Tatrow, K., & Montgomery, G.H. (2006). Cognitive behavioral therapy techniques for distress and pain in breast cancer patients: A meta-analysis. Journal of Behavioral Medicine, 29(1), 17–27.

To determine effectiveness of cognitive behavioral therapy (CBT) techniques in alleviating distress and pain in patients with breast cancer

• Databases searched PsycINFO, MEDLINE, CancerLIT, and CINAHL 1974–June 2004. Investigators completed manual searches of reference lists and published meta-analyses.
• Search keywords were anxiety, cognitive behavioral therapy, distress, depression, hypnosis, imagery, pain, relaxation, and treatment outcomes.
• Studies were included in the review if they
  • Were randomized controlled trials (RCTs) with no-treatment or standard-care controls
  • Provided data sufficient to allow calculation of effect sizes
  • Measured pain and distress
• Studies were excluded if they did not measure pain directly.

The search retrieved 61 studies. Investigators eliminated some on the basis of the exclusion criterion; meta-analysis involved 20 studies. Authors defined CBT very broadly: Interventions included relaxation; guided imagery; hypnosis; biofeedback; and approaches combining cognitive restructuring, skill trainingt, and other strategies to enhance problem solving and coping. Authors do not describe the specific procedure used to evaluate study quality.

• Across the 20 studies, the sample was composed of 1,703 patients. Sample range was 16–215 patients.
• Seven studies were included in the meta-analysis regarding interventions for pain. All patients in these seven studies, 474 patients over all, had breast cancer.

Authors noted that 69% of patients in treatment groups did better than controls with respect to pain. Regarding pain, effect size (d) was 0.49 (95% CI 0.09–0.90, p < 0.05). Subgroup analysis comparing individual versus group interventions showed that intervention format had no effect on pain.

Findings suggest that CBT techniques produced a small to medium effect, as defined by the study, regarding the reduction of the pain of women with breast cancer.

• Authors defined CBT very broadly, including such techniques as hypnosis and relaxation therapy, which would not usually be considered CBT approaches.
• The number of studies and overall sample size for analysis regarding pain was relatively small. Authors did not report heterogeneity.
• Though effect size was statistically significant, whether a reduction of 0.49 on a pain scale is clinically meaningful is unclear.

Findings suggest that, for women with breast cancer, CBT may be a helpful adjunct to pain management. This finding is limited by the broad definition of CBT that this study employed. This study determined effect size on the basis of statistics. Future studies should define the type of change that is meaningful to patients—that is, change in symptoms.