Evidence Synthesis and Evaluation in Nutrition

Abstract

Chronic non-communicable diseases affect a large proportion of the population and are associated with significant morbidity, mortality, and social and economic impact. Large cohort and modelling studies estimate that a substantial proportion of these conditions can be attributed to dietary habits. Clinicians, guideline developers, policymakers, and researchers use systematic reviews that address the relationship between dietary exposures and health outcomes to advise the public on optimal dietary habits, formulate recommendations and policies, and plan future research. A growing body of evidence, however, suggests that there are serious problems with current methods for evidence synthesis and evaluation in nutrition, examples of which include overreliance on expert opinion and consensus, failure to follow standard systematic review methods, and the application of inconsistent criteria for the assessment of the certainty of evidence. These issues have led to ineffective (at best) or harmful (at worst) dietary recommendations and policies and the proliferation of research that cannot be confidently applied to guide dietary decisions. The objective of this thesis is to advance methods for evidence synthesis and evaluation in nutrition. The thesis begins by reviewing contemporary challenges in evidence synthesis and evaluation for dietary guideline development and offering novel insight on opportunities for future improvement. The thesis subsequently provides a descriptive analysis of limitations of recently published systematic reviews and meta-analyses of nutritional epidemiology studies. This thesis then presents two systematic reviews and meta-analyses addressing the health effects of red and processed meat consumption that serve as examples of the application of rigorous systematic review methods in nutrition. This thesis ends by describing opportunities and challenges for future evidence synthesis and evaluation in nutrition.