Healthcare Risk Adjustment and Predictive ModelingThis text is listed on the Course of Reading for SOA Fellowship study in the Group & Health specialty track. Healthcare Risk Adjustment and Predictive Modeling provides a comprehensive guide to healthcare actuaries and other professionals interested in healthcare data analytics, risk adjustment and predictive modeling. The book first introduces the topic with discussions of health risk, available data, clinical identification algorithms for diagnostic grouping and the use of grouper models. The second part of the book presents the concept of data mining and some of the common approaches used by modelers. The third and final section covers a number of predictive modeling and risk adjustment case-studies, with examples from Medicaid, Medicare, disability, depression diagnosis and provider reimbursement, as well as the use of predictive modeling and risk adjustment outside the U.S. For readers who wish to experiment with their own models, the book also provides access to a test dataset. |
Common terms and phrases
Actuaries admissions age/sex AHCCCS algorithm analysis applied assessment average beneficiaries benefits calculated capitation rate Chapter chronic claims data clinical CMS-HCC cohort Condition Categories condition-based risk Connector COPD cost coverage data set decision trees demographic depression developed Diabetes diagnosis codes discussed disease DRGs DxCG eligible enrollment episode estimate evaluation example Figure function grouper grouper models HCCs health insurance health plan health risk healthcare hospital identify independent variables individual inpatient intervention linear regression logistic regression Medicaid Medicare Advantage Medicare Part D method node odds ratio outcomes outpatient overall parameters patient payment pharmacy physician PMPM population predictive accuracy predictive modeling predictors premium procedure random forests regression model reimbursement relative risk score response risk adjustment risk factors risk score ROC curve sample specific statistical Table techniques tree underwriting values weights