P120 The Relationship Between Sleep Architecture And Cognition In Late-Life Depression

INTRODUCTION: Depression in older people is associated with changes in sleep, however associations between sleep architecture and cognition have not yet been delineated. We examined sleep architecture in older people with and without depressive symptoms, and relationships with neuropsychological performance. METHODS: Adults over 50 years underwent overnight polysomnography and memory and executive function tests. Depression and controls groups were defined by a Geriatric Depression Scale-15 cut off score of 6. Sleep architectural outcomes included amount of slow wave sleep (SWS), rapid eye movement (REM) sleep, REM onset latency (ROL), NREM slow wave activity (SWA, 0.5–4 Hz), N2 sleep spindle density and REM density. RESULTS: The sample comprised of 71 participants with depressive symptoms and 101 controls (mean age both groups = 64, mean GDS-15 dep= 9.3, con= 1.8). There were no significant group differences in time spent in SWS, REM, REM density or SWA. Those with depressive symptoms had later ROL (p=.008) and less N2 sleep spindles (p=.03) compared to controls. A differential association was observed with less SWS being associated with poor memory recall in the depression group only (z=.342, p=0.008). No associations between sleep and executive function performance were observed. DISCUSSION: The link between less time in SWS and poorer memory in those with depressive symptoms could suggest that SWS is particularly pertinent for cognition in depression or that both sleep and cognition mechanisms are influenced by depressive state. Further studies are needed to determine if changes in sleep are linked with underlying neurobiological changes.