Objective and subjective sleep measures are associated with HbA1c and insulin sensitivity in the general population: Findings from the ORISCAV-LUX-2 study - 20/05/21

Doi : 10.1016/j.diabet.2021.101263

Gloria A. Aguayo ^a,⁎ , Jessica Pastore ^a, Anne Backes ^a, Saverio Stranges ^a,^b, Daniel R. Witte ^c,^d, Nico J. Diederich ^e, Ala’a Alkerwi ^f, Laetitia Huiart ^a, Maria Ruiz-Castell ^a, Laurent Malisoux ^a, Guy Fagherazzi ^a
on behalf of the Oriscav-Lux Study Group
^a Department of Population Health, Luxembourg Institute of Health, Strassen, Luxembourg
^b Department of Epidemiology and Biostatistics and Family Medicine, Schulich School of Medicine and Dentistry, University of Western Ontario, London, Ontario, Canada
^c Department of Public Health, Aarhus University, Aarhus, Denmark
^d Steno Diabetes Center Aarhus, Aarhus, Denmark
^e Department of Neurosciences, Centre Hospitalier de Luxembourg, Luxembourg City, Luxembourg
^f Directorate of Health, Luxembourg-Hamm, Luxembourg

^⁎Corresponding author at: Deep Digital Phenotyping Research Unit, Department of Population Health, Luxembourg Institute of Health, 1A-B, rue Thomas Edison, L-1445 Strassen, Luxembourg.Deep Digital Phenotyping Research UnitDepartment of Population HealthLuxembourg Institute of Health1A-B, rue Thomas EdisonStrassenL-1445Luxembourg

Sous presse. Manuscrit accepté. Disponible en ligne depuis le Thursday 20 May 2021
Cet article a été publié dans un numéro de la revue, cliquez ici pour y accéder

Highlights

•	Objective sleep measures are more frequently associated with HbA1c levels than are subjective sleep measures.
•	Sleep patterns and HbA1c are mostly associated with men, and with those who are either overweight and/or depressed.
•	Use of accelerometers to assess sleep patterns could help to better target subjects at high risk of diabetes.

Le texte complet de cet article est disponible en PDF.

Abstract

Aim

To analyze the association of objective and subjective sleep measures with HbA1c and insulin sensitivity in the general population.

Methods

Using a cross-sectional design, data from 1028 participants in the ORISCAV-LUX-2 study from the general population in Luxembourg were analyzed. Objective sleep measures were assessed using accelerometers whereas subjective measures were assessed using the Pittsburgh Sleep Quality Index (PSQI) questionnaire. Sleep measures were defined as predictors, while HbA1c and quantitative insulin sensitivity check index (QUICKI) scores were considered outcomes. Linear and spline regression models were fitted by progressively adjusting for demographic and lifestyle variables in the total sample population as well as by stratified analyses using gender, obesity status, depressive symptoms and diabetes status.

Results

In fully adjusted models, total and deep sleep durations were associated with lower HbA1c (mmol/mol) levels, whereas sleep coefficients of variation (%) and poor sleep efficiency, as measured by PSQI scores (units), were associated with higher HbA1c levels. In stratified models, such associations were observed mainly in men, and in subjects who had depressive symptoms, were overweight and no diabetes. In addition, total sleep, deep sleep, coefficients of variation and poor sleep efficiency as measured by PSQI revealed non-linear associations. Similarly, greater insulin sensitivity was associated with longer total sleep time and with PSQI-6 (use of sleep medication).

Conclusion

Associations were more frequently observed between sleep characteristics and glycaemic control with the use of objective sleep measures. Also, such associations varied within subgroups of the population. Our results highlight the relevance of measuring sleep patterns as key factors in the prevention of diabetes.

Le texte complet de cet article est disponible en PDF.

Keywords : Accelerometer, General population, HbA1c, Pittsburgh Sleep Quality Index, Sleep patterns, Type 2 diabetes