Development, validation, and application of a South African Dietary Polyphenol Data Quality Assessment Tool.

Abstract

There is a wealth of South African scientific literature studying and reporting on dietary polyphenols in research articles and academic literature. The evaluation of data quality is critical to ensure polyphenol data compiled from various scientific journals and academic texts are acceptable in terms of compilation standards. This study aimed to develop, validate, and apply a country specific South African dietary polyphenol data quality assessment tool (SADPDQAT). The evaluation tool considered nine evaluation categories with a maximum score of 5. For each literature source assessed, a quality score and confidence code was assigned. Validation analysis revealed generally consistent results between raters with an average percentage agreement of 65 %, no scores falling outside of Bland-Altman limits of agreement and a moderate internal consistency α = 0.533 which was improved to 0.616. Applying the tool to 383 scientific studies showed that most literature obtained a quality score between 70 and 79 % resulting in a confidence code of "B" (n = 166, 43 %) and a score between 60 and 69 % resulting in a confidence code of "C" (n = 125, 33 %). The SADPDQAT can be used to evaluate a variety of dietary polyphenol containing literature and aids in setting priorities for further research in this area.