Diabetes is a condition caused by the pancreas failing to produce the hormone insulin or producing insufficient quantities. Another cause is insulin resistance - the body cells’ inability to react to insulin. Insulin is a hormone produced by the pancreas and acts as a key, allowing glucose into the body’s cells. Glucose is a vital source of energy for cells and is the main fuel for the brain and body’s processes. In diabetics, blood sugar levels rise too high and this can damage blood vessels and cause nerve damage but it also has a negative effect on health overall, increasing cholesterol levels and the risk of heart disease.

Type 1 diabetes

Type 1 diabetes is less common and typically develops early in life when the immune system attacks the insulin-producing cells in the pancreas and destroys them. It results in the body being unable to produce insulin and therefore use glucose. Type 1 diabetics have to inject insulin on a regular basis.

Evidence is increasing that a combination of susceptible genes and early exposure to cow’s milk is responsible for this self-harming reaction of the body. It might be also triggered by a virus or other infection. Several gene variants have been identified as contributing to type 1 diabetes susceptibility but only a small proportion of genetically susceptible individuals – less than 10 per cent – go on to develop the disease (Knip et al., 2005). This implies that environmental factors are necessary to trigger the autoimmune reaction which destroys insulin producing cells.

If an individual has a certain combination of genes making them more susceptible to type 1 diabetes, the environmental trigger is the crucial factor in the disease development but if the trigger is avoided, diabetes may be avoided. The hypothesis that cow’s milk is the main trigger was put forward in the 1990’s (Karjalainen et al., 1992; Gerstein, 1994; Ǻkerblom and Knip, 1998) and has been progressively more accepted ever since. The theory is as follows (Campbell and Campbell, 2004; Knip et al., 2005):

A baby with a susceptible genetic make-up is exposed to cows’ milk early in life, perhaps through an infant formula. The baby’s immune system might be further compromised by a virus infection, increasing the risk for type 1 diabetes. When the milk proteins reach the small intestine they are not fully digested – ie broken down into individual amino acids – but are instead broken down into amino acid chains. These fragments may be absorbed through the gut wall into the blood where the immune system recognises them as foreign intruders and begins attacking them through an immune response. Coincidentally, the structure of some of these fragments is identical to the surface structure of insulin producing cells (β-cells) in the pancreas (Karjalainen et al., 1992; Martin et al., 1991) and the body cannot distinguish between the two. Pancreas β-cells are therefore attacked and eventually destroyed by the immune system as well as the milk protein fragments and the infant becomes diabetic. Type 1 diabetes is irreversible as the cells cannot regenerate.

The process of β-cell destruction can be fast and aggressive, leading to disease manifestation within a few months, or it can be slow and last for years, in some cases even more than 10 years with β-cells being gradually destroyed over this period with each exposure (Knip et al., 2005). However, the fast progression of the disease is rare (Knip, 2002).

Research has established which milk proteins are responsible for this dramatic autoimmune reaction. Karjalainen et al. (1992) suggest that the main one is bovine serum albumin (BSA), which is different in structure to human albumin (milk protein). They tested the blood of type 1 diabetic and non-diabetic children for the presence of antibodies against incompletely digested BSA. The results were astonishing - all diabetic children had antibody levels as much as seven times higher than the healthy children and there was no overlap in the antibody levels between the diabetic and healthy children – ie all diabetic children had high levels but none of the non-diabetic children did.

After that, a number of studies ensued and all but one found markedly elevated levels of BSA antibodies in the blood of diabetic children (Hammond-McKibben and Dosch, 1997).

Another protein abundant in cow’s milk is β-casein, which also generates a specific immune response (Cavallo et al., 1996). The structure of human β-casein is similar in many respects to bovine β-casein (from cow’s milk) but 30 per cent of the molecule is different in structure. This difference is assumed to be the reason why the immune system reacts to it. Again there are structural similarities between bovine β-casein and the surface molecules of β-cells in the pancreas, just as there is with BSA, provoking an immunological cross-reactivity - the immune system attacks β-casein molecules as well as the β-cells.

A Chilean study conducted around the same time focused on the combination of susceptible genes and cow’s milk (Perez-Bravo et al., 1996). The findings revealed that genetically susceptible children weaned too early onto cow’s milk-based formula had 13 times greater risk of developing type 1 diabetes than children breast-fed for at least three months and who did not have susceptible genes.

