Statement of the Problem: Language impairment is a hallmark characteristic of dementia in Alzheimer’s disease (AD) and is a critical indicator for monitoring its progression (Snowdon et al., 1996). Accurate evaluation of language is crucial because it comprises essential diagnostic criteria for the disease (DSM-IV, 1994; McKhann et al., 1984) and it distinguishes between language difference and language disorder in culturally and linguistically diverse populations. This is especially important because the rate of newly acquired AD is significantly higher in African Americans relative to Caucasian individuals at similar ages (Tang et al., 1998; 2001). Also, Green et al., (2002) found that first-degree relatives of African Americans with AD have a higher cumulative risk of dementia than do Caucasians with AD. Finally, the rapidly growing percentage of non-white elderly in the United States (Census, 2000) makes it imperative that cultural and linguistic factors are considered when evaluating for AD.

The results of studies comparing the cognitive abilities (e.g., memory, attention, language) of persons with dementia across different ethnic groups are inconsistent. Some studies have shown African American persons with dementia to perform poorly on standardized measures compared to whites (Fillenbaum et al., 1990; Gurland et al., 1992; Teresi et al., 1995; Welsh et al., 1995), except when controlling for education (Murden et al., 1991) and when performing higher-level language functioning tasks of verbal reasoning (Murden, 1997; Ripich et al., 1997). Evidence in the literature suggests that there are differences in performance on language tasks for persons with dementia when ethnicity and race are considered.

One problem with previous studies of cognitive abilities in culturally diverse populations with dementia is that language has been typically measured through isolated tasks, such as naming, verbal fluency, and repetition. Such tasks may not provide a true representation of an individual’s linguistic abilities. Instead, language should be evaluated comprehensively in order to accurately assess cognitive-linguistic skills and should also include indices of higher-level language functioning, like reasoning, problem solving, and judgment.

Another problem is that cognitive-linguistic measures for persons with dementia, namely the Arizona Battery of Communication Disorders in Dementia (ABCD; Bayles & Tomoeda, 1993), only included whites in the standardization sample. Speech-language pathologists commonly use the ABCD despite mandates that assessment instruments be employed with individuals for whom they have been normed, including culturally diverse populations that differ in “ethnic composition and social background [,]”(NIH, 1987). More importantly, it is the only comprehensive cognitive-linguistic measure designed to assess mild–moderate dementia available. Since non-white persons with dementia were not included in the standardization of the instrument, it may not be valid for culturally and linguistically diverse populations, including African Americans, who are the largest ethnic minority in the U.S. (Census, 2000).

Study Aims: The proposed study will test for differences in performance of African-American and Caucasian adults with dementia of the Alzheimer’s type (DAT) using comprehensive cognitive-linguistic measures. Therefore, the specific aims for the proposed exploratory study are to determine the extent to which differences in performance between African-American and Caucasian adults with DAT are present by using two measures:

1. Cognitive-linguistic measures using the ABCD;
2. Higher-level language measures using the Test of Language Competence-Expanded (TLC-E; Wiig & Secord, 1989).

The outcome of the proposed study will have direct implications for cognitive-linguistic evaluation of African Americans with dementia.