Abstract P060: Major Dietary Risk Factors for Chronic Diseases: A Systematic Review of the Current Evidence for Causal Effects and Effect Sizes
Background: Diet habits contribute to development of CVD and diabetes. Estimating the impact of diet on these diseases requires identification and quantification of causal effects of dietary factors. Objectives: To assess major dietary risk factors for CVD and diabetes, evaluate current evidence for causal effects, and identify the best unbiased effect estimates on risk. Methods: For multiple dietary risk factors, we evaluated WHO and similar criteria as part of the Global Burden of Diseases (GBD) study to assess probable or convincing evidence for causal effects, including consistency, dose-response, plausibility, and temporality. We performed systematic searches of online databases from 2008 to 2011, including hand-searches of references and author contacts, to identify systematic reviews and meta-analyses of well-designed observational or interventional studies. Meta-analyses were evaluated based on number of studies, design, definition of diet factors and outcomes, sample size, number of events, length of follow-up, statistical methods, evidence of bias, and control for confounders. Meta-analyses with largest numbers of studies and events and least evidence for bias were identified. Effect sizes and uncertainty were quantified per defined units of exposure, including pooling of categorical dose-response estimates using fixed-effects generalized least squares for trend estimation (GLST). Results: We identified 15 dietary risk factors having probable or convincing evidence of causal effects on CVD or diabetes. For 13, data were identified to provide the best pooled unbiased effect size on disease (Table). Conclusions: This systematic evaluation provides the best evidence-based quantitative estimates of the effects of major dietary factors on CVD and diabetes. These findings enable estimation of quantitative impacts on diseases burdens of suboptimal intakes of these factors in specific populations, and also highlight gaps in knowledge related to causality or effect sizes of other dietary factors.