Verisk Health White Papers and Research

Verisk Health’s scientists, clinicians and technologists conduct ongoing research that continually refines and improves the analytic capabilities of products we offer our healthcare industry clients. From time to time, we publish papers of interest on topics related to risk adjustment and prediction.

Please note that registration is required to view white papers below.

Medical vs. Pharmacy Risk Solutions for the Purpose of Risk Adjustment

A key differentiator among product modules in our Sightlines™ DxCG Risk Solutions suite is the input data required for processing. Both the Medical Risk Solutions and Pharmacy Risk Solutions modules are used to quantify clinical risk, and clients often ask when each should be used.

In this white paper, we will discuss the distinct use cases and technical characteristics that guide the selection of the Medical versus Pharmacy Risk Solution modules.

Likelihood of Hospitalization (LOH)

In order to improve health quality while reducing the cost of care, employers and Care Management Organizations need to prospectively identify the small proportion of individuals who account for a large share of future expenditures. Moreover, it is important to identify those individuals in this population whose costs are “actionable”, i.e., people for whom specific care opportunities exist prior to a major medical event that—if addressed—could help to avoid the event or reduce its severity.

With this in mind, we developed a model that predicts the likelihood of future hospitalization rather than future costs. While every hospitalization may not be preventable, such a model has a better potential for producing a good list of “actionable” candidates for a care management program.

The focus of the Likelihood of Hospitalization (LOH) model is to identify, at the population level, a small subset of individuals who are likely to be hospitalized in the near future. Using demographic and administrative claims data from inpatient and outpatient settings, the model predicts future admissions in a general population.

Verisk Health DxCG Medical Classification System – Version 7

The DxCG Medical Classification System was initially developed to predict the resources needed to treat patients so that a provider or health plan could be fairly compensated for the expected risk of managing a person or group. The clinical and statistical validity have been established and its uses have grown to include predictions of many diverse measures and outcomes for a wide range of business applications.

This white paper documents the seventh revision of the DxCG Medical Classification System.

Performance of Risk Adjustment and Predictive Models

There are two fundamental categories of models used to quantify clinical risk and cost: risk adjustment models and prediction models. While the use cases for each model category (and the technical characteristics) are very distinct, there is often confusion about when to use each model – and when not to.

In this white paper, we first describe the distinct use cases, technical characteristics, and commonly made mistakes in selecting risk adjustment versus prediction models. We then review a quantitative assessment of each model’s performance.

Estimating the Impact of Disease and Treatment on Life Expectancy in Abu Dhabi

Health status and life expectancy reflect a nation’s demographic, socioeconomic, and public health conditions. One of the ultimate goals of economic and social development is to achieve better health outcomes and improve overall population welfare. Equipped with proper tools and measures, a forward-thinking government may identify populations at risk, allocate resources appropriately, and design health and wellness intervention programs to achieve better outcomes.

In 2009 the Health Authority–Abu Dhabi (HAAD) set out to improve quality of care and overall population health. The agency wanted to understand and predict future costs, disease progression and the life expectancy of its residents. In this study we outline a set of innovative predictive models developed by Verisk Health for the HAAD that project medical and mortality risk simultaneously several years into the future.

Using Prospective Risk Score for Budgeting

DxCG methodology is the industry standard for understanding healthcare risk. Clients frequently ask how to use the Sightlines DxCG Risk Solutions suite of predictive models to better respond to business challenges. In this white paper, we will discuss how to use the DxCG prospective models for budgeting purposes.