The NutriNet-Santé study is an ongoing investigation into the relationship between nutrition and health. In this Special Feature, we look at some of the project’s findings and speak with Principal Investigator Dr. Mathilde Touvier, who has been involved in the study since its inception.

People buying vegetables at a marketShare on Pinterest
Ludovic Marin/AFP via Getty Images

A range of factors influences health, including genetics, lifestyle, environmental factors, and diet. Unraveling the complex relationships between these factors is a challenge.

For many reasons, it is incredibly difficult to investigate the role of nutrition in health and disease. For instance, no two people eat the exact same diet, and very few people eat the exact same food 2 days in a row.

As it is neither feasible nor ethical to ask thousands of people to follow a strict diet for 10 years to see what happens, researchers have to find other ways of unpicking the links between diet and disease.

The best way to tackle any difficult health-related question is to generate as much good quality data as possible, and this is the NutriNet-Santé study’s raison d’etre.

Beginning in 2009, the NutriNet-Santé study was the first internet-based study of its kind. By the start of 2021, the team was regularly collecting data from 171,000 people aged 15 years and older, making it the largest ongoing nutrition study in the world.

Now running in France and Belgium, the team is also seeding similar projects in Canada, Mexico, and Brazil.

Specifically, the researchers set out with the following aims:

  1. Investigate the relationship between nutrition, health, lifestyle factors, and mortality.
  2. Examine the factors that influence dietary patterns, such as economic and cultural factors.

The researchers keep a biobank of serum, plasma, and urine from about 20,000 people. They also collect stool to monitor and analyze gut bacteria.

Alongside questions about food intake, the NutriNet-Santé team collects information about food packaging, cooking practices, mode of production, physical activity, tobacco, drugs, environmental factors, and domestic and professional exposures.

Importantly, data from the NutriNet-Santé study are linked with medical and insurance records to improve accuracy regarding medications, diagnoses, and long-term sick leave. The study is financed entirely by public institutions.

Article highlights:

In recent years, ultra-processed foods have become a nutritional pariah, and the NutriNet-Santé study has played no small part in this.

Over recent years, data from the NutriNet-Santé study have revealed associations between diets high in ultra-processed foods and an increased risk of cancer, cardiovascular disease, mortality, depressive symptoms, type 2 diabetes, obesity, and gastrointestinal disorders.

As an example, one paper based on data from the NutriNet-Santé cohort, which appeared in the BMJ in 2018, concluded:

“In this large prospective study, a 10% increase in the proportion of ultra-processed foods in the diet was associated with a significant increase of greater than 10% in risks of overall and breast cancer.”

Another study using their data, which also appeared in BMJ, concludes:

“[H]igher consumption of ultra-processed foods was associated with higher risks of cardiovascular, coronary heart, and cerebrovascular diseases.”

Yet more research, which appeared in BMC Medicine in 2019, investigated ultra-processed foods and their links with depression. The authors write:

“Overall, [ultra-processed food] consumption was positively associated with the risk of incident depressive symptoms, suggesting that accounting for this non-nutritional aspect of the diet could be important for mental health promotion.”

Next, the NutriNet-Santé researchers plan to investigate the causative aspects of this relationship. They will drill down into the specific compounds that might be responsible.

According to Dr. Touvier, thanks to the researchers’ wealth of nutrition data, they can now “estimate the additive exposure for more than 330 additives authorized in Europe and also detect the mixtures of additives that are consumed by the population.”

They have already identified these cocktails of additives and are now searching for relationships between particular mixtures of compounds and specific disease states.

Share on Pinterest
An example of how the Nutri-Score appears on packaging.

Understanding which foods are associated with which health conditions is important, but the next step — changing behavior — can be even more challenging.

To address this, the NutriNet-Santé study focused on food labeling. Although product labels already provide information about levels of fat, sugar, and other ingredients, as Dr. Touvier pointed out, quickly gauging whether a product is healthful as you rush around a grocery store is not easy.

With this in mind, the NutriNet-Santé group designed Nutri-Score. This simple label gives each product a score from A to E, with A being the most healthful and E the least healthful.

The scoring system takes into account the amount of sugar, saturated fats, sodium, protein, fiber, fruits, vegetables, nuts, and legumes to provide a score.

“The score has been validated against many health outcomes, using the NutriNet-Santé cohort [and other cohorts]. So we showed that people who ate foods with a better score had less risk of cancers, cardiovascular disease, obesity, and so on. We also validated it in other independent cohorts,” explained Dr. Touvier.

“We also used the NutriNet-Santé cohort to validate how the score is understood and used by participants to rate nutritional quality.”

And, even more importantly, the research has shown that in a grocery setting, people with access to this type of labeling choose more healthful foods. As the authors of one paper explain:

“The Nutri-Score was associated with a higher nutritional quality of purchases in experimental and large scale trials.”

In France, some companies are already applying this label on a voluntary basis. To date, 520 firms, including a total of 690 brands, have registered to use the scoring system.

