[Five by five test. A brief instrument for the detection of cognitive impediment in clinical settings]

Many psychological tests for the detection of cognitive impediment and dementia are inadequate by being too prolonged, inconsistent or with poor specificity or sensibility. It is presented a brief test of evocation of proper names by classes (TIG) and their properties psicometrics and utility are compared with the Mini-mental State Examination (MEM).

We selected 196 patients from a neurological clinic, with suggestive symptoms of cognitive impediment. We excluded individuals with focal or chronic cerebral disorders, sensorial disturbances, severe psychiatric diseases and toxicity by drugs. Patients were by grouped the Clinical Dementia Rating scale and they were evaluated neuropsychologically. The internal consistency of TIG was examined and a cross-validation procedure was performed to test for the predictive validity. Likewise, the convergent validity was tested against the scale of dementia of Blessed and the MEM. Finally the diagnostic efficiency of both tests, the effects of the demographics variables and the clinical utility were analyzed. We used non-parametrics statistics, analysis of sensibility upon curves ROC and analysis of the post-test probabilities.

The TIG has adequate internal consistency and validity. The diagnostic efficiency was high (area under the curve ROC 0,939 [IC 95%: 0,908-0,971], p < 0,0001). The introduction of the demographics covariates in a logistic model wisely improved the diagnostic efficiency. The TIG has some better fulfillment than the MEM. Neither the MEM nor the TIG are tests suitable for communitary screening of cognitive defect.

The TIG is briefer, easier to apply and more efficient than the MEM. Possibly it is useful to detect cognitive impediment in clinical contexts. This study is preliminary.