Background: Various physical work demands are shown to be associated with sickness absence. However, these studies have: (a) predominantly used self-reported data on physical work demands that have been shown to be inaccurate compared with technical measurements, (b) principally focused on various physical work demands in 'isolation', i.e. ignoring their co-dependency - compositional nature -, and (c) mainly used register data on long-term sickness absence. The present article describes the protocol of a study with the objective of investigating the association between technically measured compositional data on physical work demands and prospective long- and short-term register-based data on sickness absence.
Methods: 'The technically measured compositional Physical wOrk DEmands and prospective association with register-based Sickness Absence study (PODESA)' comprises data from two Danish cohorts (NOMAD and DPhacto) primarily on blue-collar workers. In the PODESA cohort, data on 1108 workers were collected at baseline (between 2011 and 2014). The cohort data comprise, e.g., self-reported information on descriptives, lifestyle, workday, and health, as well as accelerometer-based measurements of physical work demands (physical activity, movements, and postures). These baseline measurements are linked with prospective register-based data on sickness absence for up to four years after baseline. The prospective association between physical work demands and sickness absence will be analysed using a Compositional Data Analysis approach.
Discussion: PODESA provides a unique possibility of unravelling which combinations of physical work demands are associated with prospective sickness absence. PODESA employs technically measured information on physical work demands (taking into account the compositionality of physical work demand data) and prospective sickness absence data. The findings from PODESA can be used to develop strengthened preventive interventions for sickness absence. Results are expected in 2019-2021.
Keywords: Accelerometers; Compositional data analysis (CoDA); Physical activity at work; Sick-leave; Time-use epidemiology.