However, it cannot be ruled out, that other factors, which we did not adjust for, could lead to residual confounding. The relative short time between baseline and follow-up check details may provide us limited power to detect change in health behaviour. However, such a prolonged time frame would also have limited the number of employees remaining in the

same workgroup. Among the other limitations of our study is the use of self-reported data. Also, for the workers in the home care units, contact with co-workers, and thus co-worker influence, may be limited. Unfortunately, the study questionnaire did not allow us to measure collegial ties. However, it is possible that we would find stronger cluster effects in teams with stronger interaction. Finally, the homogeneity of the sample (workers in the eldercare sector) was useful for reducing many potential confounders, but may limit the generalizability of the results. A final issue concerns workgroup size; Christakis and Fowler found an effect of co-workers on smoking cessation in small firms (up to six employees) but not in large firms (Christakis and Fowler, 2008). This may be due to the environment in larger firms, which provides more opportunities

to find co-workers with similar health behaviour. However, in sensitivity analyses, we found no effect of workgroup on smoking cessation when restricting our analyses to groups with less than 10 members. MLN8237 chemical structure We found modest evidence for clustering in baseline smoking, amount smoked and BMI within workgroups. This could be due to social learning or selection into and out of workgroups. Furthermore, we saw weight increase in workgroups

with high average BMI and smoking cessation in workgroups with a large number of smokers. Enhanced understanding and recognition of these lifestyle cluster effects may improve future health promotion programmes at worksites. The authors declare out that there are not conflicts of interest. The authors wish to thank Vilhelm Borg and Birgit Aust for their contribution to the design of the cohort study and the data collection. The cohort study was financed by the Danish Government through a grant (17.21.02-50) to the National Research Centre for the Working Environment. The writing of this manuscript was funded by a grant (#40-2009-09) from The Danish Working Environment Research Fund. The funding sources did not partake in the design, interpretation of the results, writing of the manuscript, or decisions regarding publication. “
“People are increasingly interested in taking health checks to prevent or early detect diseases or to be reassured about their health status. A health check is a service providing information, interpretation and guidance around the offer and conduct of one or more tests.