Lifetime diet and cognition in older people.

Dietary intake may impact upon the trajectory of older-age cognitive change and decline via nutritional mechanisms that contribute to brain health and functioning, and to the risk of chronic diseases associated with poorer late life cognitive outcomes. Diet is a modifiable environmental exposure. As such, it provides an avenue for intervention to promote better cognitive functioning and delay or prevent cognitive impairment and dementia that are placing an increasing burden on older individuals, their families and the health system. The majority of studies that have investigated the nutritional determinants of healthy cognitive ageing have done so within populations older than 65, but the long-term aetiology of cognitive change extends years and even decades prior to the onset of noticeable decline. Thus, the salience of diet to an individual’s cognitive ageing trajectory is likely to extend back into the life-periods prior to older age. Even the longest prospective studies do not have dietary records extending over the lifetime; therefore, determining the relationship, if any, between earlier-life diet and later-life cognitive health requires an alternate approach to gathering lifetime dietary data. The objective of this thesis was to develop a retrospective dietary reporting instrument to measure intake from multiple life-periods, then to investigate cross-sectional and longitudinal associations between lifetime diet and cognitive performance in cognitively healthy older-adults. Studies 1 to 3 investigated the reliability and validity of the Lifetime Diet Questionnaire (LDQ); the dietary assessment instrument developed in the context of the thesis to measure past diet. Study 1 was a preliminary study of dietary recall using the foods and frequencies of the LDQ. The strength of associations was tested between young adults’ (n=203) recall of earlier adolescent diet, and one or more family members’ recall of the same individual’s diet over the same period. Study 2 assessed the test-retest reliability of LDQ’s five life-periods in older adults (n=51). Both measures of reliability fell within acceptable limits. In Study 1, the average association between family members recall of an individual’s past intake was 0.73, while in Study 2, the average test-retest reliability of the questionnaire across all life-periods in an older sample was 0.81. Study 3 (n=352) recruited participants from the EPOCH trial (a randomised controlled trial of Omega-3 fish oil on older-age cognitive change). The validity of long-term dietary recall was investigated by testing the associations between lifetime dietary patterns extracted from the LDQ, and the EPOCH participants’ demographic and cardiovascular health variables. Lifetime dietary patterns were related to the demographic variables of age, sex, education, income, parental background, and childhood physical activity; patterns from childhood and adulthood also predicted cardiac medication use and cholesterol level in older age. Studies 4 and 5 used the same cohort to examine the relationships between LDQ dietary patterns and cross-sectional cognitive performance (Study 4) and 18-month cognitive change over 4 time points (Study 5). After controlling for relevant covariates and current dietary intake, all dietary patterns from the childhood period predicted baseline level of cognitive performance, and a ‘non-traditional Australian’ pattern in middle age predicted 18-month cognitive change. These preliminary findings have implications for the relevance of diet as a lifetime determinant of older-age cognitive health.Thesis (Ph.D.) -- University of Adelaide, School of Psychology, 201