Patient-friendly summary
If you read nothing else
Bottom line
Multidisciplinary biopsychosocial rehabilitation produces small but durable reductions in pain and disability for chronic low back pain — better than usual care (moderate quality evidence) and physical treatment (low quality evidence) — and improves the odds of being at work versus physical treatment, but not versus usual care.
Moderate evidencePublished
Evidence hierarchy
Study participants
Adults with chronic low back pain (over 3 months), mean symptom duration over one year, many having already failed conservative treatment
Full research — for clinicians and curious readers
Study Summary
This Cochrane systematic review and random-effects meta-analysis pooled 41 randomised controlled trials (6858 participants with chronic low back pain) to assess the long-term effects of multidisciplinary biopsychosocial rehabilitation — programmes combining a physical component with psychological and/or social/work-targeted components delivered by clinicians from at least two professional backgrounds. Compared with usual care, multidisciplinary rehabilitation modestly reduced pain and disability (moderate quality evidence); compared with physical treatments it reduced disability and roughly doubled the odds of being at work one year later (low to moderate quality evidence). For work outcomes it was more effective than physical treatment but not more effective than usual care. The authors describe these benefits as small but durable beyond one year, weighed against the cost and resource intensity of such programmes.
Key Findings
| Finding | Detail | Impact |
|---|---|---|
| Multidisciplinary rehabilitation modestly reduced long-term pain and disability versus usual care | Moderate quality evidence: standardised mean difference 0.21 (95% CI 0.04 to 0.37) for pain, equivalent to about 0.5 points on a 0-10 scale, and 0.23 (0.06 to 0.40) for disability, equivalent to about 1.5 points on the 24-point Roland-Morris index. | High |
| Multidisciplinary rehabilitation reduced long-term disability versus physical treatments | Low quality evidence (19 trials): standardised mean difference for disability 0.68 (0.16 to 1.19). The effect on pain was sizeable (0.51) but marginally failed to reach significance (-0.01 to 1.04); high statistical heterogeneity was present (I2 >90%), strongly influenced by one outlying study. | Medium |
| Roughly doubled odds of being at work one year later versus physical treatment | Moderate quality evidence (eight trials): odds ratio 1.87 (1.39 to 2.53) for being at work 12 months after intervention. | High |
| No added work benefit over usual care | Moderate quality evidence (seven trials): odds ratio 1.04 (0.73 to 1.47) for being at work compared with usual care, with similar findings short and medium term. | Medium |
| Similar outcomes to surgery but fewer adverse events | Two trials found little difference between rehabilitation and surgery for pain, disability, or work, but more adverse events occurred with surgery (odds ratio 28.25, 3.77 to 211.93 — very imprecise owing to low absolute rates). | Medium |
| Optimal dose and intervention intensity remain unclear | Subgroup analyses could not confirm that more intensive programmes (>100 hours, daily contact) produced better outcomes; the optimal dose of multidisciplinary rehabilitation remains unknown. | Low |
Strengths
- Largest collection of trials and participants on this topic to date (41 RCTs, 6858 participants), providing the first quantitative long-term effect-size estimates
- Conducted with Cochrane best-practice methods; key decisions on selection, analysis, and synthesis made in advance of searches
- Risk of bias independently assessed by two raters and incorporated into GRADE quality ratings
- Conservative, transparent downgrading of evidence quality for risk of bias, heterogeneity, and imprecision
- Sensitivity analyses showed effect estimates were not unduly inflated by high-risk-of-bias studies
Limitations
- High statistical heterogeneity in multidisciplinary rehabilitation versus physical treatment comparisons (I2 >90%), heavily influenced by one outlying study, making the average effect potentially unrepresentative of individual studies
- Work absenteeism was inconsistently measured across trials, hampering definitive conclusions on this outcome
- The vast majority of studies were conducted in Europe, so caution is needed when applying results to other healthcare systems
- Blinding of patients, clinicians, and assessors was not possible owing to the nature of the interventions and patient-reported outcomes
- Subgroup analyses on symptom severity and intervention intensity were inconclusive, leaving the optimal dose unknown
- Methodological quality of included studies was mixed; only 13 of 41 (32%) were rated as low risk of bias
Key Takeaways for Patients
What This Means for You
For long-standing back pain, a coordinated programme that addresses your body, mind, and work life together tends to work better over the long run than usual GP care or physical treatment alone.
The improvements in pain and disability are real but modest, and they last beyond a year.
Compared with hands-on or exercise-only physical treatment, these programmes roughly double your chances of being back at work a year later — though they did not beat usual care on work outcomes.
These programmes had outcomes similar to back surgery but with fewer harmful side effects, supporting careful, selective use of surgery.
The best 'dose' or intensity of these programmes is still unknown, and they can be costly and time-consuming, so they are most worthwhile for people with significant physical and psychological impact from their pain.