In 2000, an extensive study of children from 40 different countries confirmed a link between diet and type 1 diabetes (Muntoni et al., 2000). The study set out to examine the relationship between dietary energy from major food groups and type 1 diabetes. Energy intake per se was not associated with type 1 diabetes but energy from animal sources (meat and dairy foods) showed a significant association whereas energy from plant sources was inversely associated with diabetes. In other words, the more meat and milk in the diet, the higher the incidence of diabetes and the more plant-based foods in the diet, the lower the incidence.

In the meantime, it was discovered that there are five autoantibodies - antibodies which will attack their own host body – and the presence of these autoantibodies can predict the development of type 1 diabetes (Knip, 2002). In addition to the two which attack β-cells, together with two supporting antibodies, there is one which will attack insulin itself. It was suggested that cow’s insulin present in formula milk increases the formation of these insulin antibodies (Vaarala et al., 1999). A Finnish study of children at increased risk of type 1 diabetes (having at least one close relative with the disease) showed that the immune system of infants given cow’s milk formula as early as three-months old, reacted strongly to cow’s insulin by forming these specific antibodies (Paronen et al., 2000).

Results of another study following infants from birth (Ǻkerblom et al., 2002) showed that exclusively breastfeeding for only a short period followed by the introduction of cow’s milk, predisposed these children to β-cell-destroying autoimmune reactions by inducing formation of four culprit autoantibodies. Other population studies have shown that if three or four of these antibodies are present in blood, the risk of developing type 1 diabetes in the next five to 10 years is 60-100 per cent (Knip et al., 2005).

A recent study showed that introducing cow's milk into the diet of infants between six and 12 months of age increased four-fold the likelihood of developing type 1 diabetes later in childhood (Villagrán-García et al., 2015). And another study examining the link between diabetes and cow’s milk warned that early introduction of cow’s milk is a risk factor for type 1 diabetes (Kamal Alanani and  Alsulaimani, 2013).

There is enough evidence linking cow’s milk consumption to type 1 diabetes that it’s advisable not to consume it or feed it to children. If the autoimmune reaction is triggered, it’s irreversible and type 1 diabetes has to be managed with insulin and possibly other drugs throughout life. A plant-based diet can provide all nutrients even for infants and children and can help prevent autoimmune reactions.

Type 2 diabetes

In type 2 diabetes, the body can still make some insulin but not enough or it fails to react to insulin as it should (insulin resistance). Lifestyle and environmental factors play an enormous role in type 2 diabetes. Therefore, even individuals with susceptible genes, or people who have already developed type 2 diabetes, don’t necessarily have to live with the condition for the rest of their lives.

Comparing diets and diabetes rates in different countries reveals that as carbohydrate intake goes down and fat intake goes up, the number of diabetics rapidly increases (Campbell & Campbell, 2004; Barnard, 2007). The difference cannot be ascribed to genetics as when people move to countries where the ‘Western’ style diet predominates and they adopt these eating habits, their rates of type 2 diabetes increase above the national average (Tsunehara et al., 1990).

Research suggests that eating just one serving of meat per week significantly increases the risk of diabetes. A study looked at the link between meat intake and the occurrence of diabetes in 8,000 adult Seventh Day Adventists, all of whom were non-diabetic at the start of the study (Vang et al., 2008). Those who followed a ‘low-meat’ diet over the 17 years of this long-term study had a staggering 74 per cent increase in their risk of developing type 2 diabetes compared to participants who followed a meat-free diet for the same period. Part of this difference was attributable to obesity and/or weight gain but even after allowances were made for this, meat intake remained an important risk factor.

A study published in 2004 produced an outstanding discovery (Petersen et al., 2004) and confirmed the findings of previous studies (Phillips et al., 1996; Krssak et al., 1999). The researchers tested young healthy adults, whose family members were diabetic, for insulin resistance. Those who tested positive were found to have microscopic drops of fat in their muscle cells. This fat interfered with the cells’ ability to correctly react to insulin. Even though their bodies produced sufficient insulin, fat inside their cells inhibited the correct reactions.

Muscle cells normally store small quantities of fat as an energy reserve but in the insulin-resistant people fat had built up to levels which were 80 per cent higher than in other young (healthy) people. Even though the affected people were slim, fat had nevertheless accumulated in their cells. The fat particles were intramyocellular lipids which start accumulating many years before type 2 diabetes manifests. It was then confirmed by other studies that insulin resistance in muscle and liver cells is strongly linked to fat storage in these tissues (Delarue and Magnan, 2007; Morino et al., 2006).

