Nutritional factors and the risk of type 1 diabetes

Overall, the mother's diet during pregnancy does not appear to have a significant effect on the child's risk of type 1 diabetes [1]. Indirectly, though, a low level of vitamin D in the mother during the last trimester has sometimes been associated with a slightly higher risk of type 1 diabetes [2] [3]. However, the link has not been confirmed in all the data [1]. Correcting that level through supplements does not modify the risk of type 1 diabetes [4].

The likely explanation is that the higher risk does not come from the low vitamin D itself, but from the reason it was low. Changes in the pregnant woman's diet regarding the amount of gluten or the intake of omega-3 fatty acids did not significantly change the child's risk of type 1 diabetes [1] [5].

Yes, a birth weight over 4000 g is associated with an approximately 10% higher risk of type 1 diabetes. The risk rises linearly with birth weight, by 3% for every 500 g [6].

High birth weight is largely caused by an excess of insulin in the fetal circulation, in response to the abundance of available nutrients. The most widely discussed hypothesis is that overloading the pancreatic beta cells makes them more visible to the immune system, thereby raising the risk of autoimmunity being triggered after birth. The link between high birth weight and type 1 diabetes remains even when the mother's diabetes is taken into account, so it is not explained by it [6].

Yes, it matters. Both introduction in the first 3 months and introduction from 7 months onward were associated with a higher risk of autoimmunity, the stage that precedes type 1 diabetes [7]. The optimal window for introducing cereals is between 4 and 6 months, ideally while the child is still being breastfed. This advice coincides with the generally recommended timing for introducing solid foods [7].

In other words, it is not only what your baby eats that matters, but also when they start eating it. Introduced too early, cereals meet an immature gut, without the "protection" of breast milk. This may increase the risk of the immune system reacting in the wrong way [7].

Possibly, but the evidence is not firm. Introducing root vegetables (carrot, potato, celery root) before 4 months has been linked to a higher risk of advanced autoimmunity or type 1 diabetes [8]. With cereals and egg introduced early, the extra risk shows up afterwards, up to the age of 3, and then disappears [8].

Fruit introduced at 4–6 months, not earlier, is even linked to a lower risk of type 1 diabetes [1]. The recommendation is to start introducing solid foods between 4 and 6 months, without rushing and without specifically avoiding any healthy food [1].

Yes, they can hasten the onset, but they are not the cause of the disease. Sugar-sweetened drinks, like any other source of sugar, do not appear to be responsible for triggering the autoimmunity specific to type 1 diabetes. Once autoimmunity has already begun, however, sugar intake can bring forward the appearance of clinical disease (progression to stage 3). The effect is stronger in children at high genetic risk [9].

In children with autoimmunity specific to type 1 diabetes, sugar forces the pancreas to produce larger amounts of insulin. This extra demand stresses the beta cells and makes them "more visible" to the immune system, which then attacks them more quickly. Sugar-sweetened drinks should be avoided in the first year of life and limited afterwards, whatever the genetic risk [9].

The hypothesis that cow's milk proteins trigger type 1 diabetes was tested in a large dietary prevention trial and was not confirmed. Replacing cow's milk with a hydrolyzed-protein formula (infant formula milk) did not reduce the risk of type 1 diabetes [10].

Introducing cow's milk later (after the age of 2–3 months) appears to have a protective effect, as does prolonged breastfeeding. There is no point in avoiding cow's milk proteins in order to prevent type 1 diabetes. Extensively hydrolyzed formulas were tested for exactly this purpose and did not reduce the risk. Breastfeeding for as long as possible and delaying the introduction of cow's milk remain sound infant nutrition recommendations. Their effect on type 1 diabetes risk is at most small [10].

Yes. Children who gain weight faster than average have a higher risk of both autoimmunity and progression to clinical diabetes [11] [12] [13]. Obesity in childhood (assessed between 1 and 12 years of age) is linked to roughly twice the risk of type 1 diabetes [14].

