Subscale validity of the Mattis Dementia Rating Scale

Abstract

The Mattis Dementia Rating Scale (DRS), a widely used dementia screening instrument, generates five subscale scores in the areas of Attention (ATT), Initiation-Perseveration (IP), Construction (CN), Conceptualization (CON), and Memory (MEM). The present study sought to determine the criterion validity of the DRS subscales in a sample of 50 patients with Alzheimer's disease (25 with mild and 25 with moderate dementia). Subject performance on the five DRS subscales was correlated with performance on five well-validated neuropsychological criterion measures using Pearson r and stepwise regression. On a univariate level, each DRS subscale correlated most strongly with its assigned neuropsychological criterion measure. On a multivariate level, each DRS subscale emerged as the Step 1 predictor of its assigned criterion measure, with the exception of DRS CN, which was the Step 2 predictor. The results suggested that overall the DRS subscales are valid measures of their respective constructs and have value for both clinical and research purposes in mild and moderate dementia.

Construct and Concurrent Validity of the Dementia Rating Scale-2 Alternate Form

“…Regarding the subscale validity of the DRS-2: AF, convergent and discriminant correlational analyses conducted between four DRS-2: AF subscales and four criterion measures provide robust support for the subscales' construct validity in our communitydwelling sample of older adults without a reported diagnosis of dementia. Furthermore, the results are similar to the convergent validity studies conducted by Brown et al (1999) and Marson, Dymek, Duke and Harrell (1997) with the original DRS. It should be mentioned again, however, the Construction subscale was not included in these analyses due to its negatively skewed distribution; almost all of our participants received full credit on this subscale.…”

“Vascular Depression” Predicts Verbal Fluency in Older Adults

“…Previous studies have found that the DRS demonstrates high levels of reliability (e.g., split-half reliability = .90; Gardner, Oliver-Munoz, Fisher, & Empting, 1981) and validity in distinguishing among normal elders and those with varying levels of dementia (Vitaliano et al, 1984). The subscales of the DRS have been found to correlate highly with other well-established corresponding criterion measures (Marson, Dymek, Duke, & Harrell, 1997). DRS subscales have been found to correlate with selective regional brain volumes in Alzheimer's disease (Fama et al, 1997).…”

Utility of the Mattis dementia rating scale to assess the efficacy of rivastigmine in dementia associated with Parkinson’s disease

“…Moreover, the motor requirement of some items means that they are poorly suited to patients with movement disorders. The Mattis DRS was not designed with reference to the clinical description of a particular dementia, and moreover has the advantage of being able to investigate five cognitive domains whose construct validity has been demonstrated [14,18,23]. Notably, executive functions (the impairment of which greatly contributes to the development of PDD) are assessed.…”

“…This scale was selected because it is one of the most commonly used scales in patients with parkinsonism: it is sensitive [3] and comprises items which were specifically designed for executive function assessment [18]. The scale's sensitivity to treatment was also compared with the more usual outcome measures.…”