Predictive Modeling

(useful as guigeposts, but still imperfect)

Health insurers are growing more comfortable with the use of predictive modeling applications to improve care for high-risk patients already enrolled in the health plan. Now, a few health plans are starting to explore the potential for these sophisticated software programs to help set premiums for patients before enrolling or renewing coverage.

Predictive modeling programs analyze a few years of past medical claims data - sometimes incorporating lab, pharmacy and other services - to predict the likelihood that a given patient will use a lot of services in the future. Often, the most expensive patients in terms of past claims history aren't those most likely to cost a lot in the future. The analysis has proven useful in identifying and stratifying patients into risk corridors prior to enrollment in disease management programs. Using predictive modeling for underwriting purposes is getting more and more prevalent in the market. Some consultion firms will now use artificial intelligence to help payers measure quality, efficiency, disease severity and outcomes.

Please let us know if you want to learn more about how predictive modeling can help you and your employees--get the most from your medical and Rx plans for the least possible cost.