Part D. Chapter 5: Food Sustainability and Safety - Continued
The topic of Question 1 is new for a DGAC review and involves an emerging area of scientific investigation that is not readily addressed by traditional study designs such as randomized controlled trials and prospective cohort studies. The literature related to sustainable diets and dietary patterns involves a combination of food pattern modeling, Life Cycle Assessment (LCA) methodology (examines all processes in the life cycle of each food component - from farm to plate to waste), and determination of the environmental outcomes of the full LCA inventory. Because of the unique nature of these studies, a modified NEL systematic review was conducted for Question 1 on dietary patterns and sustainability. Databases included PubMed, Cochrane, Navigator, and Embase and the search covered from January 2000 to March 2014. For this topic and question, it was necessary to use different methods from those described in an original NEL protocol because not all methods in the protocol could be applied. This is sometimes necessary, according to the Cochrane Collaboration, but requires that methods from the original protocol that could not be implemented in the current review be summarized.21 Due to the nature of the evidence, the NEL 6-step process was tailored for the purposes of this systematic review, with modifications to step 3 extract data and assess the risk of bias. A description of the NEL systematic review process is provided in Part C: Methodology. A new data extraction grid was developed with emphasis on modeling studies, LCA methodology, and environmental outcomes. The LCA is a standardized methodological framework for assessing the environmental impact (or load) attributable to the life cycle of a food product. The customized grid was then used by NEL abstractors to extract data from the included articles and this informed the evidence synthesis (see Appendix E-2.37 Evidence Portfolio). In addition, NEL abstractors used a different tool to assess individual study quality, not the NEL Bias Assessment Tool (BAT). This alternative tool, the Critical Appraisal Checklist used by the British Medical Journal, was appropriate for studies that used a modeling design. This checklist assesses studies that use modeling to extrapolate progression of clinical outcomes, transform final outcomes from intermediate measures, examine relations between inputs and outputs to apportion resource use, and extrapolate findings from one clinical setting or population to another. To attain a high score, studies must report the variables that have been modeled rather than directly observed; what additional variables have been included or excluded; what statistical relations have been assumed; and what evidence supports these assumptions.22-24 The checklist included key components of the British Medical Journal checklist for economic evaluations, together with the Eddy checklist on mathematical models. This Critical Appraisal Checklist was reviewed and tested for applicability by two sustainability experts who served as consultants to the DGAC.
Question 2 on nutrient profiles in farm-raised versus wild-caught seafood was addressed using data analysis from the USDA-Agricultural Research Service (ARS) National Nutrient Database for Standard Reference, Release 27 (http://www.ars.usda.gov/ba/bhnrc/ndl).25 The section on finfish and shellfish products included nutrient profiles for both farm-raised and wild-caught seafood for some species. These data were augmented using a USDA-funded report on fatty-acid profiles of commercially available fish* in the United States that assessed additional farmed species and compared results with the USDA-ARS NND.26 Because this question was answered using data analysis, it was not graded (as described in Part C: Methodology). For Question 3 on contaminants in farm-raised versus wild-caught seafood, the DGAC used an expert report, the Report of the Joint Food and Agriculture Organization of the United Nations (FAO) and the World Health Organization (WHO) Expert Consultation on the Risks and Benefits of Fish Consumption, 2011.27 This report was chosen as the most updated and comprehensive source of scientific information on the net health assessment of seafood consumption, including a comparison between wild-caught and farm-raised seafood related to contaminants. Data on levels of chemical contaminants (methyl mercury and dioxins) in a large number of seafood species were reviewed, as well as recent scientific literature covering the risks and benefits of seafood consumption. The sections of the report that were used to address the question were “Data on the composition of fish” and “Risk-benefit comparisons.” Lastly, to address Question 4 on the worldwide capacity to produce enough nutritious seafood, the Committee used the FAOs report on the State of World Fisheries and Aquaculture, 2012.20 This was considered the most current and comprehensive source on this topic, specifically the sections on “Selected Issues in Fisheries and Aquaculture” and the “Organization for Economic Cooperation and Development (OECD)-FAO Agricultural Outlook: chapter on fish.” The DGAC focused on matters that directly address world production as it affects the supply of seafood for the U.S. population, particularly as the U.S. relies on significant amounts of imported seafood (~90 percent).
