Addressing long-term sickness absence: moving beyond disease, illness and work-related factors for effective preventionAlthough only a relatively small percentage of workers report sick for longer than a few days or weeks, it should be noted that long-term sickness absence constitutes the major part of the already substantial costs of sickness absence for companies and society (1). Moreover, long-term absence from work has been associated with a further detrimental impact on the mental health and wellbeing of the workers involved (2). These adverse effects, on top of the diminished social contacts with colleagues and coworkers and related alienation from work, often result in a return-to-work (RTW) barrier that grows as the duration since reporting sick from work increases. Moreover, the tendency of human beings to try to stick to the existing circumstances in uncertain situations (3), such as persisting sickness behavior, might further hamper the RTW process. Taken together, all these factors can lead to a downward spiral of deteriorating health and increased difficulties with reintegrating back into work. And as such, the process of longterm sickness absence can lead to permanent work disability. So its prevention should be an important component of occupational health policy, not only to reduce sickness-absence-related costs but also to improve workers' health and eventually avoid work disability.Establishing evidence-based models that depict the process leading up to long-term sickness is a logical and important step in establishing and improving preventive strategies. Most of the formulated current sickness absence models have a strong focus on health (4, 5) but other factors -often work-related or addressing social protection -have also been included. The large, multinational study by Mortensen et al (6) in this issue of the Scandinavian Journal of Work, Environment and Health contributes to the Literature as it considers the relevance of a factor outside the domain of work and health, namely informal caregiving (ie, unpaid assistance with the daily activities of sick, disabled, or elderly relatives), as a predictor of longterm sickness absence. The authors report that informal caregiving was associated with a higher risk of sickness absence among women [hazard ratio (HR) 1.13, 95% CI 1.04-1.23]. Mortensen et al opted for the inclusion of their underlying theoretical causal model in the form of a directed acyclic graph (DAG), a graphical representation of the assumed causal associations introduced by Robins (7). A DAG not only provides relevant and graphically illustrated information on the underlying theoretical model, but it also guides the statistical analyses and selection of potential confounders, as well as possibly preventing the reporting of spurious associations due the inclusion of colliding factors. A stimulating and helpful discussion for further use of the DAG, and its associated collider bias, can be found in the work of Glymour et al (8), for example. As presented by Mortensen et al (6), the...