Lifecourse investigation of the cumulative impact of adversity on cognitive function in old age and the mediating role of mental health: longitudinal birth cohort study

Objective To investigate the accumulation of adversities (duration of exposure to any, economic, psychosocial) across the lifecourse (birth to 63 years) on cognitive function in older age, and the mediating role of mental health. Design National birth cohort study. Setting Great Britain. Participants 5362 singleton births within marriage in England, Wales and Scotland born within 1 week of March 1946, of which 2131 completed at least 1 cognitive assessment. Main outcome measures Cognitive assessments included the Addenbrooke’s Cognitive Examination-III, as a measure of cognitive state, processing speed (timed-letter search task), and verbal memory (word learning task) at 69 years. Scores were standardised to the analytical sample. Mental health at 60–64 years was assessed using the 28-item General Health Questionnaire, with scores standardised to the analytical sample. Results After adjusting for sex, increased duration of exposure to any adversity was associated with decreased performance on cognitive state (β=−0.39; 95% CI −0.59 to –0.20) and verbal memory (β=−0.45; 95% CI −0.63 to –0.27) at 69 years, although these effects were attenuated after adjusting for further covariates (childhood cognition and emotional problems, educational attainment). Analyses by type of adversity revealed stronger associations from economic adversity to verbal memory (β=−0.54; 95% CI −0.70 to –0.39), with a small effect remaining even after adjusting for all covariates (β=−0.18; 95% CI −0.32 to –0.03), and weaker associations from psychosocial adversity. Causal mediation analyses found that mental health mediated all associations between duration of exposure to adversity (any, economic, psychosocial) and cognitive function, with around 15% of the total effect of economic adversity on verbal memory attributable to mental health. Conclusions Improving mental health among older adults has the potential to reduce cognitive impairments, as well as mitigate against some of the effect of lifecourse accumulation of adversity on cognitive performance in older age.


Causal mediation analysis: Statistical equations
Defining total effect, average direct effect (ADE) and average causal mediation effect (ACME) (adapted from Imai et al., 2011): Let  = exposure,  = mediator, and  = outcome.The ACME can be defined using the below formula: Where ACME equals the difference in outcome due to the change in mediator from the value expected under no exposure (M (t0)) to the value expected under exposure (M (t1)), holding the exposure t at a constant.Fixing the exposure and changing the mediator allows the elimination of other mechanisms and isolates the mediation effect via the mediator.
The ADE can be defined using the following: Where ADE equals the difference in outcome due to a change in the exposure from not exposed ( 0 ) to exposed ( 1 ), holding the mediator at a constant that would be expected under exposure (M (t)).This represents the effect of exposure on the outcome that is not hypothesised to be transmitted via the mediator.The total effect equals the sum of the ACME and ADE.BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s)

Sensitivity analyses
Repeating steps from main analysis for a subset of participants with complete data on all three cognitive measures (N=1720) Table S6.Linear regression models showing the association between duration of exposure to adversity (any, economic, psychosocial) across the lifecourse and cognitive function at 69 years (sensitivity analyses).
Keele L, Tingley D, Yamamoto T. Unpacking the Black Box of Causality: Learning about Causal Mechanisms from Experimental and Observational Studies.Am Polit Sci Rev 2011; 105: 765-89.

Table S2 .
Linear regression models showing the association between duration of exposure to adversity (any, economic, psychosocial) across the lifecourse and cognitive function at 69 years.
a Adjusted for sex b Adjusted additionally for educational attainment and childhood (cognition and emotional problems) covariates.

Table S3 .
Linear regression models showing the association between duration of exposure to adversity (any, economic, psychosocial) across the lifecourse and mental health (Psychological distress), and between mental health (psychological distress) and cognitive function at 69 years.

Table S4 .
Causal mediation analysis showing the average direct, mediated (via mental health), and total effect of duration of exposure to adversity (any, economic, psychosocial) on cognitive function at 69 years.

Table S5 .
Linear regression models on the interaction between economic and psychosocial adversity on verbal memory at 69 years.

Duration of exposure across the lifecourse to:
a Adjusted for sex b Adjusted additionally for educational attainment and childhood (cognition and emotional problems) covariates.

Table S7 .
Causal mediation analysis showing the average direct, mediated (via mental health), and total effect of duration of exposure to adversity (any, economic, psychosocial) on cognitive function at 69 years.Adjusted for socio-demographic (sex, educational attainment) and childhood (cognition and emotional problems) covariates.Abbreviations: ADE: average direct effect; ACME: average causal mediation effect.BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s)