Relação entre duração do sono noturno e desempenho cognitivo: um estudo com idosos da comunidade

Abstract

The changes resulting from the aging process are related to the increase in the incidence and prevalence of chronic noncommunicable diseases and may lead to a decline in functional and cognitive capacity as well as sleep disturbances. Sleep is an essential function of life that acts as a repairer and regulator of various organic systems, being indispensable for physical and mental health. Thus, the objective of this research was to identify correlation between the duration of self-reported nocturnal sleep and the cognitive performance of the elderly attended at Family Health Units (USFs). It was a quantitative, descriptive and cross-sectional study. The convenience sample consisted of 156 elderly people enrolled in USFs in the city of São Carlos (SP), divided into four groups according to the duration of self-reported nocturnal sleep. Data collection was performed through interviews with the characterization questionnaire, Addenbrooke Cognitive Examination - Revised (ACE-R) and Pittsburgh Sleep Quality Index (PSQI). Statistical analyzes Pearson's Chi-Square test, Kruskal-Wallis test for descriptive and comparative analyzes and Multiple Linear Regression were used. All stages of this work obeyed the ethical guidelines of research involving human beings. The elderly were mostly female (80.1%), married or with partners (n = 84, 53.8%), with a mean of 70.4 years (± 6.89), low level of schooling (M = 3.65 years, ± 3.27) and income (M = 2,007.97 reais, ± 1,160.99). According to the PSQI categorization, 55.1% (n = 86) had good sleep quality. The elderly had a mean of 61.94 points (± 17.35) in ACE-R. Comparison of the ACE-R domains with respect to dormant hours per night (quartiles) showed differences between groups in the domains of verbal fluency (X2 = 10.060; gl = 3; p = 0.018), the same was not observed in the other domains , as well as for total ACE-R scores (X2 = 6.770, gl = 6, p = 0.080). The post-hoc comparative analyzes showed that elderly patients who slept for more hours, on average 8.85 hours (Q1), had lower verbal fluency scores (M = 4.92; ± 3.15) when compared to the elderly who slept (M = 6.90, ± 2.48) (U = 558,000, p = 0.004) and 4.52 hours (Q4) (M = 6.15, ± 2.74) (U = 794.00, p = 0.045). The adjusted model with application of the stepwise method eliminated the variables duration of sleep, sex, age and medication use, considering only the schooling as a predictor of cognition (β = 0.609, t = 9.492, p = 0.000). Being the equation that describes this relation: Cognition = 50,407 + [3,196x (schooling in years)]. For the verbal fluency domain, the independent variables of schooling (β = 0.444, t = 6.224, p = 0.000) and sleep duration (β = -0.165, t = -2.319, p = 0.022) remained in the model. The equation that describes the relationship is: Verbal Fluency = 6.158 + [0.401 x (schooling in years)] + [-0.265 x (sleep duration in hours)]. For the sample studied, it is concluded that sleep duration is related to the verbal fluency domain, and the elderly with up to six hours of sleep presented higher cognitive scores when compared to long-sleeper elders, but no relation of duration of sleep with total cognition scores.