RESUMO
OBJECTIVE: An increasing scientific literature recognizes that traditional cut-off scores for cognitive screeners may not be optimal for use in patients who differ in race/ethnicity from the screeners' normative/reference group. There is also literature on how racial/ethnic contextual factors, such as stereotype threat or perceived discrimination, may influence performance on cognitive testing. The current study examined the characteristics of SLUMS (a cognitive screening measure) performance in a large (n = 602) sample of Black (n = 229) and White (n = 373) veterans in a VA hospital located in the Southern United States. METHOD: SLUMS data were gathered from retrospective electronic chart review between January 2013 and February 2020. Race/ethnicity of veterans was gathered by chart review and race of hospital providers who administered the SLUMS by personal communication. RESULTS: Black veterans were 1.99 times more likely to be classified by total SLUMS score as being within the dementia range compared with White veterans. Differences in item level performance were only found between Black and White veterans with ≥ high school education. Race of clinical provider (i.e., Black or White) administering the SLUMS did not significantly impact veteran performance on the SLUMS. CONCLUSION: This is the first large sample study of differences in SLUMS performance between Black and White veterans. Findings replicate earlier research on Black and White performance differences on individual SLUMS items and provide an analysis of examiner-examinee racial discordance. This study underscores the importance of researching cognitive measures in groups who differ from the original normative/references samples.