Factors to Consider in Selecting a Cognitive Assessment Battery
There are numerous factors to consider in the selection of a cognitive measure or an assessment battery for an intervention trial. Of most importance are the measurement characteristics and the psychometric properties of the selected measures (see Chapter 14). Most of the available neuropsychological measures have been used with a wide variety of populations and have established norms, known psychometric properties, and sensitivity to change. Many are also available in languages other than English. Other important factors to consider when selecting these measures include the breadth of the battery with respect to domains assessed, number of measures, duration of the battery, frequency of assessments, and mode of administration as well as, of course, whether the assessment reflects the domain being targeted by the intervention. Each of these issues is discussed in turn.
Breadth of the Assessment Battery
In many intervention trials, the use of an extensive assessment battery that contains multiple measures of multiple domains is not needed. In some conditions where performance is highly intercorrelated across tests, a carefully selected, briefer assessment battery may provide the same amount of information as a much longer assessment (Keefe et al., 2004; Keefe, Poe, et al., 2006). A briefer battery may also be appropriate in situations where the intervention is targeted at a specific cognitive domain. Also, in some instances, a condition can be identified through the presence of a single salient deficit. For instance, diagnostic exclusion of possible Alzheimer’s disease can be accomplished through a very abbreviated assessment of delayed recall memory (Welsh, Butters, Hughes, Mohs, & Heyman, 1991), with this deficit leading to substantial separation from the performance of various other diagnostic groups, including even patients with schizophrenia (Davidson et al., 1996).
An advantage of a more detailed assessment is that the identification of multiple effects of a cognitive enhancing intervention is possible only with a wide-ranging assessment. Such an assessment would likely be undertaken in the early development phases of an intervention, as regulatory agencies require investigators to declare their primary outcome measure prior to the initiation of a trial (Buchanan et al., 2011) or an experimental intervention is being developed with no clear understanding of the breadth of its benefit. An additional use of a more detailed assessment in early phase studies is that of the detection of any possible adverse effects of a treatment. For example, if a cognitive remediation procedure is designed to improve problem solving but induces anxiety or distractibility, this could not be detected unless an assessment of all of these domains was employed. Regulatory agencies may require relatively comprehensive cognitive batteries so that any deleterious effects of a new treatment on cognition can be detected (Buchanan et al., 2011).