Methodological components.

RTW = return to work, PRO = patient reported outcomes, Y = yes, (Y) = probably Yes, N = No, (N) = probably no, ? = unclear, n.a. = not applicable.1Participants analysed in the group to which they were initially assigned.2“No” if protocol published and all outcomes correctly reported; “?” if no protocol published and selective reporting not obvious.3From personal correspondence.4Participants were probably not aware of the intervention.5RTW-outcomes not published, incomplete outcome information (see table 3).6Results presented in subgroups, incomplete outcome information (see table 3).7Primary outcome not mentioned in protocol.8Incomplete outcome information (see table 3).9Cluster randomised trials: No risk of recruitment bias. Baseline information of individual clusters not reported. Effects of RTW-outcomes not corrected for possible design effects (risk of inflated precision).