The explanation is known as the "accelerator hypothesis". Excess weight makes the pancreas produce more insulin (for the whole body), which overloads the beta cells and hastens their recognition and destruction by the immune system. Keeping to a healthy weight in the first years of life might lower the risk, but this benefit has not yet been proven [15].

The total number of calories does not appear to matter directly for triggering autoimmunity. They do, however, start to matter more and more once the autoantibodies specific to type 1 diabetes have appeared. In general, the quality of the calories matters more than their quantity [9] [13] [16].

Indirectly, though, a calorie excess does matter, because it leads to weight gain when it is not matched by a corresponding energy expenditure. Rapid weight gain accelerates progression to clinical diabetes (stage 3). The chain of consequences starts with too many calories for the child's needs, which lead to weight gain and then to insulin resistance. The demand for insulin rises, the beta cells are overloaded, and their autoimmune destruction, very likely set off by something else, accelerates [13] [15].

Yes, but mainly for children who already have autoantibodies. High-glycemic-index foods (white bread, white rice, French fries, sweets, juices) do not appear to trigger autoimmunity. However, they significantly speed up the move from autoimmunity to clinical diabetes (stage 3) [16].

High-glycemic-index foods cause larger blood glucose peaks after meals, so the pancreas has to respond with a greater amount of insulin. This repeated pressure on the beta cells makes them more vulnerable to immune attack, even if that attack was initially set off by something else. Low-glycemic-index foods (legumes, whole grains, whole fruit instead of juice) are a good choice anyway, but their effect on progression to type 1 diabetes has not been proven [16].

Yes. In children who already have autoantibodies, a high sugar intake significantly increases the risk of developing clinical diabetes [9].

It is worth remembering that sugar does not trigger autoimmunity, but it accelerates it once it has appeared. In short, sugar does not "cause" type 1 diabetes, but it brings its onset forward if the process has already started. Limiting added sugar is recommended for all children, not only for those at risk of developing type 1 diabetes [9].

No. Although type 1 diabetes and celiac disease share some common genetic factors, avoiding gluten does not prevent diabetes. The only caveat is to introduce cereals to your baby between 4 and 6 months [7].

The prevalence of celiac disease in children with type 1 diabetes is nevertheless 5–6% (compared with 1% in the general population). This justifies periodic screening for celiac disease in all children with type 1 diabetes. There are no current recommendations to avoid gluten in order to prevent type 1 diabetes. A gluten-free diet is needed only if celiac disease is diagnosed.

Probably yes, but only in the phase when autoimmunity is being triggered, and especially through the marine forms. Children with a higher intake of omega-3 fatty acids (from oily fish, fish oil, cod liver oil) appear to have a significantly lower risk of developing autoantibodies. The protective mechanism probably involves reducing general inflammation and balancing the immune response [17].

Whatever protective effect omega-3 fatty acids may have, it disappears completely once autoimmunity has appeared. Once autoimmunity has begun, omega-3 fatty acids cannot slow progression to clinical diabetes. Including oily fish regularly (salmon, sardines, mackerel) in children's diet is a good idea anyway [17].

Fiber may have a positive indirect effect, through the gut bacteria. When fiber reaches the colon, the good bacteria ferment it and produce short-chain fatty acids (acetate and butyrate), which have anti-inflammatory effects and modulate (regulate) the immune system [18].

The microbiome is the collection of bacterial colonies in the gut. Healthy children carry more bacteria able to produce these fatty acids than children who go on to develop type 1 diabetes. However, simply increasing fiber intake is no guarantee of a favorable microbiome. The relationship between fiber, bacteria and immunity is complex. An adequate fiber intake is recommended for all patients, but not specifically to prevent type 1 diabetes [18].

