A brief work-oriented intervention targeting musculoskeletal and mental health disorders: Exploring work disability outcomes in a prospective study.

The two groups were not matched on how severe the complaints were. The comparison group had significantly fewer sick-leave days at the start, which suggests they were less unwell, and that alone could explain much of the difference in results. The researchers had no data on symptom severity, so they could not rule out that people referred to the clinic were simply sicker than the comparison group in ways the analysis did not capture. Most patients in this study were still a long way from qualifying for a disability pension, so this was a rare outcome and the study may have had too few events to detect real differences. People were not randomly assigned to the clinic or to usual care, so those referred may have differed from those not referred in ways that were never measured. The outcome studied was disability pension, while the programme was designed mainly to reduce sick leave. Results for sick-leave days might differ. The programme was flexible and patients chose which parts to attend, so no single component can be credited with the results. Sick-leave spells shorter than 16 days are not recorded in the national register used here, which limits how precise the comparison between groups can be.

The study was funded by Vestfold Hospital Trust, the same health trust that ran the outpatient clinic being evaluated. The authors declared no conflicts of interest.

Do not read the higher disability pension risk among the earliest-referred patients as proof that the programme harms people referred early. The two groups were not matched on how unwell they were, the comparison group had significantly fewer sick-leave days at the start, and most patients here were still far from qualifying for a disability pension, so who was referred may explain the difference as much as what the programme did.