*The term “fish” in this chapter refers to finfish, which includes aquatic species such as salmon, tuna, and trout.
For Question 5, the DGAC used an overview of systematic reviews (SRs)/meta-analyses (MA) to address the relationship between usual caffeine/coffee consumption and health. This approach allowed the DGAC to address the broad scope of the evidence on usual caffeine and health, which heretofore had not been addressed by a DGAC. The DGAC used a modification of the method described by the Cochrane Collaboration to conduct the review.28 The steps included development of analytical framework, determination of inclusion/exclusion criteria, description of search strategy and databases used, determination of methodological quality using the Assessment of Multiple Systematic Reviews (AMSTAR) tool, data extraction, summary of results and key findings, and development of conclusion and grade for each outcome, as well as implications of the evidence and research recommendations. Overlap of studies included across the SRs/MA for the same health outcome was determined and recorded; however, SRs/MA were not excluded for overlap. This approach allowed the Committee to assess and consider whether SRs/MA on the same topic independently assessed similar results and arrived at generally similar conclusions. The focus of this review was to summarize the existing SRs/MA on this question, not to re-synthesize the evidence or to conduct a new meta-analysis or meta-synthesis.
For the overview on usual caffeine/coffee consumption and health, the target population was healthy adults and adults at risk of chronic disease, as well as youth ages 2 years and older. The intervention or exposure was caffeine/coffee consumption. The outcomes were clinical endpoints: 1) chronic diseases, including cardiovascular, type 2 diabetes, and cancer, and total mortality, 2) neurologic and cognitive diseases, including Alzheimers and Parkinsons disease, and 3) pregnancy outcomes, including miscarriage and low birth weight. The included studies were SRs/MA and qualitative SRs; the date range was from 2000 to 2014. Data were extracted for all SRs/MA with emphasis on MA results, including categorical and dose-response MA, fixed or random effects models, heterogeneity and sources of heterogeneity, sub-group analysis, and publication bias (see Appendix E-2.39b Systematic Review/Meta-Analysis Data Table [Excel - 56 KB]). The methodological quality of the included SRs/MA was determined using AMSTAR. Overlap of studies included across the SRs/MA for the same health outcomes was determined and recorded; however, SRs/MA were not excluded for overlap. Rather, the emphasis was to determine consistency across studies.
For Question 6 on high-dose caffeine and health, a duplication assessment found two SRs and these were used in lieu of conducting a full NEL SR. The details of duplication assessment are provided in Part C: Methodology, and the Review of the Evidence for this question provide further detail.
For Question 7 on aspartame and health, the European Food Safety Authority (EFSA) Scientific Opinion on the Re-evaluation of Aspartame as a Food Additive was used. This was conducted by the EFSA Panel of Food Additives and Nutrient Sources Added to Food (ANS).29 The Panel based its evaluation on original study reports and information submitted following public calls for data as well as previous evaluations and additional literature that was available up to February 2013. The 2015 DGAC considered only the human studies and related conclusions from the EFSA report; animal studies and in vitro studies were not considered.
Lastly, this chapter provides a topic update from the 2010 DGAC on consumer behaviors and food safety. Tables on this topic were updated to include the most recent recommendations. Federal sources that were used for the update include: 1) Centers for Disease Control and Prevention (CDC) - Hand washing: Clean Hands Save Lives;30 2) Food and Drug Administration (FDA) - Food Facts, Raw Produce: Selecting It and Serving It Safely, 2012; Food Safety for Moms-to-Be: Safe Eats - Meat, Poultry & Seafood;31 and 3) USDA/Food Safety and Inspection Service (FSIS) Food Safety Fact Sheets.32