Job Strain Tied to Sleep Disturbances Among US Workers Over Time

Higher job strain is associated with an increase in sleep disturbances over time in US workers, according to a study published in the American Journal of Industrial Medicine.

Researchers conducted a study to evaluate different formulations of job strain in predicting sleep outcomes using data from Midlife in the United States (MIDUS), a longitudinal study addressing behavioral, psychological, and social factors in adults aged between 25 and 74 years. Data collection was performed in 3 waves (1995-1996, 2004-2006, and 2013-2017).

The researchers measured job strain using self-reported data and defined using Karasek’s Job-Demand-Control model, which combines high job demands with low job control. Job strain was formulated in 6 approaches, including standard quadrant, simplified quadrant, linear, quotient, logarithm quotient, and quartile. Sleep disturbances were quantified from 4 self-reported items and assessed on a 5-point Likert scale. Longitudinal associations were estimated using Generalized Estimating Equations (GEE), and corrected Quasi‐likelihood Information Criterion (QICu) was performed to confirm the fit for each model.

A total of 1721 participants were included (mean age, 51.1 years [SD, 9.2]; White race, 86.6%; women, 52.5%). Baseline descriptive statistics of job strain indicated that 24.4% of participants experienced high job strain. All formulations of job strain were significantly associated with sleep disturbances; associations were slightly attenuated after adjusting for demographic, socioeconomic, and health-related characteristics at baseline.

All 3 continuous measures of job strain approaches showed significant associations with sleep disturbances, with similar results observed in linear and quotient job strains. In the quartile job strain model, the medium-high and high quartiles showed significant associations (relative risk [RR], 1.11 and 1.26, respectively).

The logarithm quotient approach consistently showed the lowest QICu values across all models, followed by the linear and quotient approaches. The simplified approach had the highest QICu values across all models.

On examination of the single effects of baseline job demands and job control, both were significantly associated with sleep disturbances across follow-up.

Study limitations included self-reported data and limited generalizability of the sample.

“Implementing regulations to monitor and redesign workload, as well as promote workers’ autonomy, may help improve sleep health, therefore potentially reducing the burden on both individual and healthcare system levels,” the researchers wrote.