Pain is an unpleasant experience that may be the manifestation of an underlying disease or injury; treatment of the underlying condition may resolve pain. However, pain may persist and become chronic despite all possible medical treatments having been pursued. Chronic pain (defined as pain lasting more than three months) affects about 20% of people worldwide and can be very disabling, leading to depression, anxiety, and social isolation. In this scenario, psychological treatment aims to reduce impairments, distress, and catastrophic thinking associated with pain.

Two types of psychological treatment, namely cognitive behavioural therapy (CBT) and behaviour therapy, help to remodel dysfunctional thoughts, feelings, and behaviour that typically worsen pain and quality of life.

This review investigated whether psychological therapy was helpful for adults experiencing non-malignant chronic pain (except for headache) compared to other active treatments, placebo, or waiting list controls.

This review is important for

People with chronic pain, their loved ones/caregivers, health professionals caring for this population, general practitioners, researchers, and policymakers.

Outcomes of this review

This is an update of a Cochrane Systematic Review (CSR) published in 2009. The outcomes analysed were pain experience, disability, negative mood, and catastrophic thinking. The CSR included 42 trials, but only 35 provided usable data. They were published and/or registered up to September 2011, with 4,788 adult participants. Results showed that compared with active control there was no evidence for a positive effect of behaviour therapy on any outcome analysed, except for a small improvement in mood immediately post-treatment. On the other hand, compared with active controls CBT had a small improvement in disability and catastrophic thinking but not in pain or mood. Compared with the usual treatment or waiting list, CBT had small to moderate improvement in pain, disability, mood, and catastrophic thinking immediately post-treatment but was not sustained at six months follow-up except for a small improvement in mood. In conclusion, psychological therapies may be useful to people with chronic pain, but there is insufficient data on the quality or type of treatment and which type of patients treatment may be most beneficial.

These results must be interpreted with caution because (i) there was a lack of coherence in the theory underlying many studies; (ii) it is not clear how CBT is able to induce changes; (iii) there was heterogeneity among patients and (iv) among treatment procedures; (v) outcomes may not be independent even if they were analysed as such; (vi) appropriate models to guide intervention were lacking, and (vii) control groups were poorly designed.

Author’s conclusion of the review

Compared with treatment as usual or waiting list, CBT slightly improved pain immediately post-treatment. The same effect could be observed for CBT but not for behaviour therapy on disability, which was partially sustained at six months follow-up. CBT is effective in changing mood and catastrophic thinking when compared with treatment as usual or waiting list, with some evidence that this effect is sustained at six months. Behaviour therapy had no effect on mood but showed a positive effect on catastrophic thinking immediately post-treatment. Overall CBT could be a useful treatment for the management of chronic pain.

Future recommendations

The authors recommended that new randomized controlled trials comparing CBT to simple alternatives should be stopped unless a new population or treatment are investigated.

This is because clear evidence of small to moderate effect emerges from the current review and the ones cited and is unlikely to change from further studies.

Nevertheless, trials may be conducted to examine (a) response trajectories; (b) responder analyses; and (c) clinical effectiveness focusing on minimizing costs and adverse events and maximizing benefits. (d) A better theory for the mechanism of change needs to be elaborated and tested, and (e) further studies should find out which components of CBT are effective and for which type of patient.

Robin Kuruvila Sentinella