Diagnosing the early signs of the disease is a major challenge. There are several screening tests used for the diagnosis of dementia., but these tests are influenced by a patient s age and educational level, which justifies the need for research that validates such instruments conforming them to the cultural reality of the country . The present study aimed at, clinically validating a battery cognitive screening brief (BCSB) for early diagnosis of dementia in elderly adults in a geriatric clinic of a public hospital. Methods: 51 elderly adults with a clinical diagnosis of probable Alzheimer s disease ( mean age 81.3 (5.7) ; education 3.7 (2.8) ) and 123 older adults without dementia ( mean age 78.6(7.3) ; education 4.7( 3.5) ) were assessed. Test performance was compared between groups and data was analyzed through a quantitative method . The cut off points for each variable were generated by a ROC analysis . Results: There were no significant differences between groups with respect to education and gender, but there was a difference with respect to age (p equal 0,01). People with DA scored significantly lower on all cognitive tests ( p any less 0.05, for all tests ) . The group with AD showed deficits in the Lawton scale, both as self-reported by the informant (p any less 0.01 and p equal 0.01). The loss was not significant in the range of basic activities Katz (p more 0.01). The results of the ROC analyses indicated that measures of memory were the most sensitive measures to discriminate between individuals with AD and controls. Conclusion : The BCSB was accurate in identifying people with AD in a geriatric outpatient clinic at a federal hospital, in a sample with clinical and cognitive frailty.