Risk‐adjustment is designed to predict healthcare costs to align capitated payments with an individual's expected healthcare costs. This can have the consequence of reducing overpayments and incentives to under treat or reject high cost individuals. This paper seeks to review recent studies presenting risk‐adjustment models.
This paper presents a brief discussion of two commonly reported statistics used for evaluating the accuracy of risk adjustment models and concludes with recommendations for increasing the predictive accuracy and usefulness of risk‐adjustment models in the context of predicting future healthcare costs.
Over the last decade, many advances in risk‐adjustment methodology have been made. There has been a focus on the part of researchers to transition away from including only demographic data in their risk‐adjustment models to incorporating patient data that are more predictive of healthcare costs. This transition has resulted in more accurate risk‐adjustment models and models that can better identify high cost patients with chronic medical conditions.
The paper shows that the transition has resulted in more accurate risk‐adjustment models and models that can better identify high cost patients with chronic medical conditions.
