P4-084: Interactive effects of family history of Alzheimer's disease and childhood learning problems on midlife memory performance

Abstract

Family history of AD in a first-degree relative increases risk of AD. Dementia risk may also be increased for persons with lower mental ability or less cognitive stimulation in childhood (Frisch et al., 2005; Whalley et al., 2000). We examined self-identified learning problems during childhood as a potential moderator of cognitive performance in a cohort of middle-aged persons with a parental family history of AD. Participants were 783 cognitively intact adult children (mean age = 52) of persons with AD and 282 control participants whose parents survived to at least age 70 without dementia. All were enrolled in the Wisconsin Registry for Alzheimer's Prevention. Participants were asked if they had any learning problems in school. Ten percent answered “yes” and were asked additional questions to determine the extent and severity of these difficulties; responses were combined to form a single index of childhood learning problems (LP). Neuropsychological testing, APOE genotyping, laboratory tests, and health and lifestyle questionnaires were also completed. Relationships between the LP index and each of five factor analytically derived cognitive dimensions were examined in regression analyses, with age, sex, education, and depressive symptoms as additional predictors. The LP index was a significant predictor (p < .05) of verbal ability, working memory, and cognitive speed/flexibility for both family history groups. By contrast, on the verbal memory factor (composed of learning and delayed recall scores on the Rey Auditory Verbal Learning Test), there was a significant LP × Family History interaction (p = .006), such that a history of learning problems was associated with reduced memory performance only among children of AD parents, not among controls who lacked a family history of AD. Although childhood learning difficulties were self-identified, they emerged as significant predictors of several dimensions of cognitive performance at midlife. More importantly, this aspect of personal history interacted with family history of dementia in predicting verbal memory performance. This outcome illustrates the importance of including family history of dementia in prospective studies of dementia risk to allow detection of potential interactions between familial risk and other factors. Longitudinal follow-up is under way.