Dalmases M, Benítez ID, Mas A, Garcia-Codina O, Medina-Bustos A, Escarrabill J, Saltó E, Buysse DJ, Roure N, Sánchez-de-la-Torre M, Rué M, Barbé F, de Batlle J.PLoS One. 2018 Apr 18;13(4):e0194495. doi: 10.1371/journal.pone.0194495. eCollection 2018.

Abstract

OBJECTIVE:

To describe the overall sleep health of the Catalan population using data from the 2015 Catalan Health Survey and to compare the performance of two sleep health indicators: sleep duration and a 5-dimension sleep scale (SATED).

METHODS:

Multistage probability sampling representative of the non-institutionalized population aged 15 or more years, stratified by age, gender and municipality size, was used, excluding nightshift-workers. A total of 4385 surveys were included in the analyses. Associations between sleep health and the number of reported chronic diseases were assessed using non-parametric smoothed splines. Differences in the predictive ability of age-adjusted logistic regression models of self-rated health status were assessed. Multinomial logistic regression models were used to assess SATED determinants.

RESULTS:

Overall mean (SD) sleep duration was 7.18 (1.16) hours; and SATED score 7.91 (2.17) (range 0-10), lower (worse) scores were associated with increasing age and female sex. Alertness and efficiency were the most frequently impaired dimensions across age groups. SATED performed better than sleep duration when assessing self-rated health status (area under the curve = 0.856 vs. 0.798; p-value <0.001), and had a linear relationship with the number of reported chronic diseases, while the sleep duration relationship was u-shaped.

CONCLUSIONS:

Sleep health in Catalonia is associated with age and gender. SATED has some advantaged compared to sleep duration assessment, as it relates linearly to health indicators, has a stronger association with self-rated health status, and provides a more comprehensive assessment of sleep health. Therefore, the inclusion of multi-dimensional sleep health assessment tools in national surveys should be considered.

Pubmed