Sleep problems were more prevalent among women than men and more prevalent in the British than the Finnish cohort. At baseline, 21% of Finnish women and 31% of British women reported frequent sleep problems. The corresponding figures among men were 16% and 22%. At follow-up, the prevalence of frequent sleep problems was 27% among Finnish women and 34% among British women, respectively, and 20% and 27%, respectively, among men ().
Distributions (%) of key study variables in the Finnish Helsinki Health Study and the British Whitehall II occupational cohorts.
In contrast, differences in the level of the economic difficulties and their changes were minor between the two cohorts. Frequent economic difficulties at baseline and follow-up were reported by 10% of the Finnish, and by 7% of the British cohort (). Over the follow-up, 43% of women and 50% of men in the Finnish cohort reported no economic difficulties, while in the British cohort 46% of women and 51% of men reported no economic difficulties at either phase. There was a tendency for economic difficulties to decrease over the follow-up period, with 21% and 24% of participants in the Finnish and British cohorts reporting a decrease in economic difficulties compared with 16% and 10% reporting an increase.
Strong associations were observed between persistent frequent economic difficulties and sleep problems. These remained after adjustment for age, sex, baseline sleep problems, childhood economic difficulties, marital status, education, occupational class, household income at baseline and follow-up, housing tenure, and employment status at follow-up in the Finnish cohort (OR 1.72, 95% CI 1.35–2.18). In the British cohort, the corresponding association was found in the age and sex adjusted model (OR 1.35, 95% CI 1.05–1.73) (). However, when stricter criteria for classification of persistent frequent economic difficulties were applied (prevalence 2.4%), the association remained strong and similar to the Finnish cohort also in the British cohort throughout the modelling. In the pooled analyses, persistent frequent economic difficulties also remained associated with sleep problems after full adjustments, and no interaction between cohort and economic difficulties was found (data not shown).
Associations between changes in economic difficulties and sleep problems at follow-up. Odds ratios (OR) and their 95% confidence intervals (CI).
An increase in economic difficulties over the follow-up (from “none” at baseline to “frequent” at follow-up) was also associated with sleep problems at follow-up in the Finnish cohort after full-adjustment (OR 1.81, 95% CI 1.22–2.68). In the British cohort, evidence for this association was strong and statistically significant only in the age and sex adjusted model (OR 1.64, 95% CI 1.00–2.69), although the estimates remained equal, in terms of effect size, after full adjustment. An increase in economic difficulties from occasional to frequent was also associated with sleep problems in the Finnish (OR 1.73, 95% CI 1.30–2.32) and the British (OR 1.51, 95% CI 1.00–2.27) cohort, after full adjustments but an increase from none to occasional had no effect. There was no strong evidence in either cohort that decreases in economic difficulties were associated with sleep problems.