Multiple risk factor interventions for primary prevention of coronary heart disease

Abstract

BACKGROUND: Multiple risk factor interventions using counselling and educational methods assumed to be efficacious and cost-effective in reducing coronary heart disease (CHD) mortality and morbidity and that they should be expanded. Trials examining risk factor changes have cast doubt on the effectiveness of these interventions. OBJECTIVES: To assess the effects of multiple risk factor interventions for reducing total mortality, fatal and non-fatal events from CHD and cardiovascular risk factors among adults assumed to be without prior clinical evidence CHD.. SEARCH STRATEGY: We updated the original search BY SEARCHING CENTRAL (2006, Issue 2), MEDLINE (2000 to June 2006) and EMBASE (1998 to June 2006), and checking bibliographies. SELECTION CRITERIA: Randomised controlled trials of more than six months duration using counselling or education to modify more than one cardiovascular risk factor in adults from general populations, occupational groups or specific risk factors (i.e. diabetes, hypertension, hyperlipidaemia, obesity). DATA COLLECTION AND ANALYSIS: Two authors extracted data independently. We expressed categorical variables as odds ratios (OR) with 95% confidence intervals (CI). Where studies published subsequent follow-up data on mortality and event rates, we updated these data. MAIN RESULTS: We found 55 trials (163,471 participants) with a median duration of 12 month follow up. Fourteen trials (139,256 participants) with reported clinical event endpoints, the pooled ORs for total and CHD mortality were 1.00 (95% CI 0.96 to 1.05) and 0.99 (95% CI 0.92 to 1.07), respectively. Total mortality and combined fatal and non-fatal cardiovascular events showed benefits from intervention when confined to trials involving people with hypertension (16 trials) and diabetes (5 trials): OR 0.78 (95% CI 0.68 to 0.89) and OR 0.71 (95% CI 0.61 to 0.83), respectively. Net changes (weighted mean differences) in systolic and diastolic blood pressure (53 trials) and blood cholesterol (50 trials) were -2.71 mmHg (95% CI -3.49 to -1.93), -2.13 mmHg (95% CI -2.67 to -1.58 ) and -0.24 mmol/l (95% CI -0.32 to -0.16), respectively. The OR for reduction in smoking prevalence (20 trials) was 0.87 (95% CI 0.75 to 1.00). Marked heterogeneity (I(2) > 85%) for all risk factor analyses was not explained by co-morbidities, allocation concealment, use of antihypertensive or cholesterol-lowering drugs, or by age of trial. AUTHORS' CONCLUSIONS: Interventions using counselling and education aimed at behaviour change do not reduce total or CHD mortality or clinical events in general populations but may be effective in reducing mortality in high-risk hypertensive and diabetic populations. Risk factor declines were modest but owing to marked unexplained heterogeneity between trials, the pooled estimates are of dubious validity. Evidence suggests that health promotion interventions have limited use in general populations.