Using the Resident Assessment Instrument-Mental Health (RAI-MH) to determine levels of care for individuals with serious mental illness

Abstract

This paper outlines the development of an algorithm to determine appropriate levels of care (LOC) for individuals with a serious mental illness (SMI). The algorithm, drew on several domains of the Resident Assessment Instrument-Mental Health (RAI-MH) to support a statistical model that would explain a maximum of variance with the gold standard, a consensus-based global rating of required LOC. The RAI-MH model explained 67.5% of the variance. The validity of the model was further examined by determining how the discrepancy between the current and predicted levels of care related to psychiatric outcomes. The results demonstrated that undersupported clients experienced significant negative psychiatric outcomes compared to clients receiving adequate care. Although the model based on the RAI-MH is not perfect, the results warrant further research to determine its usefulness in predicting required LOC.