Prevention of Diabetes

A chronic condition of hyperglycemia, which has devastating effects on people’s lives, is common and affects something like 400 million people around the world. Eighty percent of those people live in low- and middle-income countries. The disease affects people of working age. In those countries it is very expensive to treat and it’s very expensive to have, because it impacts on income. So, the condition is a problem of low- and middle-income countries, that’s increasing in frequency and is a clinical problem. Because it impacts on such large populations it’s a public health problem, but because it’s affecting disadvantaged populations and is impacting on their economy, it is actually a challenge to sustainable development.

Now the question before us really is how can we respond to this challenge. The prevention of diabetes is possible. This is the most encouraging news from the last twenty years. That news comes from randomized controlled trials undertaken initially in the United States of America and in Finnland, and then in various other populations around the world. It was possible to demonstrate that if you found people with pre-diabetes, a condition of raised glucose levels, but without formal diabetes, then it was possible by individual level intervention in those people to reduce progression to diabetes. In all those trials a combination of diet, weight loss and physical activity intervention gave rise to an approximate hardening of the risk of progression to diabetes.

Certainly in the Finnish study it was a combination of five behavioral targets: achievement of a normal level of obesity or body mass index, having regular physical activity, eating a diet that was low in total fat, a diet that was low in saturated fat, and a diet that was high in fiber. If people achieved four or five of those behavioral targets, the risk of progression of diabetes in the Finnish study was zero. But if people were achieving none of those targets, it was about forty percent over three years. So, we know that individual level intervention can reduce progression to diabetes over the short term.

In follow-up studies in Finland and also in China it was clear that that effect is long term and lasted beyond the duration of that individual level intervention. The studies in China also demonstrated that not only was that lifestyle intervention associated with a reduction of risk of progression to diabetes, but also with evidence of reduction of risk of the complications of diabetes, specifically retinopathy – the eye disease associated with diabetes, and importantly demonstrating that it was associated with reduction of cardiovascular mortality. That has been demonstrated in the Chinese study in women only for some reason, not in men. Further studies will investigate whether this effect is seen in both.

The important fact is that the long-term benefits of these lifestyle changes persist, and that they give rise to changes in real chemical outcomes. Not only that, but cost-effectiveness analysis has demonstrated that the costs of intervening are relatively small in comparison to benefits, and that these sorts of interventions are affordable for many health care settings. The challenge is whether they are the only approach. Is that a sufficient way of dealing with the condition?

Many countries around the world are contemplating changes to their health care systems to implement primary prevention strategies of this type. But is that enough? I think, the general consensus is that it is not, and these individual approaches prevention important and evidence-based as they are, need to be complemented by a different approach.

That different approach takes a step back and ask the question: Are these changes that we see in populations in the way we lead our lives –in diet, physical activity – are they the result of individual problems or are we really creatures, who live in a society where we make individual choices? Those individual choices are bounded by the type of environment we live in. We might think that we have complete control on our physical activity levels, but if we live in a place where stairs are not readily accessible or it’s difficult for a walk at lunchtime, or if we constrain to our cars, then our physical activity is somewhat beyond our control.

So, I think in conjunction with thinking about diabetes as a clinical problem and thinking about its prevention as being an individual issue, we have to understand the science of what is driving physical activity and diet change in population, and then try and intervene at that level. That needs a mix of science and also political will. We have to accept that this condition is a public health problem. It’s driven by societal level changes in behavior. We have to accept that we may need to intervene at the societal level to change things.

In the case of physical activity that may mean reconfiguring our built environment to make it possible for us to re-engineer physical activity into our lives, making it possible to have healthy forms of transport, and to be physically active in work as well as in recreation. In the case of diet we have to not just impact on people’s attitudes and wishes with respect to diet, but we have to affect the food supply, the price of food, to try and promote healthy forms of eating. That may require some interventions that are or have been unpalatable to some forms of government, such as taxation of unhealthy foods, or subsidy of healthy ones. Really, if we’re to tackle the problems of obesity and diabetes at the societal level, it requires that form of radical thought coupled with evidence of effectiveness. I think it’s only through the combination of individual approaches to prevention, coupled with these societal level changes that we’re really going to give hope to 600 million people around the world, who are at risk of diabetes.

Certainly in developing countries where it’s difficult to treat the people who have the condition now, the idea that we could swamp their healthcare system with people who need to be treated individually, who have pre-diabetes, that’s really unsinkable. In those situations we need an evidence base that supports policy level interventions to prevent diabetes.

The problem of diabetes is not really one of unhealthy foods, but rather of unhealthy diets. The principal driver of the rising diabetes is the availability of calories, and particularly empty calories that can be consumed, which have no nutritional value. This is a function of the food availability and the food supply coupled with the pricing of food. Of course, at one level the availability of cheap and abundant calories is a massive public health advantage, because it’s tended to reduce the prevalence of undernutrition, but really the pendulum swung too far the other way. Throughout the second half of the 20th century the availability of cheap and abundant calories has given rise to an increasing prevalence of obesity and a consequent risk of diabetes.

Beyond that there are some specific food groups and diets that are associated with type-2 diabetes risk. First and foremost when we look at processed meat, and particularly red meat consumption, which is going up very steeply around the world, and that is are so strongly associated with type 2 diabetes risk. On the flip side people who eat diets that are rich in fresh fruit and vegetables seem to be protected from diabetes. Overall people who eat a Mediterranean form of diabetes have a lower risk. It suggests that a diet, which is relatively rich in complex carbohydrates, has high plant and vegetable consumption, low red meat consumption, moderate amount of alcohol – a sort of Mediterranean pattern – may be the one that’s most beneficial to diabetes.

The best evidence we have about how our genetic predisposition to obesity and diabetes, and lifestyle factors how they interact and interplay to give rise to diabetes comes from large prospective cohort studies, in which the risk factors have been assessed at baseline, and then people have been followed up for the occurrence of diabetes subsequently. Here in Cambridge I led an international study called “Interact” from a population study of half a million people recruited across 10 different European countries in the epic study. Those people had their diets and physical activity characterized at baseline, and then were followed up and 12 403 of those people develop type-2 diabetes. We were then able to study the combination of their genetic predisposition to diabetes with other factors, such as obesity.

Although genetic predisposition gives rise to an increase in diabetes, risk is dwarfed by the impact of the obesity. Amongst the individuals who were obese there was an increased risk in those who were obese and also were genetically predisposed to diabetes. But the risk of developing diabetes was high in all the individuals who are obese. Conversely amongst people who were thin or normal weight there was a small increase risk in diabetes that was attributable to genetic predisposition, but it was insignificant on an absolute scale. These data suggest that whilst genetics is important in predisposing one to the condition, the risk is really dominated very strongly by obesity.