Impact of reporting quality on risk of bias assessment in occupational health and safety trials

Article type
Authors
Tikka C1, Verbeek J2, de Boer AGEM3, Hoving J2, Boschmann JS2, Hulshof CTJ3
1Finnish Institute of Occupational Health
2Cochrane Work Review Group, Academic Medical Centre, University of Amsterdam
3Coronel Institute of Occupational Health, Academic Medical Centre, University of Amsterdam
Abstract
Background: It is our impression that trials used in Cochrane Work Reviews are not well reported and are often considered at high risk of bias. We hypothesized that if trials would be better reported this would facilitate the risk of bias assessment and result in a lower risk of bias. We focus on the most important aspect of randomised controlled trials (RCTs): randomisation and allocation concealment.

Objectives: We want to assess how well authors of occupational health and safety (OHS) trials report the process of randomisation and allocation concealment. In addition, we want to find out if better reporting quality is associated with a lower risk of bias assessment.

Methods: We conducted a systematic literature search in PubMed for RCTs published after 2010 in 18 OHS journals, with workers or workplaces as participants, irrespective of intervention and comparison. We excluded cross-over trials, protocols, pilot studies and studies that only report secondary outcomes, cost-effectiveness analyses, and exposure studies. We rated the quality of reporting as high, moderate, low, or very low for each article based on the number of CONSORT 2010 checklist items complied with (100%,