SLEEP EARLIER IN LIFE AND LATE-LIFE DEMENTIA: MULTI-CENTER POPULATION DATA FROM SWEDEN AND FINLAND

Abstract

Sleep disturbances commonly follow the onset of dementia. However, few longitudinal studies have assessed how sleep disturbances can increase dementia risk. The current study examines the association between sleep disturbances and dementia using population-based data. In three population-based studies (two Swedish, one Finnish), general questions were administered about sleep quality, reduced sleep hours and terminal insomnia. The short-term follow-up analyses (9–10 years follow-up time) used all three studies (N=1446). Mean baseline ages in the three datasets were 70, 70, and 84 years. The long follow-up analyses used the Finnish dataset with 22 and 32 years follow-up (N=759) and mean age was 50 years. We performed binary logistic and hazard regressions for the associations between sleep and dementia. The following potential baseline confounders were adjusted for: Data material, follow-up time, baseline age, sex, years of education (linear), alcohol consumption (linear), presently smoking, physically active, cohabitation, cardiovascular conditions, hypnotics (yes/no), APOE4 allele (for short follow-up data) and hopelessness. Short follow-up results showed significant associations in all models between terminal insomnia and a higher likelihood of dementia (fully adjusted model using all covariates, (odds ratio (OR)=1.94, p=0.027). No associations were found between reduced sleep and dementia (OR=0.99). Long follow-up results (Finnish data) showed that insomnia was associated with a higher risk for dementia (fully adjusted hazard ratio=1.24 p=0.030). In conclusion, more severe insomnia earlier in life is associated with a higher risk of late-life dementia. Individuals with sleep disturbances may benefit from interventions to improve sleep.