Dr. Young's LabThe Impact of Education on Language Test Performance in Cortical and Subcortical Dementia
STATEMENT OF PROBLEM: Level of education has been shown to affect performance on neuropsychological tests of language. Individuals with educations of 12 years and under generally have not been adequately represented in normative samples found in the literature, thereby complicating language assessment in these individuals. This study tested two different hypotheses that explored the impact of lower education on Boston Naming Test (BNT; Kaplan et al, 1993) and Controlled Oral Word Association Test (COW A T; Borkowski et ai, 1967), consisting of tests of Semantic and Phonemic Fluency, performance in individuals with and without dementia. Hypotheses tested were the following: (1a) after controlling for age and dementia, lower education will result in lower scores on the BNT and tests of Semantic Fluency and Phonemic Fluency; (1b) after controlling for age and education, differences in performance on tests of language exist among four groups: cortical dementia, subcortical dementia, mixed cortical and subcortical dementia, and healthy adults; and (2) individuals with cortical dementia as perform better on tests of Phonemic Fluency than on tests of Semantic Fluency and those with subcortical dementia demonstrate an opposite pattern of performance. In addition, normative statistics for individuals ,with 12 years of education or less in the different groups were generated.
METHODS: Dementia groups were drawn from a large archival data set of outpatients from several Chicago-area hospitals. A no-dementia group was also recruited. A brief screening interview, the Mini Mental State Exam (MMSE: Folstein et al. 1975), the COWAT, and the BNT were administered to no-dementia participants.
RESULTS: There was a total of 388 participants. Multiple regression confirmed hypotheses that language test performance \vas affected by level of education. No- dementia participants achieved higher BNT and COW A T scores than the dementia participants. Lower education accounted for lower scores on BNT and the test of Phonemic Fluency, but lower education did not affect scores on the Semantic Fluency test. Individuals with cortical dementias performed significantly better on the test of Phonemic Fluency than on Semantic Fluency, and those with Subcortical dementia showed opposite performance.
CONCLUSIONS: Because those with lower education tend to score lower on language test performance, comparing their scores to norms derived from higher education groups could lead to a misdiagnosis of language deficits associated with dementia.