Ten RCTs (n=3,138) were included in the review.
A formal validity assessment was not performed. However, the following methodological limitations were highlighted in the review: the combination of CBT with other interventions; insufficient description of the intervention; the lack of a true control group; the potential crossover between groups; the lack of baseline measures; the high number of drop-outs; and the small sample sizes and short follow-up periods.
Cardiovascular disease (7 studies, n=2,945).
Six of the 7 studies evaluating CBT in patients with cardiovascular disease found a significant reduction in symptoms of depression in the intervention group compared with controls. For each intervention type there was only one study. Interventions showing positive findings were: CBT alone (6.9 in the BDI version II versus 15 with usual care, p=0.037; no baseline measures reported); CBT combined with selective serotonin re-uptake inhibitors where indicated (9.1 on the BDI compared with 12.2 in the usual care group, p<0.001); CBT combined with relaxation training and stress management (-5.2 versus -0.2 with cardiac rehabilitation on the Symptom Checklist 90 revised depression subscale, p<0.034); CBT combined with cardiac rehabilitation (-4.25 versus -0.2 with a waiting-list group on the Hospital Anxiety and Depression Subscale, p=0.001); CBT with exercise (52% decrease in BDI scores in the intervention group versus 15% and 25% increases in Lanoxin titrate and placebo control groups, respectively, p=0.04); and CBT combined with biofeedback and health education compared with a health education class (1 study; no statistical data provided).
Stroke (1 study, n=123).
One study evaluated CBT in patients with stroke and found no significant differences between the intervention group and controls or attention controls.
Type II diabetes (2 studies n=70).
One study found a significant benefit of CBT combined with diabetes education compared with an education only group (58.3% remission from depression versus 25.9% in controls, p=0.03). The other study found no significant benefit of CBT combined with progressive muscular relaxation and problem-solving compared with a waiting-list control group.