COGNITIVE VARIABILITY AND BRAIN AGING IN LATE-LIFE DEPRESSION

Abstract

Introduction Dispersion across cognitive test scores, a measure of variability in cognitive performance, may be an indication of pathological aging. The aim of the current study was to investigate the association of test dispersion with structural brain variables and cognitive decline in non-demented older adults with major depression (MD) and healthy controls (HC). We hypothesized that those with greater dispersion would have lower hippocampal volumes (HPC) and higher white matter hypointensities (WMH) and that dispersion would predict cognitive decline at one-year follow-up. Methods Subjects included 121 MD and 39 HC subjects who participated in the NBOLD study at UConn Health. Inclusion criteria were age 60+ and a psychiatrist's diagnosis of MD for the patient group. Dementia and other major neuropsychiatric illnesses were exclusion criteria. Dispersion was calculated as the standard deviation of z-scores divided by group mean z-scores of a demographically adjusted neuropsychological test performance across a comprehensive battery. Results Dispersion positively correlated with WMH in both HC (r=0.386; p =.001) and MD (r=0.318; p=0.001), and negatively correlated with left HPC volume in MD (r=-0.191; p=0.034). Linear regression analyses accounting for baseline cognitive performance, demographics, and depression found that dispersion predicted cognitive tasks performed at a year follow up. For older adults with MD word list (WL) 1-3 (p=.038), WL delayed recall (DR) (p=.020) and symbol digit modality test (SDMT) (p=.035) were predictors. For HC dispersion was predicted for logical memory (LM) DR (p=.004) and Benton test (p=.003). Conclusions Dispersion is correlated with evidence of brain aging in older adults with MD. Future studies may look at ways to reduce dispersion in older adults and help prevent cognitive decline. This research was funded by Research reported in this poster was supported by the National Institute of General Medical Sciences of the National Institutes of Health under linked Award Numbers RL5GM118969, TL4GM118971, and UL1GM118970. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Educational and research grant from the Leo and Anne Albert Charitable Trust.