A novel approach to assess diet diversity: a development of the Nutritional Functional Diversity indicator

Matteo Di Maso; Francesca Bravi; Jerry Polesel; Luigino Dal Maso; Linia Patel; Carlo La Vecchia; Monica Ferraroni

doi:10.3389/fnut.2023.1170831

A novel approach to assess diet diversity: a development of the Nutritional Functional Diversity indicator

Front Nutr. 2023 Oct 19:10:1170831. doi: 10.3389/fnut.2023.1170831. eCollection 2023.

Authors

Matteo Di Maso¹, Francesca Bravi¹, Jerry Polesel², Luigino Dal Maso², Linia Patel¹, Carlo La Vecchia¹, Monica Ferraroni^{1

3}

Affiliations

¹ Department of Clinical Sciences and Community Health, Branch of Medical Statistics, Biometry and Epidemiology "G.A. Maccacaro", Università degli Studi di Milano, Milan, Italy.
² Unit of Cancer Epidemiology, Centro di Riferimento Oncologico di Aviano (CRO), IRCCS, Aviano, Italy.
³ Fondazione IRCCS Ca' Granda, Ospedale Maggiore Policlinico, Milan, Italy.

Abstract

Background and aims: Diversity is a key element of diet quality. The Food Variety Score (FVS) is used to assess diversity, especially in low- and middle-income countries. It sums up the number of foods consumed ignoring their nutrient content. A more suitable index should combine the number of foods consumed and their nutritional composition. We adapted the Nutritional Functional Diversity indicator (NFD), proposed by ecologists, to measure diversity in the human diet. We compared NFD and FVS evaluating subjects' distributions across quartiles of the two diversity indices. To evaluate which one reflected a higher diet quality, we estimated associations between these two diversity indices and diet quality measures, i.e., the Mediterranean Diet Score (MDS) and the Healthy Eating Index-2015 (HEI-2015). Associations were expressed by odds ratios (OR) and corresponding 95% confidence intervals (CI).

Materials and methods: We used the data of controls only derived from an integrated series of hospital-based case-control cancer studies conducted in different Italian areas. The NFD identifies groups of foods based on a set of nutrients according to a cluster analysis. Some steps are required: creating a food-nutrient matrix; clustering of the Euclidean food-food distance matrix to identify groups of foods with nutritional (dis)similarities; and calculating the NFD as the ratio between the sum of branch lengths of the dendrogram belonging to the number of foods consumed by individuals (i.e., subject-specific diversity) and the sum of all branch lengths of the dendrogram (i.e., maximal diversity).

Results: More than one quarter of individuals (28.4%) were differently classified within quartiles of the two diversity indices. For both indices, increasing the diversity level increased the risk for adhering to MDS (OR for NFD = 11.26; 95% CI: 7.88-16.09, and OR for FVS = 6.80; 95% CI: 4.84-9.54) and to HEI-2015 (OR for NDF = 2.86; 95% CI: 2.39-3.42, and OR for FVS = 2.72; 95% CI: 2.27-3.26). Associations were stronger for NFD.

Conclusion: Our findings showed a greater ability of NFD to assess diet quality quantifying the degree of diversity.

Keywords: Healthy Eating Index-2015; Mediterranean diet; cluster analysis; diet diversity; diet quality; novel approach; nutritional composition.