ACT and Multidisciplinary ACT for Women on Long-Term Sick Leave: 8-Year Outcomes from a Swedish RCT
- Three-arm RCT, women on long-term work disability for common mental disorders (CMD) or chronic musculoskeletal pain in Sweden (Karolinska / Uppsala). Randomized to unimodal Acceptance and Commitment Therapy (ACT), multidisciplinary assessment + ACT (TEAM), or standard care (Control). Work disability tracked over 8 years, health outcomes over 10. Registered NCT03343457.
- Both intervention arms had lower predicted median work disability days at every annual time point versus Control. Differences reached statistical significance only for TEAM at years 4, 7, and 8 — i.e. the unimodal ACT effect on work outcomes was directionally consistent but underpowered for late-window inference.
- On psychiatric symptom reduction, both ACT and TEAM outperformed Control at 1- and 2-year follow-up — short-term mental health gains were robust across both intervention models.
- Cost-effectiveness analysis over 8 years supported both rehabilitation models; the choice between unimodal ACT and the multidisciplinary TEAM was framed as a function of available resources and individual patient complexity, not absolute superiority of one over the other.
This is the kind of trial we almost never see. Most RCTs in chronic pain and common mental disorders stop the clock at six or twelve months, sometimes pushing to two years if a grant survives. Here a Swedish team followed a real-world cohort of women on extended sick leave for nearly a decade — and reports both health and labour-market outcomes against the same control. For practitioners working with patients who have spent more than a year out of work, this study reframes what "treatment success" looks like.
The lead paragraph for our reader is this: short-term symptom reduction is not the same outcome as eight-year work re-engagement, and these two trajectories diverge depending on which intervention package you choose.
What the data actually shows
Both ACT and TEAM beat Control on psychiatric symptoms within the first two years — that part is consistent with what we already know about third-wave CBT for CMD and chronic pain. The new signal is downstream: the multidisciplinary TEAM arm pulled ahead on work disability days specifically at years 4, 7 and 8. Unimodal ACT showed the same direction but did not reach significance at those late time points. Both interventions remained non-inferior to control on safety. The authors are careful to frame this as a stepwise resource-allocation question, not a winner-take-all comparison.
Two methodological notes that matter clinically. First, the sample is exclusively women — generalisability to mixed-gender or male-dominant disability cohorts is not established here. Second, the diagnostic envelope (CMD or chronic musculoskeletal pain) was deliberately broad. That mirrors the heterogeneity of an actual occupational rehabilitation queue but limits diagnosis-specific conclusions.
For your practice
For a patient who has been out of work for more than 12 months, unimodal ACT delivered competently is a defensible first move — symptoms will likely improve, and the long-term work-disability trajectory will be at least non-inferior to standard care. But if the patient presents with a complex profile (psychiatric comorbidity plus pain plus social-vocational deconditioning), the additional return on a multidisciplinary assessment-plus-ACT package is what shows up at year 4 and beyond. The decision is not "which therapy works." It is "how complex is this case, and what is the realistic time horizon I am being asked to influence?"
Practical implication: when you write a treatment plan for a long-term-sick-leave referral, write it on a multi-year horizon. The reimbursement system rewards short-window outcomes; the patient's life does not. If you have access to a multidisciplinary team for the high-complexity subgroup, use it. If you do not, ACT alone is still a reasonable choice — but adjust your communication with the referrer about what is and is not realistic over an eight-year arc.
Symptom relief in two years and work re-engagement in eight years are different outcomes — choose the intervention package against the time horizon you are actually being asked to influence.
Single-country (Sweden), women-only sample, broad diagnostic envelope; the unimodal-ACT effect on late work-disability outcomes was directionally consistent but underpowered for late-window significance. Long follow-up reduces internal validity protections (cross-over, secular changes in welfare policy).