Objectives: Health plans expend effort to try and predict the less than 1% of their members who will need extra support in order to maintain their health status by avoiding hospitalizations and emergency department visits. In care coordination, low risk predicts low cost and acuity while high risk forecasts increased cost and acuity. However, this definition of risk leaves out members who have improved or declined in health status over time.

Methods: Blue Cross and Blue Shield of Louisiana (BCBSLA) has taken an innovative approach to risk scoring by adding a “rising risk” dimension to their predictive models. The dimension “rising risk” classifies risk scores across rising or falling traditional risk categories. BCBSLA creates the “rising risk” dimension by taking a member’s standardized risk of hospitalization over a 12-month period and fitting a line through those points (creating a slope). Members are then assigned into one of three model categories: low risk, medium to high risk, and high to very high risk. For each of the three categories, a member can be rising, flat or falling.

Results: Majority of members will not change rising risk levels over time but a subset of members will be deemed a “rising risk”. These members (low, medium-high, high-very high) appear to have the highest medical per member per month (PMPM) costs. When stratifying members by those enrolled in disease management (DM) programs by participation status, DM at medium and low-risk categories. By adjusting the slope thresholds, BCBSLA can identify select members to act on within each risk category, which maximizes care and reduces costs.

Conclusions: The benefit of “rising risk” is that it contributes additional and unique information about a member’s health status. This knowledge can lead to quicker interventions that will maximize care and reduce costs.