January 2010
Reviewer: Jeff C. Huffman, MD
Telephone-delivered collaborative care for treating post-CABG depression: a randomized controlled trial.
Background: Depression after coronary artery bypass graft (CABG) surgery is common and independently associated with morbidity and mortality. There had been no randomized trials of depression treatment for post-CABG patients.
Methods: The authors performed a randomized controlled trial of a telephone-delivered collaborative care depression management program among patients with depressive symptoms shortly after CABG that persisted at 2 week follow-up. In a standard collaborative care format, the nurse care managers coordinated care between patient, physician depression specialists, and primary care physicians (PCPs) such that patients had choice of treatment (e.g., workbook, medication, specialist referral) and the care managers provided ongoing evaluation to allow iterative treatment as needed. Study outcomes included health-related quality of life (HRQL; measured via mental/physical components of SF-36), depression symptoms (measured via Hamilton Depression Rating Scale [HDRS]), functional status (Duke Health Status Index [DHSI]), and cardiac readmissions.
Results: A total of 302 patients with depression (positive initial PHQ-2 evaluation and PHQ-9 score =10 at 2 week follow-up) were enrolled were randomized to 8 months of collaborative care vs. usual care (feedback to treaters). Patients in the collaborative care arm of the study had significantly greater improvements of mental HRQL (3.2 point difference in SF-36; p=0.02), depression symptoms (3.1 point difference in HDRS; p=0.001), and functional status (4.6 point difference in DHSI; p=0.001); there were no significant between-group differences in physical HRQL or cardiac readmissions. Fifty percent of intervention patients had a 50% or more reduction of depression symptoms, compared to 29.6% in the usual care arm. Between-group differences were generally more marked for men.
Discussion: This was a carefully-designed study of a relatively standard collaborative care model (with the added cost and convenience benefit of telephone-based interventions) in a very important population of patients. On one hand, the findings are not overwhelming, with moderate-seeming improvements in depression, function, and HRQL, and no impact on physical HRQL (though in men, the differences in physical HRQL were significant) or readmission, despite a focused and persistent intervention targeting depression.
On the other hand, this is a very efficient, cost-effective, often well-accepted, and generalizable model of depression care that in this case did appear to improve a variety of very important outcomes in a group of patients who suffer substantial morbidity. The effect size of the intervention on depression, as noted by the authors, was greater than in other major depression trials in cardiac patients (e.g., SADHART, CREATE, and ENRICHD), and such an effect occurred in a population recruited from a wide range of settings. It is probably not realistic to expect that focused depression treatments will have major effects on ‘objective’ cardiac outcomes (e.g., readmissions) unless trials are both very large and have very long periods of follow-up.
One final note about this trial is that the investigators recruited patients with relatively mild depression symptoms; the mean PHQ-9 score was approximately 13, representing relatively milder depression, and less than 40% of the cohort met criteria for major depression. One wonders if the between-group differences would have been more marked if only MDD patients would have been included (given the likelihood of spontaneous recovery in some patients on the milder spectrum), or whether patients with true MDD often require more intensive interventions. In any event, the methods used in this study helped the whole cohort of subjects in several domains, and this important study paves the way for future studies that can drill down to how best to use this intervention in a very important population.