Beyond nutrients: alternative approaches to analysing dietary intake data

Person examining the nutrition information on a box of food

Dr Kathryn BeckDr Kathryn Beck, Senior Lecturer in Human Nutrition and Dietetics, Massey University, New Zealand, and World Cancer Research Fund International (WCRF) Academy Fellow.

Reduced carbohydrates, rich in antioxidants, and low in fat – common claims on food packaging, and part of the marketing of some of the most popular diets in the world.

Traditionally, nutrition research has focused on individual nutrients, and associated health outcomes. While this approach is important (for example, to detect micronutrient deficiencies) it also has its limitations. People don’t eat nutrients in isolation, but as part of an overall dietary pattern containing meals and snacks. It is unlikely a single nutrient will cause a particular chronic disease.

In the past 20 or so years, nutrition research (and the analysis of dietary intake data) has moved towards assessing foods and the combinations of food we eat (dietary patterns). For example, healthy dietary patterns (those that are high in fruit, vegetables and wholegrains) derived using statistical methods or based on dietary indices tend to be associated with positive health outcomes.

New approaches to dietary intake analysis

Relatively recent approaches to the analysis of dietary intake data beyond nutrients were presented at the International Union of Nutritional Sciences (IUNS) 21st International Congress of Nutrition (ICN) in Buenos Aires on 15–20 October. Two examples of these methods focused on the categorisation of foods according to their level of processing; and on meal pattern analysis and the circadian rhythm across the day.

Man taking a sandwich from the fridge

 

Prof Carlos Monteiro from the University of Sao Paulo in Brazil and colleagues presented on the intake of ultra-processed food and nutrient profiles in the diet. Using the NOVA classification, foods were grouped according to the extent and purpose of processing they undergo. The four groups include unprocessed or minimally processed foods; processed culinary ingredients; processed foods and ultra-processed foods.

Nutrition datasets from seven countries (Chile, Colombia, US, Canada, Australia, UK and Brazil) were explored. The contribution of ultra-processed foods to total energy intake ranged from 15.9% in Colombia to 57.9% in the US. In Brazil, an increased intake of ultra-processed foods was associated with higher intakes of total fat, saturated fat, trans fat, free sugars and diet energy density; and a decreased intake of protein and diet fibre density. Similar findings were observed in the other countries.

Prof Heiner Boeing from the German Institute of Human Nutrition, Potsdam-Rehbrucke, presented his group’s work on the analysis of meal patterns. His research group focuses on the circadian rhythm of dietary intake (known as nutrichronobiology), and how this is associated with different metabolic activity throughout the day.

For example, are foods eaten together at a particular meal and time of day (eg breakfast) associated with a metabolism following a circadian rhythm? This work also involves exploring how habitual dietary patterns relate to meal patterns and distributions across the day using data from 24-hour recalls.

Assessing overall patterns

These are just some examples of recent developments in the analysis of dietary intake data. Assessing overall dietary patterns and how foods are eaten in combination, focusing on levels of food processing, and the distribution of food intake over the day offer alternative and complementary approaches to traditional nutrient-based analysis. These methodologies provide the potential for further insights into understanding the food we eat and its influence on health outcomes.