OBJECTIVES

To estimate the clinical effectiveness of active management (AM) in general practice versus AM plus a group-based, professionally led cognitive behavioural approach (CBA) for subacute and chronic low back pain (LBP) and to measure the cost of each strategy over a period of 12 months and estimate cost-effectiveness.

DESIGN

Pragmatic multicentred randomised controlled trial with investigator-blinded assessment of outcomes.

SETTING

Fifty-six general practices from seven English regions.

PARTICIPANTS

People with subacute and chronic LBP who were experiencing symptoms that were at least moderately troublesome.

INTERVENTIONS

Participants were randomised (in a ratio of 2:1) to receive either AM+CBA or AM alone.

MAIN OUTCOME MEASURES

Primary outcomes were the Roland Morris Disability Questionnaire (RMQ) and the Modified Von Korff Scale (MVK), which measure LBP and disability. Secondary outcomes included mental and physical health-related quality of life (Short Form 12-item health survey), health status, fear avoidance beliefs and pain self-efficacy. Cost-utility of CBA was considered from both the UK NHS perspective and a broader health-care perspective, including both NHS costs and costs of privately purchased goods and services related to LBP. Quality-adjusted life-years (QALYs) were calculated from the five-item EuroQoL.

RESULTS

Between April 2005 and April 2007, 701 participants were randomised: 233 to AM and 468 to AM+CBA. Of these, 420 were female. The mean age of participants was 54 years and mean baseline RMQ was 8.7. Outcome data were obtained for 85% of participants at 12 months. Benefits were seen across a range of outcome measures in favour of CBA with no evidence of group or therapist effects. CBA resulted in at least twice as much improvement as AM. Mean additional improvement in the CBA arm was 1.1 [95% confidence interval (CI) 0.4 to 1.7], 1.4 (95% CI 0.7 to 2.1) and 1.3 (95% CI 0.6 to 2.1) change points in the RMQ at 3, 6 and 12 months respectively. Additional improvement in MVK pain was 6.8 (95% CI 3.5 to 10.2), 8.0 (95% CI 4.3 to 11.7) and 7.0 (95% CI 3.2 to 10.7) points, and in MVK disability was 4.3 (95% CI 0.4 to 8.2), 8.1 (95% CI 4.1 to 12.0) and 8.4 (95% CI 4.4 to 12.4) points at 3, 6 and 12 months respectively. At 12 months, 60% of the AM+CBA arm and 31% of the AM arm reported some or complete recovery. Mean cost of attending a CBA course was 187 pounds per participant with an additional benefit in QALYs of 0.099 and an additional cost of 178.06 pounds. Incremental cost-effectiveness ratio was 1786.00 pounds. Probability of CBA being cost-effective reached 90% at about 3000 pounds and remained at that level or above; at a cost-effectiveness threshold of 20,000 pounds the CBA group had an almost 100% probability of being considered cost-effective. User perspectives on the acceptability of group treatments were sought through semi-structured interviews. Most were familiar with key messages of AM; most who had attended any group sessions had retained key messages from the sessions and two-thirds talked about a reduction in fear avoidance and changes in their behaviour. Group sessions appeared to provide reassurance, lessen isolation and enable participants to learn strategies from each other.

CONCLUSIONS

Long-term effectiveness and cost-effectiveness of CBA in treating subacute and chronic LBP was shown, making this intervention attractive to patients, clinicians and purchasers. Short-term (3-month) clinical effects were similar to those found in high-quality studies of other therapies and benefits were maintained and increased over the long term (12 months). Cost per QALY was about half that of competing interventions for LBP and because the intervention can be delivered by existing NHS staff following brief training, the back skills training programme could be implemented within the NHS with relative ease.

TRIAL REGISTRATION

Current Controlled Trials ISRCTN37807450.

FUNDING

The National Institute for Health Research Health Technology Assessment programme.