Meta-Analysis: Secondary Prevention Programs for Patients with Coronary Artery Disease

Although supervised exercise programs reduce mortality in survivors of myocardial infarction, the effects of other types of cardiac secondary prevention programs are unknown. PURPOSE: To determine the effectiveness of secondary cardiac prevention programs with and without exercise components. DATA SOURCES: The authors searched MEDLINE (1966-2004), the Cochrane Central Register of Controlled Trials, EMBASE, CINAHL, SIGLE, and the Cochrane Effective Practice and Organization of Care Study Registry. They also contacted primary study authors and hand-searched bibliographies provided by the Centers for Medicare & Medicaid Services. STUDY SELECTION: Randomized clinical trials. DATA EXTRACTION: Two reviewers chose studies and extracted data independently; random-effects summary risk ratios were calculated. DATA SYNTHESIS: The authors identified 63 randomized trials (21 295 patients with coronary disease). The summary risk ratio was 0.85 (95% CI, 0.77 to 0.94) for all-cause mortality, but this result differed over time with a risk ratio of 0.97 (CI, 0.82 to 1.14) at 12 months and 0.53 (CI, 0.35 to 0.81) at 24 months. The summary risk ratio was 0.83 (CI, 0.74 to 0.94) for recurrent myocardial infarction over a median follow-up of 12 months. Effects were similar for programs that included risk factor education or counseling with a structured exercise component (risk ratio, 0.88 [CI, 0.74 to 1.04] for mortality and 0.62 [CI, 0.44 to 0.87] for myocardial infarction), for programs that included risk factor education or counseling without an exercise component (risk ratio, 0.87 [CI, 0.76 to 0.99] for mortality and 0.86 [CI, 0.72 to 1.03] for myocardial infarction), and for programs that were solely exercise-based (risk ratio, 0.72 [CI, 0.54 to 0.95] for mortality and 0.76 [CI, 0.57 to 1.01] for myocardial infarction). Most of these programs improved quality of life or functional status, but effect sizes were small.

Although these programs may reduce total health care costs, published data on the costs of the programs are inadequate to conclusively comment on their cost-effectiveness.