Cognitive ageing and late-life depression across cultures

Student thesis: Doctoral ThesisDoctor of Philosophy


This thesis examined the role of education and depression symptoms to late-life cognitive inequalities. The inquiry was conducted in several stages.
The first line of investigation examined whether a more highly educated recently born cohort had higher levels of cognitive performance and slower rates of cognitive decline in late life compared to a less highly educated earlier born cohort of the same age (65 to 75 years old at baseline). Data were used from the Longitudinal Ageing Study Amsterdam (LASA). Findings from Linear Mixed Models suggest that the later born cohort had higher baseline levels of processing speed, inductive reasoning and general cognitive performance, but showed similar or steeper rates of cognitive decline compared to the earlier born cohort. Education duration accounted for cohort differences in baseline levels of inductive reasoning and general cognitive performance, but it did not account for cohort differences in cognitive decline over 6 years of follow-up.
The second line of research examined which symptom dimensions of depression may show stronger associations with cognitive dysfunctions in late life. This research question was addressed in two separate psychometric studies, one conducted among older adults in Latin American countries (10/66 study) and the other conducted among older adults in the Netherlands (LASA study). Multiple Indicators Multiple Causes Models were used to examine cross-sectional associations between cognitive abilities and latent depression dimensions in the context of adjustment for differences in item response behaviour due to country of residence, age, gender and cognitive function levels. Findings from the 10/66 study suggest that poorer delayed recall performance was related to higher affective suffering symptoms and higher motivational symptoms. Motivational symptoms had a stronger negative association with verbal fluency performance than affective suffering symptoms. In both studies item response biases were of small magnitude and did not affect substantive conclusions.
The third line of investigation informs on the longitudinal direction of influence and timing of the association between cognitive abilities and depression symptoms (conceptualized as a unitary construct and as specific symptom dimensions). This inquiry was conducted in two separate studies: the 10/66 study with a follow-up duration of 3 years (using cross-lagged path analyses), and the LASA study with a follow-up duration of 13 years (using cross-domain latent growth curve models). Findings from the 10/66 study suggest that most prospective associations between cognitive abilities (i.e., verbal fluency, immediate recall and delayed recall) and overall depression symptoms / specific symptom-dimensions were bidirectional. Baseline motivational symptoms of depression were not related to follow-up verbal fluency performance, whereas baseline delayed recall performance was not related to follow-up affective suffering levels. Findings from the LASA study suggest that poor initial memory performance predicted an increase in overall depression symptoms and a specific increase in depressed affect over time, whereas processing speed decline was accompanied by an increase in overall depression symptoms and a specific increase in somatic symptoms.
In conclusion, findings from this thesis suggest that education alone has a limited role in accounting for cohort differences in cognitive ageing. The interplay between depression and cognitive functioning is complex. Whether depression co-occurs, precedes, accompanies, or follows cognitive dysfunctions may depend on the depression symptom dimensions experienced and on the timing of the assessments.
Date of Award2017
Original languageEnglish
Awarding Institution
  • King's College London
SupervisorMatthew Prina (Supervisor) & Martin Prince (Supervisor)

Cite this