Press releases

The consumption of antioxidant-rich foods is associated with a lower risk of type 2 diabetes

10 Nov 2017 | By INSERM (Newsroom) | Public health

A lower risk of type 2 diabetes has been observed among individuals consuming food rich in antioxidants. This effect is largely contributed by fruit, vegetables, tea and other hot beverages, as well as moderate consumption of alcohol, as shown in a recent study from an Inserm research group, published in Diabetologia, the journal of the European Association for the Study of Diabetes (EASD)

A diet rich in fruit and vegetables has previously been associated with a lower risk of certain cancers and cardiovascular conditions. An Inserm team (Health across generations, Center of Research in Epidemiology and Population Health, Villejuif, France) has now shown that such a diet is similarly associated with a reduced risk of type 2 diabetes.

The team already suspected there might be a link on the basis of previous studies showing that certain antioxidants, such as vitamins C and E, lycophenes or flavonoids, were associated with a reduction in type 2 diabetes risk. However, these studies looked only at isolated nutrients, not at the total antioxidant capacity of the diet. The researchers therefore wanted to verify whether overall diet, according to its antioxidant capacity, is associated with diabetes risk. Using data from the E3N cohort comprising French women recruited from 1990, then aged between 40 and 65 years, they followed 64,223 women from 1993 to 2008, all of whom were free from diabetes and cardiovascular disease at the time of inclusion in the study. Each participant completed a dietary questionnaire at the beginning of the study, including detailed information on more than 200 different food items. Using this information, together with an Italian database providing the antioxidant capacity of a large number of different foods, the Inserm researchers calculated a score for ‘total dietary antioxidant capacity’ for each participant. The group then analysed the associations between this score and the risk of diabetes occurrence during the follow-up period.

The results show that diabetes risk diminished with increased antioxidant consumption up to a level of 15 mmol/day, above which the effect reached a plateau. Increasing dietary antioxidants to this level could be achieved through eating antioxidant-rich foods such as dark chocolate, tea, walnuts, prunes, blueberries, strawberries or hazelnuts, to name just a few.

Women with the highest antioxidant scores had a reduction in diabetes risk of 27% compared with those with the lowest scores. ‘This link persists after taking into account all the other principal diabetes risk factors: smoking, education level, hypertension, high cholesterol levels, family history of diabetes and, above all, BMI, the most important factor’, clarifies Francesca Romana Mancini, the first author of this study. The foods and drinks that contributed the most to a high dietary antioxidant score were fruits and vegetables, tea and red wine (consumed in moderate quantities). The authors excluded coffee from the analysis, despite its high antioxidant levels, because the antioxidants in coffee have already been shown to be associated with reduced type 2 diabetes risk, and might therefore mask the effects of antioxidants from other sources.

‘This work complements our current knowledge of the effect of isolated foods and nutrients, and provides a more comprehensive view of the relationship between food and type 2 diabetes’ explains Guy Fagherazzi, the lead researcher in charge of diabetes research in the E3N study. ‘We have shown that an increased intake of antioxidants can contribute to a reduction in diabetes risk’. This now raises the question why: ‘We know that these molecules counterbalance the effect of free radicals, which are damaging to cells, but there are likely to be more specific actions in addition to this, for example an effect on the sensitivity of cells to insulin. This will need to be confirmed in future studies’, concludes Francesca Romana Mancini.

The E3N study (; Etude Epidémiologique auprès de femmes de la MGEN [Mutuelle Générale de l’Education Nationale]) is a prospective cohort study of around 100,000 French female volunteers born between 1925 and 1950, and followed from 1990. Since 1990, the women have completed and returned self-administered questionnaires every 2–3 years. The questionnaires cover their lifestyle and their  health status over time. The ‘lost to follow-up’ rate is very low because the MGEN is able to assist in following up non-responders.  The E3N study is supported by four founding partners: Inserm, la Ligue contre le Cancer, l’Institut Gustave Roussy and the MGEN. The E3N study is now continued via the E4N study (, a study awarded the status of  ‘Investment for the future’ (Investissement d’Avenir) in 2011 by the French National Research Agency (Agence Nationale de Recherche), and which aims to follow the children and grandchildren of the E3N participants, as well as the biological fathers of the children, with a view to studying the health of three generations of family members in relation to modern lifestyle.
Researcher Contact
Guy Fagherazzi Unité Inserm 1018 "Centre de Recherche en Epidémiologie et Santé des populations" Equipe “Générations et Santé” Email : rf.yssuorevatsug@izzarehgaf.yug Tel : (0)1 42 11 63 74  
Dietary Antioxidant Capacity and Risk of Type 2 diabetes in the large prospective E3N-EPIC cohort study. Francesca Romana Mancini,1,2,3 Aurélie Affret,1,2,3 Courtney Dow,1,2,3 Beverley Balkau,1,2,4, Fabrice Bonnet,1,2,5 Marie-Christine Boutron-Ruault,1,2,3 Guy Fagherazzi1,2,3 1: INSERM U1018, Center for Research in Epidemiology and Population Health, Villejuif,France 2: University Paris-Saclay, University Paris-Sud, Villejuif, France 3: Gustave Roussy, F-94805, Villejuif, France 4: University Versailles, Saint Quentin, University Paris-Sud, Villejuif, France 5: CHU Rennes, Université de Rennes 1, France Diabetologia