Allina Health Optimizes Risk Adjustment, Revenue Capture, and Sees a 6x ROI with Evidently

Challenge: Allina Health asked Evidently for assistance in optimizing capture of Hierarchical Condition Category (HCC) codes within its patient population. Allina Health had already implemented a robust clinical workflow to enable accurate HCC coding through provider education and training.  To further support providers in this work, Allina Health had also tailored its existing Electronic Health Record (EHR) tools to ease the administrative burden of entering appropriate codes.  Despite these measures, a gap remained in  the ability  to search unstructured data, as well as data outside  the EHR database (such as scans and electronic information sent by organizations outside of Allina Health).

Accurate coding is critical to stratify an individual patient’s risk and assign appropriate clinical resources. There are also significant downstream financial implications if patient complexity is inaccurately reflected.  The gap in Allina Health’s tools to support providers in HCC coding directly impacted the organization's ability to accurately reflect the scope and depth of care provided to patients.  “At Allina Health, we’re focused on equipping our workforce with modern tools to help alleviate the administrative burden so they can focus on what matters most—patient care. Evidently helps us achieve more accurate risk adjustment without asking providers to devote more time to non-clinical tasks," explained Dr. Karen Sedivy, Director of Provider Informatics at Allina Health.

Solution: To address this critical need, the organization partnered with Evidently, a leader in AI-driven healthcare solutions, which is able to analyze vast datasets and identify potential HCC opportunities. Within the initial 90 days, Evidently efficiently processed over 61,000 encounters for over 37,000 patients. The system then presented actionable diagnostic suggestions to providers for review, enriched with relevant clinical context, including associated medications, laboratory results, and AI-generated summaries, all in the native EHR workflow. Risk adjustment accuracy was increased without any provider behavior change needed, streamlining the review process.

“As a company, we aim to be a partner with our customers, and work with them to enhance existing workflows rather than recreating new ones. Our HCC work with Allina Health is a perfect example of this in action. We supercharged existing workflows with more robust data, and demonstrated a financial benefit to the organization,” described Dr. Kalie Dove-Maguire, Chief Product Officer at Evidently.

Outcome: The integration of Evidently yielded substantial improvements in HCC capture and risk adjustment accuracy. Providers engaged with the AI-generated insights, accepting the suggested diagnoses at a consistent rate of 20-30%. This proactive identification of previously uncaptured HCC codes would have otherwise been burdensome for providers to manually find, saving them chart review time while reducing their administrative task load.  In addition, the more accurate RAF scores allowed Allina Health to better assign clinical resources to meet patient care needs.  Finally, the risk adjustment translated directly into a significant expected financial impact driving a meaningful 6x ROI. This case study underscores the power of targeted AI interventions in optimizing value-based care models and ensuring accurate reflection of patient complexity, resulting in improved ability for health care organizations to care for their patients and ensure appropriate reimbursement for that care.