Diagnostic cost group (DCG) and hierarchical coexisting conditions (HCC) models for Medicare risk adjustment

Description:

A set of risk assessment models is developed for measuring health status of Medicare enrollees and for use in risk adjustment of capitated rates for managed care organizations. Models use administrative data, elements on claims or encounter records to predict utilization. Equations are developed that use variables for demographic hospitalization indicators. Regression methods are used to develop weights for the included characteristics. Prospective models predict utilization for the year following the recording of clinical information ; concurrent models predict costs for the year concurrent with the clinical information. DCG models assign each person to a single group according the most expensive condition. Diagnoses from inpatient or inpatient and ambulatory settings may be used. HCC models characterize an individual by the presence of a condition in any of multiple classes. Within a group of related conditions, such as cancers, there are classes that vary by expected cost. Only the costliest condition in such a hierarchy is captured. Disease groups that are not related are not hierachically ranked. The models can be used to adjust capitated payments more precisely for the expected utilization of managed care enrollees than simple demographic models. Demographic models are used currently to pay Medicare risk HMOs.

subjects: Medicare,  Finance,  Mathematical models,  Economics,  Economic Models,  Risk Adjustment