An introduction to the why and how of risk adjustment

Abstract

Department of Veterans Affairs (VA) health services researchers often adjust for the differing risk profiles of selected patient populations for a variety of purposes. This paper explains the major reasons to conduct risk adjustment and provides a high level overview of what risk adjustment actually does and how the results of risk adjustment can be used in different ways for different purposes. The paper also discusses choosing a diagnostic classification system and describes some of the systems commonly used in risk adjustment along with comorbidity/severity indices and individual disease taxonomies. The factors influencing the choice of diagnostic classification systems and other commonly used risk adjustors are also presented along with a discussion of data requirements. Statistical approaches to risk adjustment are also briefly discussed. The paper concludes with some recommendations concerning risk adjustment that should be considering when developing research proposals.

KEYWORDS:

• Risk adjustment
• case mix
• outcomes
• performance
• veterans affairs

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 This discussion concentrates on the mean as the parameter of interest for purposes of exposition. For broad comparisons across providers based on overall performance, a focus on the mean is appropriate. For other applications, however, such as premium prediction and pricing, the entire distribution needs to be modelled correctly to produce unbiased predictions for individual observations.

2 This focus is appropriate for performance improvement tools that rank medical centers by their performance, such as the VA’s Strategic Analytics for Improvement and Learning (SAIL) value model (https://www.va.gov/QUALITYOFCARE/measure-up/Strategic_Analytics_for_Improvement_and_Learning_SAIL.asp).

3 Recommendations concerning race and ethnicity and socioeconomic status as risk adjustors are evolving as evidenced by a recent National Quality Forum report [34] on this topic. Researchers should check on current recommendation before including these factors as risk adjustors.

Additional information

Funding

This work was supported by U.S. Department of Veterans Affairs [grant number SDR 13-426].

Notes on contributors

W. Bruce Vogel

Dr. W. Bruce Vogel, is presently an Associate Professor in the Department of Health Outcomes & Biomedical Informatics in the College of Medicine at the University of Florida where he also serves as Director of the Division of Health Outcomes. An economist by training, he has extensive experience in a wide variety of health services research studies, including work related to Medicaid managed care, risk adjustment, and cost-effectiveness. In addition, he has served as a Research Health Scientist at the VA Center for Innovation in Disability and Rehabilitation Research where he also served as Director of the Methodology Core. His research has appeared in Medical Care, Health Services Research, Health Care Financing Review, Journal of Gerontology, Advances in Health Economics and Health Services Research, and other scholarly publications.

Guoqing John Chen

Guoqing John Chen, M.D., Ph.D., M.P.H., received his medical degree from the Tongji Medical University, Wuhan, China, and Ph.D./M.P.H. from the University of North Carolina at Chapel Hill, North Carolina. He completed a gerontology fellowship at Wake Forest University School of Medicine, Winston-Salem, North Carolina. Dr Chen joined the faculty at the Kansas University Medical Center in October 2013 and also serves as Director in the Division of Health Services Research, Department of Internal Medicine. His research interests are in the areas of quality improvement, mobile-based care delivery, outcomes and health services research, and economic evaluation in medicine. He has expertise in use of EHR and large claim-based datasets, e.g. Medicare, Medicaid, and VA data, and has extensive experience in conducting randomized clinical trials for quality improvement and comparative effectiveness. He has led multidisciplinary research teams with a track record of collaborating with clinician and non-clinician investigators in studies of outcomes, quality improvement, and economic evaluation in medicine. He has been as PI or Co-Investigator on several federal-funded projects with over 60 peer-reviewed publications.