Food-based dietary guidelines: methods used to synthesise evidence and grade recommendations




Poster session 3


Tuesday 25 October 2016 - 10:30 to 11:00


All authors in correct order:

Blake P1, Durão S2, Naude C3, Bero L1
1 Charles Perkins Centre and Faculty of Pharmacy, University of Sydney, Australia
2 Cochrane South Africa, South African Medical Research Council, South Africa
3 Centre for Evidence-Based Health Care, Stellenbosch University, South Africa
Presenting author and contact person

Presenting author:

Celeste Naude

Contact person:

Abstract text
Background: Evidence-based guidelines are informed by rigorously conducted systematic reviews and use structured consensus frameworks, such as GRADE (Grading of Recommendations, Assessment, Development and Evaluations), to grade evidence quality and strength of recommendations. Given advances in these methods there is a need to evaluate methods used to develop dietary guidelines for population health.

Objectives: To describe the methods used for evidence synthesis and grading of recommendations in national food-based dietary guidelines (FBDGs).

Methods: One author handsearched the Food and Agriculture Organization’s FBDGs database (14 January 2016). We included the latest versions of FBDGs in any language, published from 2010 onwards; aligned with the WHO definition of a guideline; and aimed at a general healthy population. We also included referenced documents on guideline development methods. One author extracted information on: country; publication date; type of evidence reviewed; methods used to conduct systematic reviews, rate the strength of recommendations, and manage conflicts of interest (COI). Data extraction was checked by the same author and questions were resolved through author discussions.

Results: We included 30 of 79 eligible FBDGs (18 English, 12 other languages). Most were based on other countries’ guidelines (16/30) and published systematic reviews or reports (13/30). Three guidelines commissioned systematic reviews. Most guidelines reported methods used to define evidence review questions (28/30), but few reported methods used to search (5/30), extract data (2/30), evaluate methodological quality (6/30), or synthesize evidence (1/30). Most used consensus to rate recommendations (27/30) and four used structured consensus frameworks. Few reported COIs (4/30) or funding sources (9/30).

Conclusions: Our study highlights discrepancies in FBDG development across countries and a dependence on other countries’ guidelines likely due to resource constraints. Governments and research organizations should implement efficient, explicit and reproducible methods for dietary guideline development that balance rigor and pragmatism.