Individual differences in the timing of dementia have been attributed to cognitive reserve (CR), thought to reflect lifelong engagement in stimulating experiences, which provide resilience against brain pathology. In older adults, dementia and depression are closely related, and some studies have linked CR with depression risk in old age. It is unclear if different ways of operationalizing CR exhibit similar association with old-age depression. We examined the association of two measures of CR with depressive burden in older adults: activity-based CR, capturing engagement in stimulating activities using proxy variables, and residual-based CR, indicating residual variance in cognition, not explained by the brain status.

We used data on 354 adults aged 60+ from the Swedish National Study on Aging and Care in Kungsholmen, followed for 15 years. Residual-based reserve was computed from a regression predicting episodic memory with a brain-integrity index incorporating six structural neuroimaging markers (white- matter hyperintensities volume, whole-brain gray matter volume, hippocampal volume, lateral ventricular volume, lacunes, and perivascular spaces), age, and sex. Activity-based reserve incorporated education, work complexity, social network, and leisure activities. Depressive burden was captured over the follow- up with the Montgomery-Åsberg Depression Rating Scale and time until clinically relevant level of symptoms (>6) was modelled using Cox proportional hazard models.

Preliminary results indicate that, upon minimal adjustment (age, sex, brain integrity status), top tertiles (ref: bottom tertile) of both activity-based (HR: 0.77; 95% CI: 0.61-0.98) and residual-based CR (HR: 0.62; 95% CI: 0.44-0.98) were associated with a lower risk of depressive burden onset over 15 years. Upon further adjustment for anthropometrics, health behaviors, and chronic disease burden, the association of activity-based CR was attenuated, whereas residual-based CR preserved its effect on depressive burden (HR [fully adjusted model]: 0.59; 95% CI: 0.40-0.88). Next steps include evaluating the ability of reserve measures to attenuate the association of brain integrity with depressive burden using interaction analysis.

Preliminary findings suggest that CR may be linked with depression development in older adults, although the association may vary depending on measurement of reserve. Association of activity- based reserve may be attributed to somatic disease pathways.