AI and ML Applications in Supplemental Health Plans: ReducingOut-of-Pocket Costs through Predictive Insights
DOI:
https://doi.org/10.47363/JMSMR/2024(5)189Keywords:
Supplemental Health Plan, Customer Outcomes, Enterprise Financials, Machine Learning, Predictive Modeling, Customer Behavior, Out-of-Pocket Costs, Service Utilization, Plan-Service Utilization, Predictive Variables, Plan Administrator, Under-Insured Products, Insurance Policyholders, Customer Segmentation, Cost Reduction, Critical Customer Segments, Insurance Products, Healthcare Cost Management, Financial Optimization, Service ImprovementAbstract
Many service, support, and cost variables within a Supplemental Health Plan directly influence both customer outcomes and enterprise financials. It is important to model these variables and uncover those that are most predictive of certain customer behaviors. This research study applies machine learning techniques to model how different customer parts of a Supplemental Plan can reduce their out-of-pocket costs and which of a set of plan-service-utilization specific variables are most predictive. These predictive models, in turn, enable the Plan Administrator to improve service for a small but critical customer segment: different parts of the insurance policyholders who have ‘under-insured’ products and have more out-of-pocket compared to the rest of the insurance policyholders. By identifying these predictive variables, the Supplemental Plan Administrator can now select the top variables that will influence costs. The result of deploying these models is an out-of-pocket reduction for the different customer segments while retaining acceptable enterprise financials.
