Melancholia in later life: late and early onset differences in presentation, course, and dementia risk. Academic Article uri icon

Overview

abstract

  • OBJECTIVES: Depression is a risk factor for cognitive decline and dementia. This risk may vary with age of onset and depression subtype. Late onset depression (LOD, 60 years and older) is associated with more cognitive decline, whereas early onset depression (EOD, before 60 years) is associated with more residual depressive symptoms. Potential differences may reflect divergent etiologies. These onset differences, however, have not been examined in the melancholic subtype of depression in older adults. METHODS: Data were obtained from the Neurocognitive Outcomes of Depression in the Elderly study. Participants (N = 284, 73% EOD-melancholic (EOD-M) and 27% LOD-melancholic (LOD-M)) were followed up over 3 years. Factor analyses examined differences in baseline depressive symptoms. Hierarchical linear growth curve models examined changes in depressive symptoms (Montgomery-Asberg Depression Rating Scale) and cognition (mini mental state examination). An annual clinical review panel assigned diagnoses of dementia. RESULTS: The LOD-M participants had more vegetative symptoms at baseline. LOD-M exhibited greater cognitive decline but fewer residual depressive symptoms than EOD-M. Among participants who remained in the study for at least 1 year, in uncontrolled analyses, a greater percentage of LOD-M compared with EOD-M developed dementia (23.0% vs. 7.8%). Whereas in logistic analyses, controlling for baseline demographics, age at onset remained a predictor of dementia, the odds ratio suggested that the effect was relatively small. CONCLUSIONS: The EOD-M and LOD-M participants have a different presentation and course. LOD-M may represent a syndrome of neuropsychiatric deterioration with expression of both depressive symptoms and cognitive decline.

publication date

  • February 12, 2014

Research

keywords

  • Dementia
  • Depressive Disorder

Identity

Scopus Document Identifier

  • 84905819067

Digital Object Identifier (DOI)

  • 10.1002/gps.4083

PubMed ID

  • 24677247

Additional Document Info

volume

  • 29

issue

  • 9