In order to understand the extent to which diet influences intracellular fat metabolism, another study was conducted (Sparks et al., 2005). Young healthy men were put on a special, high-fat diet that drew 50 per cent of its calories from fat – a diet not too different from that which many people in Western countries consume. After just three days, intracellular lipids had increased considerably, showing that accumulation of fat inside cells is fast and diet-dependant.

A study of cell metabolism in relation to insulin resistance revealed that elevated fat levels in blood and/or intramuscular fat accumulation can cause reduction in mitochondrial function which is crucial for insulin sensitivity (Hoeks, et al., 2010). Mitochondria are in every cell and they perform many metabolic functions, including taking part in sugar and fat metabolism.

Based on the above, Gojda et al. (2013) studied insulin sensitivity, intramyocellular lipids and mitochondria in healthy vegans and omnivores without close relatives with diabetes. The study participants were matched for age, weight, height and waist circumference. The tests discovered that vegans had higher insulin sensitivity, less intromyocellular lipids and slightly more mitochondria. Their glucose metabolism was overall more efficient and sensitive than in the other group. Vegans also had significantly lower cholesterol levels and their blood lipid profile was much better than that of omnivores.

Research shows that with the right diet it is possible to decrease blood sugar, limit medication, cut the risk of complications and even reverse type 2 diabetes.

One of the first studies to test the effects of a plant-based, low-fat diet and exercise on a group of 40 type 2 diabetic patients, had outstanding results - 36 of the patients were able to discontinue all medication after only 26 days (Barnard et al., 1982). The same research group later demonstrated that the benefits of this diet are long-term and last for years, if the diet is adhered to (Barnard et al., 1983).

In one of the groundbreaking studies that followed, researchers employed a combination of diet change and exercise (Barnard et al., 1994). The subjects were 197 men with type 2 diabetes and after just three weeks of a low-fat plant-based diet, 140 of them were able to discontinue their medication.

A study with a different angle involved 21 diabetics with diabetic neuropathy (characterised by numbness and shooting or burning pains in the lower limbs), who volunteered to follow a vegan, whole food diet and exercise programme for 25 days (Crane and Sample, 1994). Within 16 days, 17 of the patients reported that the neuropathic pain had been completely alleviated. Although the numbness persisted, it was noticeably improved within the 25 days of the programme.

And a new study on the same subject agrees that a low-fat vegan diet can reduce diabetic neuropathy (Bunner et al., 2015). In this study, diabetics were either assigned to a low-fat vegan diet or to a control group with no diet change. Everyone was also given a vitamin B12 supplement for the 20 weeks of the study. At the end, the vegan group achieved improved blood-sugar control with some patients needing to have their medication reduced. They also experienced healthy weight-loss, a decrease in cholesterol levels and much greater reduction of pain compared to the control group.

A 2006 study, conducted by the Physicians Committee for Responsible Medicine with the George Washington University and the University of Toronto, tested the health benefits of a low-fat vegan diet emphasising foods with a low glycemic index value (low in fast-absorbing sugars) and excluding all animal products on people with type 2 diabetes. It was compared to a diet based on the American Diabetes Association (ADA) guidelines which restricted calorie intake and limited carbohydrates (Barnard et al., 2006). Portions of vegetables, grains, fruits and pulses were unlimited. Over the 22-week study, 43 per cent of the vegan group and 26 per cent of the ADA group reduced their diabetes medications. Furthermore, the vegan group lost an average of almost one stone (13 pounds), compared with just over half a stone (9 pounds) in the ADA group.

Overall quality of this vegan diet was compared to the ADA diet on the basis of the Alternate Healthy Eating Index (AHEI), which is used to estimate the risk of chronic diseases (Turner-McGrievy et al., 2008). It employs a scoring system which assesses several dietary behaviours and rates food and nutrient intakes. The vegan group improved in every AHEI food category (vegetables, fruit, nuts and soya protein, ratio of white to red meat, cereal fibre, trans fat, polyunsaturated to saturated fat ratio) and significantly improved the overall AHEI score. The ADA group improved in only two categories (nuts and soya protein, polyunsaturated to saturated fat ratio) and did not improve the overall AHEI score of the group. An increase in AHEI score was also associated with decreases in HbA1c value (which measures blood sugar levels over time) and weight.

