Type 1 diabetes risk is known to depend upon a genetic predisposition based on HLA types (particularly types DR3 and DR4), an unknown environmental trigger (suspected to be an infection, although none has proven definitive in all cases), and an uncontrolled autoimmune response that attacks the insulin producing beta cells. Some research has suggested that breastfeeding decreased the risk in later life; various other nutritional risk factors are being studied, but no firm evidence has been found. Giving children 2000 IU of Vitamin D during their first year of life is associated with reduced risk of type 1 diabetes, though the causal relationship is obscure.[37]
Children with antibodies to beta cell proteins (ie at early stages of an immune reaction to them) but no overt diabetes, and treated with vitamin B-3 (niacin), had less than half the diabetes onset incidence in a 7-year time span as did the general population, and an even lower incidence relative to those with antibodies as above, but who received no vitamin B3.
Type 2 diabetes risk can be reduced in many cases by making changes in diet and increasing physical activity. The American Diabetes Association (ADA) recommends maintaining a healthy weight, getting at least 2½ hours of exercise per week (several brisk sustained walks appear sufficient), having a modest fat intake, and eating sufficient fiber (e.g., from whole grains). The ADA does not recommend alcohol consumption as a preventive, but it is interesting to note that moderate alcohol intake may reduce the risk (though heavy consumption absolutely and clearly increases damage to bodily systems significantly); a similarly confused connection between low dose alcohol consumption and heart disease is termed the French Paradox.
There is inadequate evidence that eating foods of low glycemic index is clinically helpful despite recommendations and suggested diets emphasizing this approach.
Diets that are very low in saturated fats reduce the risk of becoming insulin resistant and diabetic. Study group participants whose "physical activity level and dietary, smoking, and alcohol habits were all in the low-risk group had an 82% lower incidence of diabetes." In another study of dietary practice and incidence of diabetes, "foods rich in vegetable oils, including non-hydrogenated margarines, nuts, and seeds, should replace foods rich in saturated fats from meats and fat-rich dairy products. Consumption of partially hydrogenated fats should be minimized."
There are numerous studies which suggest connections between some aspects of Type II diabetes with ingestion of certain foods or with some drugs. Some studies have shown delayed progression to diabetes in predisposed patients through prophylactic use of metformin, rosiglitazone, or valsartan. In patients on hydroxychloroquine for rheumatoid arthritis, incidence of diabetes was reduced by 77% though causal mechanisms are unclear. Breastfeeding may also be associated with the prevention of type 2 of the disease in mothers. Clear evidence for these and any of many other connections between foods and supplements and diabetes is sparse to date; none, despite secondary claims for (or against), is sufficiently well established to justify as a standard clinical approach.