Depression, anxiety and cognition in community-dwelling adults aged 70 years and over.

Although there is evidence of associations between anxiety, depression and cognitive function in old age, there is little work investigating relations between those variables over an extended period of time. Therefore, we used data from the Canberra Longitudinal Study to investigate 12-year cognitive change over four measurement points in relation to anxiety and depression symptoms. Latent growth models on over 836 community-dwelling persons aged 70 years and over, recruited from the electoral roll suggested that higher depression symptom scores were associated with poorer initial performance in processing speed, verbal fluency and episodic memory while higher anxiety symptom scores were associated with verbal fluency. We found no evidence that mental health variables affected change in cognition over time. Importantly, when possible mild cognitive impairment or dementia cases were removed from the models, associations between the cognitive variables and depression symptoms disappeared while those for anxiety symptoms strengthened. The findings are consistent with the possibility that depression-related cognitive deficits represent a prodrome or risk factor for dementia while associations between anxiety and cognition may be more characteristic of normal aging.