Hospitalization prediction is a refinement to the predictive models specifically calibrated to identifying patients with risk of future hospitalization. The hospital prediction models focus on unanticipated hospitalizations and are complementary to the cost prediction models. There are five predictive model outputs related to the likelihood of hospitalization. These models are intended to be used for the indicated outcome.
A specific model was developed for capturing unexpected high pharmacy users to address an important group of patients who offered opportunities for care management. Specifically, these are patients who had future pharmacy expenditures that were much greater than what had been predicted based upon their morbidity profile alone. These are patients who may be taking unnecessarily expensive medications, taking multiple medications within the same therapeutic class (poorly coordinated care), abusing medications, or experiencing data reporting problems. The model output, the product of a two-part regression, is the probability of experiencing unexpectedly high pharmacy use.
Postponing or slowing disease progression retains an individual's quality of life and reduces healthcare spending.