1005Dietary pattern assessment methods and implications for dietary guideline development

Abstract

Abstract Background Dietary guidelines should be informed by evidence from dietary patterns research. However, variation in the methods used to assess and report dietary patterns can make this type of evidence difficult to synthesise. The aim of this systematic review was to analyse the methods used to assess and report dietary patterns. Methods Literature searches were conducted in 2019 using three electronic databases. Cohort and nested case control studies published in English between 1980 and 2019 were eligible for inclusion. A data extraction tool was designed to capture details of the dietary pattern assessment methods that were used. Results A total of 410 studies were included: 63% used index based dietary pattern assessment methods, 31% used factor analysis or principal component analysis; 6% used cluster analysis; and 6% used reduced rank regression. Some studies used multiple methods. Within studies that used similar methods, variation in the methodology existed. The level of detail used to describe each dietary pattern assessment method also varied, and in some cases, important methodological details were omitted. Conclusions There was considerable variation in the methods used to assess and report dietary patterns. To ensure evidence from dietary patterns research can be synthesised, standardised approaches for the reporting of dietary pattern assessment methods are needed. Key messages To ensure evidence from dietary patterns research can be synthesised and used to inform dietary guidelines, there is a need to standardise the methods used to report dietary pattern assessment methods.