The evidence is mixed and sometimes surprising. In children at genetic risk, a higher intake of saturated fat was associated with a lower risk of type 1 diabetes. This association did not appear in every analysis, however [19]. Higher levels of some saturated fats in infants' blood were also associated with a lower risk of autoimmunity [20].

These levels partly reflect the milk the infant receives, so they do not prove a direct effect of saturated fat [20]. At present, limiting saturated fat is recommended to prevent cardiovascular disease, but not specifically for type 1 diabetes [21].

Possibly. The incidence of type 1 diabetes was higher in areas with nitrates in the drinking water [22]. Children with a higher intake of nitrites from food (cold cuts, processed meat) had a higher risk of type 1 diabetes, but the certainty of these data is low [1].

There is biological plausibility. Streptozotocin, the compound used in the laboratory to induce diabetes in mice, is an N-nitroso compound similar to those found in processed meat. However, studies in humans are not yet sufficient for a firm conclusion. Limiting processed meat remains a general health recommendation [22].

Possibly, but evidence in humans is lacking. Emulsifiers are additives that bind water and fat in processed products. In mice, they altered the gut bacteria and caused intestinal inflammation [23]. Also in mice, foods rich in compounds formed at high temperatures (frying, charred grilling) greatly increased the risk of type 1 diabetes [24].

The presumed mechanisms include disruption of the gut bacteria, increased gut permeability ("leaky gut syndrome"), systemic inflammation and overloading of the beta cells. A diet based on foods that are as little processed as possible is advisable not only to prevent diabetes, but also for health in general. Prefer cooking at moderate temperatures (boiling, steaming, a medium-temperature oven) rather than frying or grilling [23] [24].

We don't know. No studies have directly examined whether a vegetarian diet prevents type 1 diabetes. Vegans have a lower risk of diabetes, but this mainly concerns type 2 diabetes, not type 1 [25].

In theory, a well-planned vegetarian diet could help by reducing inflammation, improving the gut microbiome and avoiding processed meat. A vegetarian diet is an accepted dietary pattern for people with diabetes, but it is not recommended specifically for prevention. If it is chosen, the diet must be varied and balanced, with attention paid to the intake of vitamin B12, iron and protein [25].

There is no evidence that the Mediterranean diet prevents the onset of type 1 diabetes. In children and adolescents who already have the disease, better adherence to this dietary pattern is associated with better glucose control. Glycated hemoglobin (HbA1c) is lower, and time in the target blood glucose range is longer [26] [27] [28].

The Mediterranean diet is recommended for people with diabetes, especially to reduce cardiovascular risk [21]. Its principles are based on:

• more vegetables and fruit;
• whole grains;
• fish (especially oily fish);
• extra virgin olive oil;
• nuts and seeds;
• moderate amounts of dairy;
• little red meat.

Zinc plays a central role in the pancreatic beta cells, and one of the major autoantibodies in type 1 diabetes attacks precisely the zinc transporter in these cells (ZnT8). However, there is no evidence that zinc supplementation prevents type 1 diabetes [29].

Vitamin C, obtained from natural sources, may have a positive effect. Children with a higher intake of vitamin C in their diet have a significantly lower risk of autoimmunity and of type 1 diabetes [1] [30]. A higher blood level of vitamin C was also associated with a lower risk of autoimmunity [31]. The practical answer is not supplements, but eating fruit and vegetables rich in vitamin C regularly, such as:

• bell peppers;
• citrus fruit;
• kiwi;
• strawberries;
• cabbage;
• fresh parsley.

Possibly, but not through supplements. A higher dietary intake of vitamin E is associated with a somewhat lower risk of autoimmunity and, possibly, of type 1 diabetes. The effect is modest [30].

However, supplementation with antioxidant vitamins (E, C, beta-carotene) is not recommended, because of the lack of clear evidence of benefit and some possible long-term safety concerns (especially for beta-carotene). The best source of antioxidants remains a varied diet, rich in fruit, vegetables, nuts and seeds [32].