Overview of Scales and Inventories

General Aspects

Mental status scales are brief, structured instruments with specific administration and scoring, and mental status inventories gather information by a list of questions. Some scales are combinations of mental status testing and inventories, for example, the Blessed Dementia Scale and the Alzheimer Disease Rating Scale. Mental status scales and inventories are either general instruments for cognitive deficits, or they are targeted instruments aimed at specific cognitive abilities, syndromes, or diseases. Inventories, as opposed to scales, tend to be targeted at behavioral disturbances. The purpose of general mental status scales is to distinguish patients with impaired versus normal cognition. General scales are sensitive for identifying screening for patients with the most common cognitive impairments, but they are not for diagnosing diseases or for brain-behavior localization. No mental status scale covers all areas of cognition; therefore they cannot be entirely comprehensive screens for all possible cognitive impairments. There are over 50 mental status scales in wide use. There is no single scale that is the “gold standard”, most have sufficient accuracy (sensitivity tempered by specificity) for dementia screening and share a number of advantages and deficiencies ( Tables 15.1 and 15.2 ).

General mental status scales vary in their coverage of cognitive areas. Most scales have items for memory and orientation, as these are affected early in Alzheimer disease and other dementias. This is followed by items for mental control/attention, language, visuospatial skills, and calculations. Few extend to examination of executive abilities, semantics, praxis, or socioemotional changes. This is evident in the content of the Mini-Mental State Examination (MMSE) and the Montreal Cognitive Assessment (MoCA), two of the most widely used scales in the clinical setting and used as illustrative examples in this chapter (also see Chapter 16 ). The MMSE items include 10 for orientation, 3 for registration, 3 for memory, 8 for language, 5 for mental control/attention, and 1 for visuospatial skills. In contrast, the MoCA items include 6 for orientation, 5 for memory, 6 for language, 6 for mental control/attention, 3 for visuospatial, and 4 for executive abilities (including clock hands) (author’s evaluation). Even within these content areas, there are differences in what is tested. For example, for language the MMSE, but not the MoCA, includes auditory comprehension, reading, and writing, whereas the MoCA, but not the MMSE includes verbal fluency. Of course, more than one cognitive domain may affect an individual item, but this comparison clearly shows the content variability that extends across mental status scales. Moreover, the number of items per cognitive domain, as well as the length of the instrument, can vary greatly.

The examiner should consider a number of other differences evident on both mental status scales and inventories. Although all aim for brevity, the administration times may vary between 5 minutes or less to 15–30 minutes or more. Scales may be constructed with relatively harder or easier items overall. For example, the MMSE is much easier than the MoCA, despite a suggested average adjustment for age and education. The order of presentation of items or inventory questions can make a difference in patient responses, for example, whether easy items are presented first, when patients are less fatigued, or presented last, when they distracted when items are less challenging. Orientation items are presented first in the MMSE and last in the MoCA. Memory performance on a scale may vary with the number of intervening items between registration and delayed recall. There is one intervening item on the MMSE, and there are six on the MoCA. On the MMSE, it is conceivable to still have the registration words in working memory when asked for them on delayed recall. Finally, mental status scales differ in their sensitivity to the effects of age, education, and sociocultural and language background, and there is little adjustment for these variables on these scales. Indeed, even slight differences in the complexity of the language used by scales and inventories may affect performance.