Stereotype threat in telephone‐administered neuropsychological assessment in older Veterans with and without cognitive impairment

Abstract

AbstractBackgroundNeuropsychological evaluation is ubiquitous in Alzheimer’s disease clinical care and research—from diagnosing cognitive deficits to evaluating treatment and outcomes. However, performance during evaluation can depend on factors other than one’s cognition, including stereotype threat. Stereotype threat occurs when people feel at risk of conforming to stereotypes about their social group, in this case, memory declines associated with aging. While stereotype threat in neuropsychological evaluation in healthy older adults has been established, we tested, for the first time, this construct in older adults with cognitive impairment.MethodTwenty older adults (range 63 – 89; mean age = 77.2, SD = 6.8) recruited from the Veteran’s Affairs Boston Healthcare system were administered a telephone Repeatable Battery of Neuropsychological Assessment (RBANS). Following baseline assessment, one‐week later participants received a stereotype threat condition prior to neuropsychological evaluation. Participants also received a reduced threat condition which de‐emphasized memory diagnosis (during a separate week, counterbalanced with the threat condition). In the fourth week and final session, supplemental measures were administered, including a telephone‐MoCA, demographics, and surveys measuring mood, dementia worry, beliefs about aging, and subjective memory complaints. Planned analyses include a mixed‐effects analysis of variance, with group (healthy; cognitively impaired) and condition (stereotype‐threat; threat‐reduced).ResultsOlder adults with cognitive impairment demonstrated lower performance across all domains of the RBANS, including list learning, story recall, fluency, digit span, delayed recall, and recognition, ηp 2 = .87‐.99. Older adults with impairment scored numerically lower but not statistically lower in list learning under stereotype threat (M = 17.4; SD = 6.6) compared to the reduced threat condition (M = 19.9; SD = 7.2). Confirmation of the analyses, including examination of covariates, will follow final subject recruitment (N = 28).ConclusionWe found that stereotype threat did not significantly impact performance of healthy older adults and older adults with cognitive impairment. We note that small, numerical differences emerged for list learning. Differences of just one point in neuropsychological evaluation can be impactful, especially with respect to norms and cutoff scores. As such, the results carry implications for neuropsychological evaluation in clinic practice and research broadly.