Assessment of the utility of cognitive screening instruments

There are a substantial and increasing variety of test instruments available to guide the clinician in making a diagnosis of dementia. An appreciation of the methods and outputs associated with test accuracy research is useful for all clinicians, not just academics. Test accuracy is best considered using a framework that clearly defines the index test, the gold standard (reference standard) used to define the condition of interest and the population in which testing will take place. By creation of a two by two table, cross classifying the results of the index test and the reference standard, we can derive various metrics describing the properties of the test. Test accuracy studies where the condition of interest is dementia present particular challenges. Using best practice statements in the conduct, reporting and assessment of study validity can assist the interpretation of test accuracy research papers and also for planning future studies. Techniques for systematic review and meta-analysis of test accuracy studies have been developed and are being applied to certain commonly used cognitive screening tests.