Nutritional researchers have a tough job getting good results from studies. Ruth Paget talks to Dr Nadia Slimani to find out why
Getting definitive answers about food can be difficult for research scientists. Diet is one of the most complex environmental factors to investigate in relation to chronic disease, in particular cancer. Estimating what people eat, and in what amounts, is an ongoing challenge for research scientists.
One of the largest studies of diet and disease is EPIC – the European Prospective Investigation into Cancer and Nutrition. This study looked at the association between eating certain foods and the risk of getting diseases such as cancer, obesity, diabetes, stroke and heart disease. More than half a million people in ten different countries are involved.
Dr Nadia Slimani, Head of the Dietary Exposure Assessment Group at the International Agency for Research on Cancer, has worked on the EPIC study since it began in the early 1990s.
Studying people’s diets
“Nutritional studies have different measurement errors, which are not only due to the individual, but also to the complexities of diet from day to day, across individuals and over time, making an accurate assessment of an individual’s dietary pattern difficult to capture,” says Dr Slimani.
“The nature of diets implies complex interactions between environmental factors, social, psychological and cultural components, and of course the individual person’s unique genetic makeup,” she adds.
To try get reliable results from nutrition studies, researchers estimate what you have eaten in the previous year and account for seasonal and personal variations.
Dr Slimani says that in the EPIC studies, there is an integrated approach using the food frequency questionnaire (where the foods eaten in the previous year are estimated) combined with open-ended recall diaries to refine dietary estimates from the questionnaires.
As well as questionnaires and diaries, researchers can look at the level of chemicals in body fluids to investigate what people have eaten. These chemicals are known as biomarkers. If you eat more proteins, for example, more nitrogen is found in your urine. Sodium levels in your blood go up if you eat more salt.
Complex nutrition databases that contain information on each type of food (including its nutritional components) and software designed for nutrition studies can be used to analyse the different parts of the diet and to link and compare people and populations.
Being honest about your habits
So why is it hard to get people to report what they’ve eaten? Being conscious of your actions can in fact make you change what you do normally. Telling someone what you eat can also seem highly personal, as food can have an emotional and psychological link.
Dr Slimani says that in the EPIC study, “when the 24-hour recall method was applied, we try to interview people without informing them in advance. We then ask them to recall what they have eaten the day before, using different probing questions and quality checks during the interview to minimise errors, forgotten items and inconsistent responses.”
Certain inconsistencies, where people say they eat less or eat better than they actually do, have been shown in a number of scientific studies. They have been particularly noted in people who are overweight or obese, who record they eat less than they do. Some studies have also seen such misreporting in women in some countries, who say they eat more healthily than they actually do.
However, there might be a rational explanation for an individual misreporting, particularly on a single day, points out Dr Slimani. “They might be travelling, have been at a party, or been sick – there are many reasons why someone on that particular day appeared to be misreporting, and this is classed as the natural day-to-day variation in an individual’s diet.”
Trained interviewers are guided by in-depth instructions and use specifically designed computer programs to encourage honest reporting. Biomarkers can help researchers account for any unusual results, too. They can also check how much energy a person uses by using a method called doubly labelled water – where you ingest or have injected a harmless isotope that can be measured to show energy expenditure.
By studying large numbers of people in different countries with different diets, EPIC addresses a large variety of diet-related questions.
There has been a change to diets since the second half of the 20th century, says Dr Slimani, and researchers have seen an increasing contribution of industrial (processed) foods and the disappearance of the more traditional diet in several European countries. This change in nutrition raises a series of new challenges for the scientists on how to monitor and measure these changes and how to investigate their roles in relation to diseases, particularly cancer.
“EPIC has been able to show associations between diet and cancer, as, for example, for red meat and more specifically processed meat and colorectal cancer, which had not been reported so strongly before. But no link between the consumption of fish or white meat [eg chicken] and this type of cancer was observed. These results, evaluated with others by the World Cancer Research Fund expert panel, contributed to the conclusion that there is strong evidence that red/processed meat is a risk factor in the development of colorectal cancer,” says Dr Slimani.
Dr Slimani outlines that future research should aim to integrate better high-quality dietary information, including biomarker measurements, genetics and novel statistical methods, to provide new insights into diet–disease risk associations.
All the pieces of evidence found in nutritional epidemiological studies are pieces of a larger puzzle, explains Dr Slimani, to be used alongside the contribution from experimental studies, clinical settings and intervention studies, which will help to confirm that a particular food, component in food or dietary pattern actually has a protective or harmful effect on a disease.Lead image:
- EPIC – European Prospective Investigation into Cancer and Nutrition
- What foods do people habitually eat? A dilemma for nutrition, an enigma for psychology (2000)
- Under- and overreporting of energy is related to obesity, lifestyle factors and food group intakes in Jamaican adults (2004)