The review concluded that most cardiac rehabilitation programmes, delivered in the home, significantly reduced depression symptoms in adults aged 64 years or older. The authors’ conclusions do not reflect the evidence presented, this together with unclear synthesis methods, and the unknown quality of the trials, means the conclusions are not reliable.

To examine the effectiveness of community-based cardiac rehabilitation on depression outcomes for adults aged 64 years or older.

Five databases including PubMed, PsycINFO and CINAHL were searched for relevant articles published in English from 1998 to January 2012. Search terms were reported.

Eligible were randomised controlled trials (RCTs) of community-based cardiac rehabilitation interventions for patients (mean age 64 years or older) with heart disease (diagnosis defined in review). Interventions could be delivered in the home or out-patient clinic or both. Trials had to report depression outcomes.

In the included trials, the interventions were delivered to out-patients or remotely by telephone and included collaborative medical care, nutrition and health education, structured exercise, Tai Chi, motivational interviewing and/or psychosocial interventions. Control groups were usual care including standard medical care, heart education, aerobic exercise, nutrition counselling and/or telephone support. The outcomes were assessed using a range of depression measures, as well as by a walking test, physical symptoms, or assessment of quality of life or anxiety. The mean age of participants was 71 years (range 64 to 80). Some trials included only men or only women. Participants had a primary diagnosis of heart failure or coronary artery disease, with a range of co-morbidities. Half the trials were conducted in the USA; the others were conducted in the UK, Canada, Italy, Japan or Taiwan.

Two reviewers independently selected trials for inclusion.

The assessment of trial quality covered: randomisation, blinding of outcome assessment, and management of attrition in the analysis.

Two reviewers independently assessed quality.

The means and standard deviations, for intervention and control groups, in depression outcomes, were extracted and were used to calculate the standardised mean difference between groups at the latest follow-up assessment. Corresponding 95% confidence intervals were calculated.

Two reviewers independently extracted the data.

The data appear to have been combined to calculate an average standardised mean difference, with 95% confidence interval (methods not reported). The results were grouped into trials of patients with heart failure and those of patients with coronary artery disease. The standardised mean difference and 95% confidence interval for each trial were recorded on a forest plot and discussed in a narrative. Publication bias was assessed by visual inspection of funnel plots.

Eighteen trials (3,827 participants; range 32 to 570) were included. Follow-up ranged from less than one month to eight months.

Overall, improvements were reported in depression scores as a result of community-based cardiac rehabilitation interventions, but these improvements were not statistically significant (average SMD 0.18, 95% CI -0.64 to 0.29; 15 trials). Other results for individual trials were reported.

There was no evidence of publication bias (details not shown).

The evidence suggested that most cardiac rehabilitation programmes, delivered in the home, could significantly improve depression. Tailored interventions combined with psychosocial interventions were likely to be more effective, for depression in older adults with heart disease, than usual care.

The review question was clearly defined with appropriate inclusion criteria. Several relevant sources were searched. The restriction to trials published in English left potential for language and publication bias, but the authors reported that there was no evidence of such bias. Appropriate methods to reduce reviewer error and bias were used throughout the review.

Trial quality was appraised using appropriate criteria, but the full results for all the included trials were not reported. Some trial details were given, but the details of each intervention were missing. The results were partly reported in a narrative, which was hard to follow given the lack of trial details. The outcomes for depression appear to have been combined in a meta-analysis, but the methods were not reported, making it impossible to determine their suitability.

The authors’ conclusions do not reflect the evidence presented, this together with unclear synthesis methods, and the unknown quality of the trials, means the conclusions are not reliable.

The authors did not state any implications for practice and research.

Not stated.