Background Depressive symptoms and cognitive decline often co-occur among older adults, and they share several mechanisms. Despite the fact that cognitive dysfunction has been linked to increased depressive symptoms, the directionality of this association remains unclear. This study aims to examine whether there is a bidirectional relationship between depressive symptoms and cognitive function in middle-aged and older English adults over a 16-year follow-up period in the English Longitudinal Study of Ageing (ELSA).

Methods Among the original ELSA sample of 11,050 older adults, a nationally representative sample of community-dwelling English older adults, 7,745 participants with relevant data who were free of dementia and stroke were included in the current analysis. These participants were examined every other year from 2002 up to wave 9 (2018-2019), resulting in a follow-up period of up to 16 years. Cognitive measures include objective tasks of memory and verbal fluency (animal naming), while depressive symptoms were evaluated by the Center for Epidemiologic Studies Depression Scale. We used bivariate dual change score models to estimate the multivariate associations between depressive symptoms and cognitive function, used interchangeably as exposures and outcomes over 16 years of follow-up.

Methods Higher depressive symptoms were cross-sectionally associated with poorer memory (β intercept  = -0.026, standard error (SE ) =  0.004, P ≤ 0.001) and verbal fluency (β intercept = −0.012, SE = 0.005, P =0.019) at baseline. Longitudinally, greater baseline depressive symptoms were also associated with faster memory loss (β linear slope=−0.129, SE =0.014, P ≤ 0.001) over time but not with a change in verbal fluency. While baseline performance on memory or verbal fluency did not predict a change in depressive symptoms, a steeper change in depressive symptoms was related to faster memory decline (β intercept  =-0.258, SE = 0.083, P = 0.002) and a greater memory change was associated with a faster change in depressive symptoms (β intercept =0.022, SE = 0.008, P = 0.009).

Conclusion Greater depressive symptoms were associated with poorer baseline memory and faster memory loss over time. A gradual change in depressive symptoms contributed to a faster memory decline and vice versa, suggesting that psychological mood and cognitive performance are intrinsically interrelated. Our results emphasise the connection between psychological mood and cognitive performance, underscoring the importance of holistic wellbeing. Healthcare systems should adopt integrated approaches that address both mental health and cognitive function. Public health campaigns can raise awareness about the link between depressive symptoms and cognitive decline. Educating individuals about the importance of mental health maintenance and regular cognitive assessments can empower them to take proactive steps towards preserving cognitive function.