Late‐life depression, allostatic load, and risk of dementia: The AGES‐Reykjavik study.

Background: The biological underpinnings of the relationship between depression and dementia are currently not completely understood. Previous research has highlighted the role of vascular, metabolic, inflammatory, and stress markers in the role of both depression and dementia. Allostatic load has been an all‐encompassing term to describe these biological factors when investigating subgroups in individuals based on health characteristics. The current study aimed to elucidate if depression subtypes exist based on modifiable allostatic load factors using latent class analysis (LCA), and whether these subtypes have differing risk on incident all‐cause dementia, Alzheimer's disease (AD), and non‐AD diagnosis using Cox regression analyses. Method: The study included 5343 individuals without dementia at baseline from the AGES‐Reykjavik Study. Vascular factors (systolic blood pressure and pulse pressure), cholesterol (high‐density lipoprotein and low‐density lipoprotein), metabolic factors (abdominal circumference, fasting glucose, and triglycerides), inflammation (c‐reactive protein), and stress factors (salivary morning and evening cortisol) were used to indicate allostatic load. Both allostatic load indicators and depressive symptoms (Geriatric Depression Scale‐15) were assessed at baseline. Incident dementia was assessed after 12‐years follow‐up. Allostatic load indicators as well as each item on the GDS‐15 were used in the LCA. Result: LCA revealed that allostatic load factors, more than the depression questionnaire, grouped the individuals in this cohort. We identified six classes, named based on their most identifying characteristics: 'Average' (34% prevalence), 'Cardiovascular + metabolic risk' (27% prevalence), 'Slim' (16% prevalence), 'Depressed' (10% prevalence), 'Hypertension' (10% prevalence), and 'Diabetic' (4% prevalence). Cox regression analyses found an increased risk for incident all‐cause dementia in the 'Slim', 'Depressed', and 'Diabetic' groups. The 'Slim' group had a specific risk for AD, and the 'Depressed' group for non‐AD dementias. Conclusion: The findings suggest subgroups exist in older individuals based on allostatic load factors, with differing risk for incident dementia. This highlights the importance of early interventions in diabetes and depression and in monitoring for dementia symptoms. [ABSTRACT FROM AUTHOR]