Obstructive sleep apnea risk is associated with a diminished cognitive profile: findings from the Canadian...

Obstructive sleep apnea (OSA) and insomnia are highly prevalent in older adults and both have been linked to cognitive decline. However, it remains unclear whether the physiological burden of OSA or the sleep fragmentation associated with insomnia more strongly predicts cognitive impairment. This study examined whether OSA risk or insomnia better was associated with memory problems and cognitive slowing among community-dwelling older Canadians. Given the cross-sectional design, the study evaluated associations rather than causal relationships.

Data were drawn from 946 participants in the Canadian Longitudinal Study on Aging (CLSA), including 473 individuals reporting a physician-diagnosed memory problem and 473 age- and sex-matched controls. OSA risk was assessed using the validated STOP-BA(N)G questionnaire. At the same time, insomnia was defined using operational research criteria based on sleep onset latency, wake after sleep onset, and duration. Cognitive processing speed was measured via Choice Reaction Time (CRT). General linear models, logistic regression, and discriminant analyses examined associations between sleep parameters and cognitive outcomes, controlling for age, sex, body mass index, income, and diabetes.

Participants with memory problems had significantly higher STOP-BA(N)G scores (F(1,931) = 7.10, p = 0.008) and slower CRTs (p = 0.05) compared to controls. Insomnia alone or comorbid with OSA (COMISA) was not associated with cognitive outcomes. In discriminant analysis, STOP-BA(N)G score (r = 0.45) and diabetes (r = 0.26) were the strongest predictors of memory problems, correctly classifying 66.6% of cases. Sex-stratified analyses demonstrated that the association between higher STOP-BA(N)G scores and memory problems remained significant in both men and women, although effect sizes were somewhat larger in men.

Higher OSA risk was associated with poorer memory and slower cognitive processing in older adults, whereas insomnia was not associated with these outcomes. Because of the cross-sectional design, causality cannot be inferred. Screening for OSA and metabolic comorbidities in patients with memory complaints may help identify individuals at increased risk of cognitive impairment.