• depression
• Alzheimer's disease
• dementia
• psychogeriatrics

Hachem et al attempt to peer through the murky waters of the relationship between depression and dementia.1

Depression was identified as a potentially modifiable risk factor for dementia by the Lancet Commission in 2017.2 Opinion remains divided as to whether depression is a true risk factor for dementia, an early warning sign of emerging dementia or a result of shared disease processes. Hachem et al attempt to peer through the murky waters of the relationship between depression and dementia.

It has been difficult to try and tease apart the exact nature of this relationship because of the long evolution time of neurodegenerative pathologies before cognitive symptoms appear, the consequent requirement for long-term follow-up in studies representative of the general population;, need to differentiate clinical diagnoses of depression from self-reported low mood as well as the need for robust cognitive measures/clinical dementia diagnoses, and information on the full range of possible confounders.

Hachem et al had high-quality information on cognitive status and whilst they did not have clinical diagnoses of depression, they used a cut-off on the CES-D which would have excluded those with subthreshold depressive symptoms. They also accounted for a broad range of confounders, including APOE genotype, with only a history of significant head injury notably absent. The authors acknowledge that including antidepressant use in their identification of individuals with depression may have led to an overestimation of depression incidence, as antidepressants are regularly prescribed for other conditions, for example, pain. Their finding that depressive symptoms begin to rise 6–8 years before a dementia diagnosis, becoming more pronounced over time, argues against depression being a true risk factor for dementia.

Hachem et al contribute to our understanding of the complex relationship between depression and dementia. Earlier this year, Paris et al used a subset of UK Biobank data to formally examine whether depression is a causal risk factor for dementia.3 Although the information on depression in the UK Biobank is somewhat limited, the authors found that, once again, depressive symptoms increased in the 5 years preceding a dementia diagnosis. They concluded that there was evidence supporting both causation—possibly due to reduced cognitive reserve—and reverse causality.3

In contrast, other cross-sectional studies—such as my own work using human postmortem brain tissue from individuals with depression who had died by suicide—found no evidence of increased Alzheimer’s disease or vascular neuropathology compared with controls.4 A recent large consortium study similarly found no evidence that Aβ levels were elevated on PET imaging or in cerebrospinal fluid in those with depressive symptoms and either normal cognition or mild cognitive impairment.5 This may suggest that the persistence or late-life emergence of depressive symptoms is key to dementia risk.

Whilst each study contributes only a small piece of the overall picture, it is gradually becoming clear that the relationship between depression and dementia is more complex than it simply being a modifiable risk factor for dementia, like smoking. Fundamentally, however, increased clarity about this relationship is urgently required to enable clinicians to provide optimal advice and treatment to older adults with depression.