Following the success of the previous studies, a 74-week clinical trial using a low-fat vegan diet was conducted (Barnard et al., 2009a). Participants were type 2 diabetics and they were randomly assigned to a low-fat vegan diet or a diet following ADA guidelines. HbA1c changes (measure of blood sugar control) from the beginning of the study to 74 weeks, or to the last available value before any medication adjustment, were -0.40 points for the vegan group and 0.01points for the conventional diet group. In patients whose medication did not need to be adjusted, HbA1c fell 1.23 points over the initial 22 weeks, compared to 0.38 points in the ADA group. Glycemic control, therefore, improved significantly more in the vegan group.

The reduction in triglycerides (fats in blood) in the vegan group was also remarkable as was the decrease in cholesterol levels (20.4mg/dl in contrast to just 6.8mg/dl in the conventional group). Both groups managed to lose weight but unlike the vegan participants, volunteers on the conventional (ADA) diet had restricted calorie intake whilst the vegan group did not.

A similar study was conducted in Europe, where half of the study participants (all type 2 diabetics) were assigned to a plant-based diet for 24 weeks and the other half served as a control group (Kahleova et al., 2011). The plant-based diet was based on fruit and vegetables, wholegrains, pulses, nuts and seeds. Both groups received a vitamin B12 supplement and were encouraged to do gentle exercise in the second half of the study. At the end, 43 per cent of participants in the experimental group but only five per cent in the control group had to have their medication reduced. An increase in insulin sensitivity was significantly greater in the experimental group than in the control group and  the experimental group also lost much more fat. This was accompanied by a reduction in oxidative stress markers in the group following a plant-based diet which indicates better health.

Volunteers participating in some of the above studies preferred the vegan diet not only because it was effective but also because they found it better than the diet previously recommended. Participants in the 74-week study were repeatedly asked to rate the acceptability of their diets (Barnard et al., 2009b) and the results showed that patients initially felt more restricted by the ADA diet and at the end of the study reported that the vegan diet was as acceptable as the conventional diet. These findings suggest that following a diet that reverses diabetes is no harder than following a conventionally recommended diet which produces only minor changes in metabolism. 

Parallel to these intervention studies, another research group focused on analysing dietary patterns of 2,875 volunteers without diabetes and determined their risk of diabetes by repeated measurements of basic indicators – blood glucose, insulin concentrations, cholesterol levels, and waist circumference (Liu et al., 2009). Their findings were clear: consumption of a diet based mainly on plant foods protects against insulin resistance, while refined grains, high-fat dairy, desserts and sweets and sugary soft drinks promote insulin resistance.

Vegetarians have lower rates of diabetes than the general population but one study looked at the vegetarian dietary patterns across North America more closely to identify any differences (Tonstad et al., 2009). The authors discovered that vegans had the lowest prevalence of type 2 diabetes, only 2.9 per cent, lacto-ovo vegetarians 3.2 per cent, pesco-vegetarians 4.8 per cent, semi-vegetarians 6.1 per cent and finally meat-eaters in this population had 7.6 per cent prevalence.

The usefulness of vegan diets was eventually endorsed even by the American Diabetes Association when in 2010, their Clinical Practice Guidelines stated that plant-based diets had been shown to improve metabolic control in persons with diabetes (American Diabetes Association, 2010).

Case study: Peter Scott, Lancashire

After being diagnosed, I became a typical example - fat around the middle, high BMI, high cholesterol and high blood sugar. I had no energy, would cough a lot despite not smoking and didn't sleep well. But then I heard about the vegan diet. After just four weeks on the diet, my blood pressure started to fall towards normal levels. All my blood readings were approaching or within normal ranges. After eight weeks on a vegan diet, I lost 1.7 stone. Four months later, I’m no longer obese and my blood results show I’m no longer diabetic. I feel fitter, I sleep well, I wake up more quickly even without coffee and I've stopped the coughing probably because I stopped drinking milk. I continue to enjoy the diet and my new way of life. Thank you again, Viva!, for your work and your help.

For resources on how to prevent, manage or reverse diabetes through diet, go to Viva!Health's diabetes resources page. For references to the above, see The Incredible Vegan Health report