records (EHRs) and other sources, to identify clinical patterns for patients with similar diagnoses and characteristics. The new CMS rule requires those plans to “ensure that they are making medical necessity determinations based on the circumstances of the specific individual, as opposed to using an algorithm or software that doesn’t account for an individual’s circumstances.”

CMS stipulates that artificial intelligence (AI) tools must still follow Medicare coverage criteria and cannot deny benefits that original Medicare covers. If plan-specific criteria is added, where Medicare criteria is believed to be too vague, the medical evidence supporting the algorithm must be disclosed.

Finally, in cases where coverage is denied based on medical necessity, review by a physician or other appropriate healthcare professional is required.

Cohere applauds CMS’s decision and is committed to the responsible use of AI to improve care outcomes for patients. Our offerings enable full compliance with CMS’s predictive technology requirements.

Cohere Health’s patented AI technology has been used by over 360 thousand medical providers across all fifty states to improve the care of 17 million members.

When leveraged appropriately, AI drives much-needed impact on the broken prior authorization process. Ninety-four percent of physicians report care delays from typical prior authorization processes. Intelligent prior authorization leverages optical character recognition (OCR), natural language processing, and machine learning models to automate the process and to avoid denials and appeals by enabling a medically appropriate request before submission. For example, manually submitted fax forms are digitized, with OCR converting documents to machine-readable text, allowing health plans to capture important clinical information previously locked away from digital interphases.

By leveraging health plan policy and claims information, machine learning models identify missing clinical information from authorization requests and deploy in-workflow notifications, known as “nudges,” to ensure authorizations are clean, accurate, and complete before submission. As a result, more service requests become eligible for auto-decisioning, saving health plans and providers resources and physicians and patients valuable time. As a result, this technology helps health plans avoid expensive, time-consuming and abrasive denials, appeals and overturns.

With prior authorization digitized, health plans can drive even greater clinical quality and safety impact for their members. Cohere’s clinical innovation solutions apply AI, machine learning and other innovative technologies to a combination of evidence-based guidelines, population health benchmarks, claims, provider patterns, and prior authorization data sets to improve outcomes for patients and avoid unnecessary utilization. Machine learning algorithms match patients to care paths best-suited to their clinical and socioeconomic situation, which enables a higher likelihood of a faster, safer and better patient outcome.

Cohere’s care path approach aligns physicians across the health plan network on evidence-based care and adds clinical context to prior authorization decisions. This added context enables the health plan to leverage these transactions as opportunities to enhance the end-to-end patient journey through innovative features like nudges and episodic authorizations. This is the opposite of using algorithms to restrict care; it’s the use of algorithms to enable better care.

Cohere’s AI technology has been responsibly applied to more than 5.5 million prior authorization requests through the Cohere Unify™ Platform and has enhanced care quality for millions of health plan members nationwide.

Again, Cohere applauds CMS’s actions around predictive algorithms and coverage decisions. Visit AI Headquarters to learn more about how AI technology, in compliance with CMS regulations, can improve patient care within the prior authorization process.

Published On: October 25th, 2023Categories: AI/ML Cohere Content, Blog

Share:

About the Author: Brian Covino, M.D., FAAOS

Dr. Brian Covino oversees more than 50 physicians as Cohere Health’s Chief Medical Officer. After practicing orthopedic surgery for more than 25 years, Dr. Covino joined Cohere in 2020 after having served as a consultant since 2018. During his years as a practicing surgeon, Dr. Covino was a partner at Knoxville Orthopaedic Clinic/OrthoTennessee specializing in joint replacement surgery. He holds a bachelor’s from Harvard University as well as an M.D. from Georgetown University School of Medicine. Dr. Covino received his surgical training at the University of Virginia Graduate School of Medical Education and completed a fellowship at The Cleveland Clinic Foundation.