Background: A number of studies have indicated a beneficial effect of tea consumption on the reduction of risk of cognitive impairment and dementia in older aged populations. However, there is a paucity of data on these associations in the very old, defined as individuals aged 85 years and over. We investigated the relationship between tea consumption in the very old and measures of global cognitive function, memory, attention and psychomotor speed.
Method: Longitudinal (5-years), population-based cohort study of individuals aged 85+ years in the North East of England, United Kingdom. Participants were community-dwelling and institutionalized men and women recruited through general medical practices (n=676). Baseline tea consumption and longitudinal measures of global and domain specific (memory, speed and attention) cognitive function were assessed. Linear mixed models, controlling for demographic (e.g. age, sex and education) and health variables were used to determine whether tea consumption was protective against cognitive decline.
Results: Tea consumption was not associated with cognitive function at baseline on any measure (unadjusted and adjusted analyses). In the linear mixed effects models adjusted for age, sex, education and disease co-morbidity, higher tea consumption was associated with significantly better attention (focused and sustained attention), and psychomotor speed (complex tasks only) over five-years follow-up. However, there was no association between tea consumption and global cognitive function, memory or performance on simple speed tasks over time.
Conclusions: In this cohort study of non-demented very old adults we found that higher (vs. lower) tea consumption was associated with better performance over time on measures of focused and sustained attention and some psychomotor speed tasks. No associations with global cognition, memory or easy speed tasks (simple Reaction Time or Word Recognition) were detected. The results have implications for the development of possible diet-based interventions focused on improving cognitive function in the very old age group. These findings need to be confirmed in a sufficiently powered and well-designed RCT with non-demented very old adults.