Caffeine Consumption and Cognitive Function at Age 70: The Lothian Birth Cohort 1936 Study
Objective: To investigate the association between caffeine consumption and cognitive outcomes in later life. Methods: Participants were 923 healthy adults aged about 70 years in the Lothian Birth Cohort 1936 Study, on whom there were IQ data from age 11 years. Cognitive function at age 70 was assessed using a battery of tests measuring general cognitive ability, speed of information processing, and memory. Current caffeine consumption (using multiple measures of tea, coffee and total dietary
... nd total dietary caffeine) was obtained by self-report questionnaire, and demographic and health information was collected in a standardized interview. Results: In age-and sex-adjusted models there were significant positive associations between total caffeine intake and general cognitive ability and memory. After additional adjustment for age 11 IQ and social class-individually and together-most of these associations became non-significant. A robust positive association, however, was found between drinking ground coffee (e.g. filter and espresso) and performance on the National Adult Reading Test (NART, p = .007), and the Wechsler Test of Adult Reading (WTAR, p = .02). No gender effects were observed, contrary to previous studies. Generally, higher cognitive scores were associated with coffee consumption, and lower cognitive scores with tea consumption, but these effects were not significant in the fully adjusted model. Conclusions: The present study is rare in having childhood IQ in a large sample of older people. The results suggest that the significant caffeine intake-cognitive ability associations are bidirectional-because childhood IQ and estimated prior IQ are associated with the type of caffeine intake in old age-and partly confounded by social class. Key words: caffeine, cognitive function, childhood IQ, aging. BMI = Body mass index; FFQ = Food Frequency Questionnaire; M = mean; MHT = Moray House Test; PCA = Principal Components Analysis; SD = standard deviation; SMS1947 = Scottish Mental Survey 1947; LBC1936 = Lothian Birth Cohort 1936; SES = socio-economic status general well-being, further research is needed to fully understand the nature of these associations and rule out chance confounding by other factors.