OBJECTIVES:

Enhanced Care and Advocacy are critical in ensuring better health outcomes, improved member experience, and priority for payers and providers is to provide this enhanced service only to selective few with complex and high needs. This study focuses on various programs leveraging AI/ML models to identify High and complex needs membership, and value derived from these programs. Complex Health membership includes members with complex, chronic conditions with high medical and service needs.

METHODS:

Complex health member identification framework was created which used advance AI/ML approaches and looked at 360-degree view of the member to identify complex and high utilizer population. Comprehensive member data including clinical, transactional, and behavioral features were used in this analysis. Complex member cohorts were identified via landscape analyses which investigated members health complexity basis comorbidities, special needs, chronic disease profile and health chapters. Various features were created across the categories such as Cost (Medical cost, Rx cost) Contact Behavior (Calls, Repeat calls), Claims (Total claims, Denials, Appeals), Prior Authorization (Denials, Appeals) and Customer Feedback (NPS, Written Grievances) were used to generate risk scores of members that provided the propensity of a member to have higher score on each of the above dimension. The study period for the analysis was from Jan 2021 to December 2021. GBM, Random Forest AI/ML models were used

RESULTS:

Model performed well with 85% accuracy level for the identification of the targeted members. The high-risk population selected in the top decile showed 15X Medical cost, 12X operational cost, and 8X member experience / NPS opportunity

CONCLUSIONS:

Study shows that there is a betterment in member’s experience, which increased NPS of the members. By assigning personalized advocates to these high-risk members, the problem of disjoint contacts was resolved, which improved the overall health outcome of the members. This also led to Medical and operational cost reduction