Work-related sleep disturbances and sickness absence in the Swedish working population, 1993-1999

To examine secular trends in work-related sleep disturbances and their association with sickness absence in the Swedish working population.

Nationally representative cross-sectional samples of the Swedish working population aged 16-64 (the biennial Swedish Work Environment Survey) in 1993, 1995, and 1999 respectively. Questionnaire data on work-related sleep disturbances were linked to records of medically-certified sick-leave spells exceeding 14 days obtained from national registers.

All Sweden.

A total of 28,424 individuals aged 16-65 with complete data (5162/5173 women/men in 1993; 4635/4764 in 1995; and 4422/4268 in 1999).

N/A.

The age-adjusted proportion of women with work-related sleep disturbances at least once a week increased from 12.3% in 1993 to 21.7% in 1999 (P < 0.001). The corresponding figures for men were 12.5% to 18.6% (P < 0.001). There was a strong cross-sectional association between work-related sleep disturbances and sickness absence in both genders and in each studied year. Using binary logistic regressions and adjusting for age, supervisory position, and geographical region, the odds ratios for sickness absence for those who reported work-related sleep disturbances every day, compared with those who answered "not at all/seldom last 3 months" varied between 3.22 (1.88-5.50) and 4.26 (2.56-7.19), with the strongest associations seen in 1999. Adjustment for health indicators, especially depressive symptoms, attenuated the relationship substantially.

Self-reported sleep disturbances attributed to work-related causes were on the rise in Sweden and were associated with medically-certified sickness absence. Most of this association seems to be accounted for by depressive symptoms.