How-To Guides

Yes, adjunctive cognitive-behavioral therapy showed a significant overall benefit when added to medication for bipolar disorder, but the effect was modest. In this meta-analysis of 12 randomized clinical trials, the pooled posttreatment effect size was d = -0.42 (95% CI, -0.51 to -0.34; P < .05). After correcting for sample size, the effect remained significant but smaller at D = -0.20 (95% CI, -0.29 to -0.11; P < .05).

At posttreatment, the strongest improvement was in treatment adherence, with an effect size of d = -0.53 (P < .05). Cognitive-behavioral therapy also significantly improved clinical symptoms (d = -0.44, P < .05), cognitive-behavioral etiopathogenetic mechanisms such as dysfunctional cognitions and coping skills (d = -0.49, P < .05), and quality of life and life/social adjustment (d = -0.36, P < .05).

No significant effect on relapse and/or recurrence was found. At posttreatment, the effect size for relapse and/or recurrence was d = -0.28, which the authors described as low and nonsignificant. Relapse and/or recurrence also remained nonsignificant across follow-up periods.

The benefits were clearest at posttreatment and were generally smaller or less consistent during follow-up. The overall effect size was significant from posttreatment to 6 months (d = -0.27; 95% CI, -0.39 to -0.16) and from more than 6 months to 12 months (d = -0.41; 95% CI, -0.57 to -0.26).

At more than 12 months, the uncorrected overall effect was still significant at d = -0.27 (95% CI, -0.40 to -0.13), but after correcting for sample size it was no longer significant (D = -0.06, VAR D = 0.03, P > .05). The authors concluded that CBT benefits were more evident at posttreatment than at follow-up.

During follow-up, the most consistent benefits were for clinical symptoms and cognitive-behavioral etiopathogenetic mechanisms. From posttreatment to 6 months, CBT had significant low effect sizes for both of these domains and a low to medium effect size for treatment adherence, while quality of life and life/social adjustment were not significantly improved.

From more than 6 months to 12 months, CBT again showed significant effects on clinical symptoms and cognitive-behavioral mechanisms, but not on quality of life, relapse/recurrence, or treatment adherence. Beyond 12 months, CBT still showed significant effects on clinical symptoms and cognitive-behavioral mechanisms, while relapse/recurrence and treatment cost were not significantly affected.

No increase in overall treatment cost was identified. At posttreatment, CBT showed no effect on treatment costs, and the authors concluded that adding CBT to medication entailed significant benefits without increasing costs. At follow-up beyond 12 months, CBT also had no significant effect on treatment cost.

This was a quantitative meta-analysis of 12 randomized clinical trials evaluating cognitive-behavioral therapy as an adjunct to standard care for bipolar disorder. Eligible studies had to be randomized trials, include a clearly defined CBT intervention, include a control group, and report enough data to calculate effect sizes.

All included studies compared CBT plus standard care, defined as medication and clinical management, against standard care alone. The authors identified studies through an English-language MEDLINE search from January 1980 to March 2008 and calculated outcomes using Cohen d, with a minus sign indicating an effect in favor of CBT.

For the posttreatment analysis, the meta-analysis included 770 patients and 67 effect sizes. Follow-up analyses included 80 patients and 21 effect sizes from posttreatment to 6 months, 636 patients and 50 effect sizes from more than 6 months to 12 months, and 459 patients and 21 effect sizes beyond 12 months.

The authors concluded that cognitive-behavioral therapy can be used as an adjunctive treatment to medication for patients with bipolar disorder. Their conclusion was based on significant posttreatment benefits for symptoms, cognitive-behavioral mechanisms, treatment adherence, and quality of life, together with no apparent increase in treatment costs.

At the same time, the authors emphasized that effects were generally in the low to medium range, were less clear during follow-up, and did not significantly improve relapse and/or recurrence. They also stated that new CBT strategies are needed to strengthen posttreatment effects and maintain benefits over time.

The main limitations were the inability to separate CBT-specific effects from general psychotherapy factors, small sample sizes in many trials, possible publication bias, and major heterogeneity across studies in outcomes, follow-up intervals, and reporting style. Only 2 randomized trials compared CBT with other psychosocial interventions, so the analysis could not rigorously test whether benefits were specific to CBT rather than to nonspecific therapeutic factors.

Sample size was also a concern because only 3 studies had more than 100 patients, and the authors noted that one large study with nonsignificant findings strongly influenced follow-up results. They also emphasized that heterogeneity in methodology limited how confidently positive and negative findings could be reconciled.