Measuring Dementia Symptoms: Using the Mini-Mental State Examination as a Surrogate for the Clinical Dementia Rating Scale when Evaluating Dementia Interventions

Abstract

Background: Recently, a number of non-pharmacological interventions for dementia symptoms were evaluated using Cost-Benefit Analysis and found to be both effective and socially worthwhile. In these studies, dementia symptoms were measured by the Clinical Dementia Rating (CDR) scale. Although the CDR is the most comprehensive and reliable instrument for measuring dementia symptoms, common practice is to use the Mini-Mental State Examination (MMSE) as a substitute. The objective of this paper is to examine the relationship between the two measures, using a large national data set, in order that one can meaningfully convert evaluations using the MMSE into the more comprehensive CDR version.

Methods: The main analysis involves using regression models to estimate the relationship between the two dementia symptom methods in a large, national, panel data set. Four regression specifications are presented according to which controls are selected. The subsequent analysis entails looking at a case study of a particular dementia intervention, that of vision correction, to see whether in practice the MMSE and CDR give similar CBA decisions results when data on both measures were actually available.

Results: We find that multiplying a MMSE outcome measure by 0.4 will provide a valid approximation for the CDR when the latter measure is not available. This approximation was reasonably robust to a whole range of controls that were included. The case study of vision correction provided further corroborative evidence that the MMSE can be a practical surrogate for the CDR.

Conclusions: The CDR can be approximated by the MMSE when carrying out a CBA of any dementia symptoms intervention. Benefits estimated using the CDR can readily be transferable to become the benefits using the MMSE.