Fifteen RCTs (n=1,492) were included in the review. All studies included in the analysis had a validity assessment score of 4 or more.
Depression: CBT was more effective than control for the treatment of depression, (five RCTs, n=621) effect size 1.21 (95% CI: 0.22 to 2.19, p=0.02). The sensitivity analysis reported that studies using an individual treatment approach for CBT showed a larger clinical effect compared to control (four RCTs) effect size 1.44 (95% CI: 0.29 to 2.59, p<0.05). A group CBT format (one RCT) found no statistically significant difference compared to control. There was evidence of statistical heterogeneity between these studies (p<0.001). Although CBT was more effective than control for short-term follow-up effect size 1.81 (95% CI: 0.72 to 2.89, p<0.01), there were no long-term effects for the intervention.
Two RCTs (n=306) that evaluated the effects of patient education on depression compared to control did not find any statistically significant differences between groups at follow-up. There was no evidence of statistical heterogeneity between these two studies.
Anxiety: CBT was more effective than control for treatment of anxiety (four RCTs, n=554) effect size 1.99 (95% CI: 0.69 to 3.31, p=0.03). The sensitivity analysis reported that studies using an individual treatment approach for CBT showed a larger effect compared to control (three RCTs, n=251) effect size 2.41 (95% CI: 1.26 to 3.55, p<0.01), but there was evidence of statistical heterogeneity (p<0.001). No statistically significant difference was found for group CBT interventions, but there was evidence of statistical heterogeneity (p<0.001). CBT showed no long-term effects on anxiety compared to control, although the intervention was effective in the short term effect size 2.87 (95% CI: 2.38 to 3.34, p<0.01).
There were no statistically significant differences between groups for the study evaluating patient education in comparison with control (one RCT, n=140).
Pain: There were no statistically significant differences between groups for pain for treatment with either CBT or patient education compared to control.
Physical Functioning: There were no statistically significant differences between groups for physical functioning for treatment with either CBT or patient education compared to control.
Quality of life: CBT was more effective than control for quality of life scores (eight RCTs, n=933) effect size 0.91 (95% CI: 0.38 to 1.44, p=0.001). The sensitivity analysis showed that studies using an individual approach for CBT had a greater effect than control (seven RCTs, n=860) effect size 0.95 (95% CI: -0.37 to 1.54, p=0.001). The study using a group intervention found no statistically significant effect, but there was evidence of statistical heterogeneity (p<0.001). CBT showed significant effect on quality of life scores for both short-term follow up, effect size 1.45 (95% CI: 0.43 to 2.47, p<0.01) and long-term follow up, effect size 0.26 (95% CI: 0.06 to 0.46, p<0.05).
There were no statistically significant differences between groups for patient education and control (one RCT, n=140).
Publication bias: There was no evidence of publication bias for studies included in the depression and anxiety analyses, but there was evidence of publication bias for studies included in the quality of life analyses (p<0.03).