Harmful alcohol habits were no more common in a sample of newly sick-listed Swedish women and men compared with a random population sample.

AIMS To estimate harmful alcohol habits in a sample of incident sick-listed individuals compared with a random sample from the general population taking social background, health and work-related factors into account. METHODS Data for this cross-sectional questionnaire study were collected in 2008 in the Västra Götaland region, Sweden. The study population (19-64 year olds) consisted of 2888 consecutive incident sick-leave sample (ISS) and 3567 individuals from a random population sample (RPS). The mailed questionnaire included Alcohol Use Disorder Identification Test and validated instruments on health and work-related factors. Socio-demographic data came from register data. Analyses were made with χ(2) tests and logistic regression analyses. RESULTS No differences in prevalence of harmful alcohol habits were found between men in the ISS (22%) and the RPS (21%). Compared with women in the ISS, a higher proportion of women in the RPS were likely to report harmful alcohol habits [14 versus 9% (P < 0.001)]. This difference was confirmed in the logistic regression analyses where women in the RPS had higher odds of having harmful alcohol habits compared with women in the ISS [odds ratio (OR) = 1.54 (95% confidence interval (CI): 1.23-1.89)]. Even after controlling for significant confounders (age, low income, high self-reported health and high level of perceived symptoms), we found that the differences in harmful alcohol habits remained [OR = 1.44 (95% CI: 1.16-1.81)]. CONCLUSIONS Harmful alcohol habits were no commoner in men and women who belonged to the sample of incident sick-leave cases. Future studies are needed to analyse the predictive value of harmful alcohol habits on sickness absence length and the time until return to work after sickness absence.