Risk of incident dementia varies with different onset and courses of depression.

Background: This study aims to examine if risk of dementia differs between adult- and late-onset depression.

Methods: 16,608 community-living dementia-free older adults were followed for 6 years to the outcome of incident dementia. Depression was diagnosed according to international diagnostic guidelines. Depression in adulthood or late life was categorized using age 65 as cutoff. Hazard ratio for dementia was estimated using Cox regression analysis.

Results: People with depression in adulthood only did not have higher dementia incidence, suggesting those in remission from adult-onset depression are not at greater risk of dementia. Conversely, having depression in both adulthood and late life was associated with higher dementia risk, and improvement in depression in late life was associated with lower risk, suggesting persistent or recurrent lifetime depression is a risk factor for dementia. Those with depression in late life only were not associated with higher dementia risk after controlling for the longitudinal changes in depressive symptoms, consistent with late-onset depression being a prodrome of dementia.

Limitations: Reverse causation is a potential limitation. This was minimized by careful ascertainment of depression and dementia cases, exclusion of individuals with suspected dementia at baseline and those who developed dementia within 3 years after baseline, and controlling for various important confounders.

Conclusions: Risk of incident dementia varies with presence and resolution of depression at different ages. Further studies are needed to test whether treating adult-onset depression may prevent dementia. Older adults with a history of depression present for an extended time should be monitored for cognitive decline.