Influence of Socio-Demographic, Occupational and Lifestyle Variables on Sleep Time

Abstract

Background: Socio-demographic, occupational and lifestyle variables influence total sleep time. Therefore, we aimed to evaluate the influence of those variables on sleep time, and to study risk factors of being a short sleeper.

Methods: The COVISTRESS international study is an online questionnaire using the secure REDCap^® software. Total sleep time was evaluated using declared bedtime and time of awakening and was analyzed as a quantitative variable and as a qualitative variable.

Results: We included 549 respondents to the questionnaire, divided into 10-year age groups ranging from <30yo to ≥60yo. The mean quantity of sleep was 7.11±1.43 hours per night. Factors that reduce total sleep time were age (coefficient -0.19, 95CI -0.33 to 0.06), being an employee (-0.46, -0.85 to -0.06), working time (-0.18, -0.31 to 0.05), smoking ≥5 cigarettes/day (-0.5, -0.95 to -0.20), high stress at work (-0.64, -0.96 to -0.32) and at home (-0.66, -0.97 to -0.35). Being a student (0.61, 0.02 to 1.19), working less than 25h per week (0.57, 0.17 to 0.97) and telework (0.46, 0.02 to 0.89) increased total sleep time. The risk factors of being a short sleeper were age (odds ratio 1.27, 95CI 1.07 to 1.51), being an employee (2.58, 1.36 to 4.89), smoking ≥5 cigarettes/day (2.73, 1.54 to 4.84) and a high level of stress at work (2.64, 1.45 to 4.82) and at home (3.89, 2.25 to 6.63). Physical activity ≥2.5 hours/week tended to decrease the risk of being a short sleeper by 35%.

Conclusion: We demonstrated the concomitant impact of sociodemographic, occupational and lifestyle behavior on sleep, which may help to build efficient preventive strategy.

Keywords: mental health; occupational variables; sociodemographic variables; total sleep time.

Figure 1

Flow chart. For quantitative analysis, “±” means more or less.

Figure 2

Overall total sleep time, as a quantitative variable (box and whisker plot) and as a qualitative variable (prevalence of sleeping <4h, 4–5h, 5–6h, 6–7h, 7–8h, 8–9h, and >9h). In the box and whisker plot (total sleep time in grey), the lower and upper sides of the box are the lower and upper quartiles (Q1 and Q3). The box covers the interquartile interval (IQR), where 50% of the data is found. The horizontal line usually splits the box in two and is the median. In that case, the median is equal to Q3. The mean is indicated by a cross on the box plot. The whiskers are the two vertical lines outside the box, that go from the minimum to Q1 (the start of the box) and then from Q3 (the end of the box) to the maximum. The whiskers are the two vertical lines outside the box, that go from 1.5 IQR to Q1 (the start of the box) and then from Q3 (the end of the box) to the 1.5 IQR. Outliers are individuals lower and upper than the end of the whiskers (lower than Q1 - 1.5 IQR and higher than Q3 + 1.5 IQR) and are indicated by a circle.

Figure 3

Total sleep time (as a quantitative and qualitative variable) depending on sociodemographic, occupational characteristics and lifestyle behavior. Number of hours of sleep (quantitative variable) were analyzed using Student’s t test or Wilcoxon-Mann–Whitney test if data were not normally distributed for 2-group comparisons and using an analysis of variance (ANOVA) or Kruskal–Wallis test if data were not normally distributed for comparisons of 3 or more groups. Number of hours of sleep were further dichotomized into less than or equal to 6 hours of sleep and more than 6 hours. Prevalence of people sleeping 6 hours or less and more than 6 hours (qualitative variable) were analyzed using Chi2 test. Symbols are used according to the significance of the differences in the various tests. “†”: p-value < 0.1, “*”: p-value < 0.05, “**”: p-value < 0.01 and “***”: p-value < 0.001.

Figure 4

Factors influencing number of hours of sleep (univariate linear regression – see Supplementary Figure 1 for multivariate linear regression). Bolded p-values are less than 0.05 and are significant.

Figure 5

Factors influencing prevalence of sleeping less than 6 hours (univariate logistic regression – see Supplementary Figure 2 for multivariate logistic regression). Bolded p-values are less than 0.05 and are significant.