Skip navigation

Cognitive behavioural therapy for distress and pain in breast cancer

 

Clinical bottom line

There is only limited support for cognitive behavioural therapy (CBT) for distress and pain in patients with breast cancer using individual interventions because the trials were small and may be biased. Larger trials using group interventions showed no benefits over control for distress or pain.


Reference:

TitK Tatrow, GH Montgomery. Cognitive behavioural therapy techniques for distress and pain in breast cancer patients: a meta-analysis. Journal of Behavioral Medicine 2006 29: 17-27.le

Background

Emotional distress is common in women with breast cancer, and pain is experienced at some time, either from the disease or its treatment, and at various stages.

Systematic review

The review sought randomised trials comparing CBT with no treatment or standard care controls with outcomes of distress and pain. CBT was broadly defined and included any intervention with components of behavioural or cognitive techniques.

Results

In this review, trial size for individual interventions ranged from 12 to 110 patients (713 total, and those for group interventions (658 patients) from 50 to 215. For individual interventions the median trial size was 55 (mean 53) patients; only one of 13 trials had more than 100 patients. For group interventions the median trial size was 110 (mean 92) patients; three of six trials had more than 100 patients.

The outcomes were reported as effect sizes. The overall effect size for 0.3 for distress and 0.5 for pain, indicating a moderate effect. Larger studies (using predominantly group interventions) had no significant benefit; smaller studies used mainly individual interventions. Trials using individual interventions had significantly larger effect sizes than those using group interventions.

Comment

We know that in other circumstances small trials tend to show greater effects of treatment than large trials. The concern here is that the difference between individual or group interventions could be because small trials used for individual interventions have some bias, making them look positive when they are not. Until individual interventions are studied in larger trials, there must be some uncertainty over the efficacy of the intervention.