Measuring risk-adjustment is crucial for avoiding risk selection incentives. Up to now, regression models have prevailed over categorical ones. However, such difference is often misunderstood or forgotten. A new article explains with all the details the comparison between both approaches.
The summary:
Regression and clinical categorical models represent very distinct approaches to risk adjustment. Users must carefully choose the model that best suites the intended application. Although clinical categorical models have many advantages in terms of communication, transparency, and stability, their initial development requires a significant effort and clinical input. Regression models usually require less initial development effort but are unstable in a changing environment and fail to provide the same degree of communication value and transparencyGreat work by Fuller et al. Though I fully support the categorical approach, my impression is that beyond such options, there are also alternatives that may fit better with morbidity data: mixed models (grade of membership). The following book explains the details (chap 17).
PS. R package
PS. Nowadays, unfortunately our government has lost its way regarding the design of the appropriate incentives in healthcare payment systems. The impact in the efficiency is huge, but nobody cares about it. There is a current effort to lie systematically in our post-truth era.
