Characteristics of psychological interventions that improve depression in people with coronary heart disease: a systematic review and meta-regression

OBJECTIVE: Despite previous intervention trials, it is unclear which psychological treatments are most effective for people with coronary heart disease (CHD). We have conducted a systematic review with meta-regression to identify the characteristics of psychological interventions that improve depression and depressive symptoms among people with CHD.

METHODS: Searches of multiple electronic databases up to March 2012 were conducted, supplemented by hand-searching of identified reviews and citation tracing of eligible studies. Studies were included if they reported a randomized controlled trial of a psychological intervention for people with CHD and included depression as an outcome. Data on main effects and characteristics of interventions were extracted from eligible studies. Standardized mean differences (SMDs) were calculated for each study and pooled using random-effects models. Random-effects multivariate meta-regression was performed to identify treatment characteristics associated with improvements in depression.

RESULTS: Sixty-four independent treatment comparisons were identified. Psychological interventions improved depression, although the effect was small (SMD=0.18, p<.001). Problem solving (SMD=0.34), general education (SMD=0.19), skills training (SMD=0.25), cognitive-behavioral therapy (CBT; SMD=0.23), and relaxation (SMD=0.15) had small effects on CHD patients who were recruited irrespective of their depression status. Among high-quality trials of depressed CHD patients, only CBT showed significant but small effects (SMD=0.31). When entered into multivariable analysis, no individual treatment component significantly improved depression.

CONCLUSIONS: CBT and problem solving should be considered for inclusion in future treatment developments and randomized controlled trials. However, the effects are small in magnitude, and there is room to develop new interventions that may be more effective.