Development of a Diet Quality score reflecting the 2017 French dietary guidelines

Paper of the month

October's Paper of the Month is from the British Journal of Nutrition and is entitled ‘Programme National Nutrition Santé – guidelines score 2 (PNNS-GS2): development and validation of a diet quality score reflecting the 2017 French dietary guidelines’ by Dan Chaltiel, Moufidath Adjibade, Valérie Deschamps, Mathilde Touvier, Serge Hercberg, Chantal Julia and Emmanuelle Kesse-Guyot.

Diet is considered a key lever for public health policies as it represents a changeable factor that may be modified through primary prevention interventions. Population dietary guidelines are thus developed and regularly updated according to the evolution of scientific knowledge in nutrition.

To quantify the potential impact of such guidelines, the association between adherence to recommendations and mortality and morbidity needs to be tested. A common, simple and efficient way to do so is by developing an index reflecting individual adherence to the recommendations.

In March 2017, French dietary guidelines were revised in order to account for the significant knowledge evolution since the previous guidelines issued in 2001. Therefore, it was necessary to also update the Programme National Nutrition Santé – guidelines score (PNNS-GS), the dietary index built upon the 2001 guidelines, to make it possible to evaluate the new recommendations.

The purpose of this study was thus to develop and to validate the PNNS-GS2, the dietary index reflecting the adherence of the 2017 French dietary guidelines. Score validation should follow a strict methodology and the development of the PNNS-GS2 was therefore further strengthened by the evaluation of its face validity (correctness of the score), content validity (completeness of the score), construct validity (association of the score to factors associated to nutrition quality) and criterion validity (correlation of the score to other comparable measures) validity.

The PNNS-GS2 includes 22 sub-components organized into 13 components, 6 of adequation and 7 of moderation. The development of the PNNS-GS2 was conducted in association with experts involved in the development of the French dietary guidelines, in order to best choose the weights and cut-offs of the score components. This makes the face and content validity rather satisfying.

Construct validity was tested in 80,965 participants of the NutriNet-Santé cohort, for whom a high PNNS-GS2 was significantly associated with sociodemographic, nutritional, clinical and biological markers of good nutrition quality, such as high education, high physical activity, non-smoking, being female, being older, having lower blood pressure and a higher intake of n-3 PUFAs, fibre or vitamin C. Criterion validity was harder to assess, as there is no consensual measure of diet quality, but it was approximated by measuring, in this sample, the correlation between PNNS-GS2 and mPNNS-GS (PNNS-GS without physical activity). The rather average value obtained was compatible with a significant improvement while measuring the same dimension.

In conclusion, the arguments in favour of a good validity of the PNNS-GS2 were coherent, which means that it is supposed to properly measure adherence to 2017 French dietary guidelines. For that reason, the PNNS-GS2 represents an important tool to evaluate how much following these guidelines is associated to a better health. This will be the subject of upcoming studies.