Dados do Trabalho

Título

Frequency and Predictors of Sleep Duration Misperceptions: Data from 2,036 participants from the ELSA-Brasil Study.

Introdução

Accumulating evidence links short and/or long sleep duration (SD) with several metabolic and cardiovascular diseases. However, the majority of the previous studies used self-reported SD. It is unclear, however, the magnitude and the factors associated with SD misperceptions that may potentially limited the utility of using subjective SD.

Objetivo

To evaluate the frequency and predictors of Sleep Duration (SD) misperceptions in a large cohort of participants not referred to sleep laboratories.

Métodos

Participants from the ELSA-Brasil study underwent clinical and sleep evaluation (including mean subjective SD and wrist actigraphy for 7 days). We considered a significant SD underestimation (underSD) and overestimation (overSD) when the differences between subjective and objective SD reached at least -1 hour/+1-hour, respectively. The Bland‑Altman method and Pearson correlation coefficients (r) were assessed for exploring potential bias and agreement between the self-reported and actigraphy data. We used logistic regression analyses for identifying characteristics associated with SD misperceptions.

Resultados

Data from 2,036 participants were used in the final analysis (42.7% males; mean age: 49±8 years). Subjective SD revealed poor correlations and low agreement with objective SD. The overall frequency of significant SD misperceptions was 39.1% (underSD: 19.4%; overSD: 19.7%). The predictors of underSD included: black race (OR: 1.88; 95% CI: 1.35, 2.64), mixed race (OR: 1.57; 95% CI: 1.18-2.09); excessive daytime sleepiness (OR: 1.46; 95% CI: 1.12-1.91); longer wake time after sleep onset time, WASO (OR: 2.38; 95% CI: 1.34-4.26), and longer objective SD (OR: 2.98; 95% CI: 2.52-3.52). Longer WASO (OR: 2.24; 95% CI: 1.33-3.79), and higher number of awakenings (OR: 1.02; 95% CI: 1.00-1.03) were independently associated with overSD. Interestingly, married status, high education levels and higher sleep efficiency were associated with good SD perceptions.

Conclusões

Subjective has poor correlations and agreement with objective measurements of SD contributing to a significant rate of SD misperceptions. While underSD was more associated with black/mixed race, daytime sleepiness, longer SD and WASO, overSD was specifically associated with markers of sleep fragmentation. Considering the available technology, these results underscore the need for stopping the use of subjective data for SD definitions.

Palavras-chave

sleep duration, actigraphy; measurement error.