Description: This study identifies large categories of conditions that long-term care insurance LTCI carriers previously might have avoided because of a lack of data. This study estimates claims, mortality, and lapse rates for several conditions. Many of the conditions are further analyzed into risk categories, some of which would be insurable at standard rates.

Description: View the following report regarding Risk Scoring in Health Insurance: A Primer.

Description: This report presents the results of a study comparing health risk scoring models.

Description: One of the Health Section's key focuses in 2019 is Initiative 18/11. Last year the SOA joined forces with the Kaiser Family Foundation (KFF) and the Healthcare Financial Management Association (HFMA) to answer the nagging question: “What can we do about the cost of healthcare?” A steering group has been formed to lead a research project as to what managed care companies can do to reduce costs without sacrificing quality. The project [called Managed Care 3.0 in the 18/11 paper] will examine historical models that have worked as well as investigate the impact that new technologies and care models may have on cost and quality, with the goal of better understanding the current environment and identify­ing potential risks and opportunities.

Description: In health insurance marketplaces, prospective insured individuals and groups are able to choose from a range of available plans. This study uses health care encounter data to simulate the choices faced by health insurance shoppers under a variety of conditions and measures the simulated effect of increased consumer choice and consumer information on medical loss ratios and, as a consequence, insurer profitability.

Description: A discussion on podcasts relative to the Predictive Analytics & Futurism Section and how they can generate CE credits.

Description: This article explores a potential alternative to the more typical regression-based specification of risk adjustment models. We demonstrate a proof of concept that a nearest-neighbors approach may hold some promise in certain applications of health-based risk adjustment.