Dr. Touvier and the team are currently trying to convince the European Union to roll out the scoring system throughout the E.U. in 2022. However, they are facing stiff resistance from certain large food corporations and their lobbying groups.

More than 40 published studies back the benefits of the Nutri-Score. Dr. Touvier hopes that “it will be adopted because, for the consumer, we really showed it will have a great impact.”

Over the years, there has been a fair amount of controversy around the benefits of eating organic produce. As Dr. Touvier explained, until recently, there were simply not enough data to make connections between these products and health outcomes.

Once again, the NutriNet-Santé study has begun to fill these gaps. According to Dr. Touvier, the team found “an association between higher concentrations of organic food and a lower risk of breast cancer and lymphoma.” Similarly, the data “showed a lower risk of obesity, overweight, and metabolic syndrome.”

The researchers have also demonstrated that people who eat more organic foods have “lower levels of pesticides in their urine.” Next, they plan to quantify exposures to the various types of pesticides and identify “cocktails” of pesticide exposures. They are in the process of assessing whether certain pesticidal cocktails might be associated with specific health outcomes.

According to Dr. Touvier, one such study, which has just been accepted for publication in the International Journal of Epidemiology, found that:

“People exposed to these cocktails of pesticides had a higher risk of postmenopausal breast cancer.”

As the COVID-19 pandemic continues, so does NutriNet-Santé’s data collection. The scientists are tracking how diets have shifted during these unusual times.

One study, for instance, showed that the pandemic has affected people differently, with some eating less healthfully and others improving their diet significantly.

“We saw a very diverse profile of participants. Some of them started to eat more sugary foods and to eat between meals because they were bored or stressed. Also, they lowered the level of physical activity. […] For these participants, we saw an increase in body weight,” explained Dr. Touvier. But the pandemic has had different effects on other people, she noted:

“Other participants began to cook more at home, so less processed foods and more homemade meals, and they increased their level of physical activity […] These participants lost weight.”

The NutriNet-Santé scientists also found that there were people who had made no change to their diet or level of activity.

As the researchers have also collected dried blood spots from more than 20,000 participants, they are able to detect anti-SARS-CoV-2 antibodies.

They are currently working on a way to link infection to diet. This research could reveal whether any dietary patterns are associated with an increased risk of SARS-CoV-2 infection or, conversely, whether any nutritional profiles are associated with a protective effect.

Investigating nutrition is challenging. Every individual eats hundreds or thousands of ingredients each week, and no two diets are the same. Also, when scientists rely on self-reported dietary information, they are likely to get imprecise answers.

The NutriNet-Santé study was the first to use online questionnaires to collect data. As trailblazers for this methodology, the researchers spent a great deal of energy making sure that the data they were collecting were accurate and meaningful.

Medical News Today asked Dr. Touvier how the NutriNet-Santé study team went about this. She explained that early on in the study, the investigators ran a series of validation studies in which they checked participants’ self-reported food intake against their “blood and urinary biomarkers” and their phone interviews with trained dietitians.

They found a “high correlation” between the information coming from participants, blood markers, and interview data.

In some cases, the data they collected were more reliable than the phone interviews. Dr. Touvier explained that some participants were more likely to enter unhealthful food in the anonymous online form than they were to mention it when speaking with a dietitian. She said:

“It’s not perfect. It’s still an observational study, but we have validated all of these methods, so they are reliable.”

Another common issue that nutrition scientists face is that one brand of frozen pizza is not the same as another, and one bag of chips can harbor entirely different ingredients than another.

To help minimize this issue, the researchers designed a barcode reader that allows participants to scan their food items directly into the system. In this way, the scientists have direct access to the precise nutritional content of the food item.

Another significant issue with nutrition research is the role of industry bias. Studies that receive funding from interested parties are more likely to report the benefits of their produce. The NutriNet-Santé study, however, is publicly funded, and the data are not open access.

As Dr. Touvier explained, “we have a moral agreement with the participants not to provide the data to the food industry.” However, the team does “share their data with other public researchers.”

Correlation versus causation is the thorn in any observational study’s heel. Finding an association is one thing, but proving that one factor directly causes another is a different challenge altogether. To ensure that the researchers chop away as many confounding factors as possible, they track a great number of variables.

For instance, they keep anthropometric data, such as height and weight; socioeconomic data; information about participants’ occupations so that they can track potential environmental exposures to chemicals, for instance; medical history; current and past medications; activity levels; and more.

The NutriNet-Santé scientists are beginning to dive even deeper into the complexities of nutrition and health. Beyond the nutritional makeup of foods, they are beginning to investigate how the packaging might interact with food items and influence health. Their innovative barcode system helps them collate this information.

These data will also help with investigating sustainability, which is another area on which the team is beginning to focus.

The wealth of data that the NutriNet-Santé team has collected and continues to collect will slowly allow scientists to answer increasingly complex questions.

As the number of participants in the study steadily grows, Dr. Touvier hopes that the NutriNet-Santé team can “continue indefinitely.” Undoubtedly, the insights that this team generates will have a beneficial effect on the health of future generations.