Search results

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.

The aim of this paper is to reveal these problems and to derive recommendations for improvement. In the field of alcohol use disorders (AUDs), two common complaints are the large treatment gap that exists because only a small percentage of people with an AUD are in treatment; and the prolonged lag that typically exists between the emergence of problematic symptoms and actual on set of treatment. However, there also are no clear definitions for these terms – “treatment gap” and “treatment lag” – and, therefore, no consensus regarding how to quantify them. For this reason, it is difficult to compare the results of studies assessing either of these measures.

A non-systematic literature search and logical-analytical investigation was performed of immanent problems related to definitions and measurements aiming to enhance understanding in this area and derive suggestions for improvement.

The following four fundamental questions were identified: How does one operationalise the need to change substance use behaviours? Which interventions can justifiably be called treatment? Is treatment always necessary? and How regularly do patients need to be in contact with a treatment system to be considered “in treatment”? Potential approaches to answering these questions are discussed and recommendations made for future studies to determine how the treatment gap and treatment lag should be derived.

The derived recommendations should make the calculation of treatment gap and treatment lag more transparent and comparable between studies. They also may serve as checklists for future studies on the treatment gap and lag in the AUD field.