The Diagnostic Cost Group Hierarchical Condition Category (DCG/HCC) payment
models summarize the health care problems and predict the future health ca
re costs of populations. These models use the diagnoses generated during pa
tient encounters with the medical delivery system to infer which medical pr
oblems are present. Patient demographics and diagnostic profiles are, in tu
rn, used to predict costs. We describe the logic, structure, coefficients a
nd performance of DCG/HCC models, as developed and validated on three impor
tant data bases (privately insured, Medicaid, and Medicare) with more than
1 million people each.