Objectives: There is an association between affective symptoms and cognition. However, the direction of this association remains unclear. This study aimed to test bidirectional relationships between affective symptoms and cognition from middle to late adulthood.
Method: Data were available from the MRC National Survey of Health and Development (NSHD), a prospective birth cohort of 5362 people born in 1946. Affective symptoms and cognition were measured at ages 53, 60-64, and 69. Latent scores of affective symptoms were derived and cross-lagged models were fitted for affective symptoms with verbal memory and processing speed.
Results: Results revealed an inverse cross-sectional association between affective symptoms and verbal memory (β=-0.18, SE=0.04, p<.001) and processing speed (β=-0.13, SE=0.06, p=.05) at age 53, but not at ages 60-64 or 69. Affective symptoms at age 53 predicted lower verbal memory at age 60-64 (β=-0.58, SE=0.27, p=.03), and affective symptoms at age 60-64 predicted lower verbal memory (β=-0.64, SE=0.29, p=.03) and processing speed (β=-1.27, SE=0.41, p=.002) at age 69. Verbal memory and processing speed did not predict subsequent affective symptoms.
Conclusion: Affective symptoms predict poorer verbal memory and processing speed over a period of 16 years, but not vice versa.
Keywords: Longitudinal; ageing; anxiety; cognition; depression; memory.