Effectiveness of Workplace Interventions in Return-to-Work for Musculoskeletal, Pain-Related and Mental Health Conditions: An Update of the Evidence and Messages for Practitioners.

The review did not combine studies in a meta-analysis because interventions, workplaces, outcomes and designs were too different, so it gives evidence categories rather than a single estimate of benefit. Many included studies were medium quality. 16 lacked randomisation or allocation concealment, 15 had substantial loss to follow-up, 22 had uneven attrition, 21 lacked evidence of intervention compliance, 27 lacked blinding of participants or personnel, 13 used non-optimal statistical analyses, and 15 failed to state the primary hypothesis clearly. The main outcome was lost time from work. Work functioning was reported in 8 studies and costs in 15 studies, so the review says less about how well people function after returning or about the cost consequences of the interventions. The review grouped studies by consensus about the primary intervention objective. This can make multi-domain programmes look effective as a whole without showing which component produced the effect. Descriptions of compensation system factors were often absent, although the samples were mostly workers with compensable injuries. The best evidence synthesis method has been criticized as at risk of bias, although it used transparent criteria for assigning evidence levels.

The named funder was the Institute for Safety Compensation and Recovery Research. The supplied text does not report additional author declarations.

Do not conclude that any workplace-based return-to-work intervention reduces lost time. The review found strong positive evidence only for multi-domain interventions for musculoskeletal or pain-related conditions and work-focused CBT for mental health conditions; traditional CBT alone had no effect on lost time, and several other categories had limited, mixed or insufficient evidence.