Genetic risk of Alzheimer’s disease is linked to short sleep duration

Abstract

AbstractBackgroundSleep disturbances including short sleep duration are common in older adults, especially in those with Alzheimer’s disease (AD). However, it is unclear to what extent sleep duration is a manifestation of AD disease process. We examined whether genetic variants related to AD influence sleep duration in middle‐aged and older adults using a Mendelian randomization (MR) approach.MethodWe examined 406,687 UK Biobank participants without dementia at enrollment, with Caucasian genetic ancestry, and who self‐reported sleep duration at baseline (2006‐2010). Sleep duration was assessed by asking: “About how many hours sleep do you get in every 24 hours? (please include naps).” A genetic risk score for AD (AD‐GRS) was calculated as a weighted sum of 23 previously identified AD‐related single nucleotide polymorphisms in individuals of European ancestry. AD diagnosis was assessed over follow‐up (to 2018) based on ICD‐9/10 codes. We evaluated whether genetic risk for AD influenced sleep duration using two steps: first we used linear regression to examine the association between AD‐GRS and sleep duration, adjusting for age, sex and principle components for genetic ancestry. Next, we conducted a two‐stage least square regression using AD‐GRS as an instrumental variable for AD diagnosis to estimate the magnitude of the effect of AD on sleep duration. We also stratified the analysis by age at baseline (≤55y or >55y).ResultParticipants (aged 56.91±8.00y) had an average sleep duration of 7.2 (Standard deviation [SD]=1.1) hours and AD‐GRS of 0.11 (SD=0.40; range:‐1.15∼1.85). Higher AD‐GRS score predicted shorter sleep duration (b= ‐0.013, 95%CI:‐0.022,‐0.005), mainly among those aged over 55y (b= ‐0.023, 95%CI:‐0.034,‐0.012) and not in those 55y or younger (b= 0.006, 95%CI:‐0.012,0.013); p for interaction by age=0.02. Analysis using AD‐GRS as an instrumental variable for AD diagnosis suggested that participants with AD diagnosis had 1.76 hrs (b=‐1.76, 95%CI: ‐2.62, ‐0.90) shorter sleep duration in those aged >55y.ConclusionUsing a novel analytical approach, we found that higher genetic risk for AD predicted shorter sleep duration among older adults. This suggests shared genetic pathways; the biologic processes that lead to AD may also affect sleep duration.