While training in endocrinology I worked in diabetes clinics where patients routinely had their blood sugar and weight measured on arrival. These were recorded in the invariably large, cardboard-covered, medical records. Some of the patients had records going back decades. This was before diabetes had become so common that general practitioners were thought competent to manage the problem. Back then everyone with type 2 diabetes came to a specialist clinic for prescriptions of metformin and sulphonylureas and the unvarying advice to lose weight. I liked flicking back through the years and recognising the handwriting of my older colleagues and occasionally myself as a junior doctor. Most interesting, however, were the weight recordings. Although there were short-term fluctuations the pattern was one of remarkable stability. When patients are first referred to clinic they are usually determined and enthusiastic and most will lose weight on learning the role that excess abdominal fat plays in glucose intolerance. A 5–10% drop in weight was common. But by six months it would start to creep up again and by two years even quite large amounts of weight lost would usually be regained.
Setting patients up to fail?
Recommending a course of treatment with only a small chance of success wouldn’t be so bad if it was framed this way from the outset. Cancer specialists do this all the time and their patients then make an informed decision to try the therapy or not. Imagine if we said to our patients ‘you can try weight loss. It would be really helpful if it worked but there is an 85% failure rate’ (Ayyad and Andersen, 2012). Many might decide they’d rather avoid the misery of dieting and accept the inevitable; or try to improve their health in other ways. But we not only fail to give realistic expectations, we do something worse. We place all the responsibility for treatment success or failure on our patients rather than ourselves, AND we imbue the treatment success with value not normally associated with effective medicine. Years after I decided not to recommend losing weight as a treatment option I had to stop myself from celebrating weight loss when it occurred. The flipside of congratulations for weight loss is disapproval of weight gain. People with diabetes are constantly judged for their compliance with blood sugar testing regimes and dietary restrictions. I figured the last thing they need was a value judgement attached to the biological phenomenon of weight maintenance.
Obesity has genetic roots
What’s surprising is our adherence to the notion of personal responsibility for weight when it is clear that body size is genetically determined and that physiology fiercely resists weight loss. What naturally slim person doesn’t have at least one slim parent? Sure they eat less than their overweight friends but it’s clear that appetite is genetically determined. You only need to look at the monogenic forms of obesity such as Prader-Willi Syndrome and leptin deficiency to see that the problem is one of hyperphagia. Professor Jane Waddell studied responsiveness to food in breast-fed babies and young children demonstrating a clear relationship to genetic predisposition to obesity (Llewellyn et al 2012; Webber et al 2008).
And it’s not the rearing environment. In 1986 the psychiatrist Albert Stunkard first examined the relative effects of genetics and rearing environments using the Danish adoption registry. The purpose of the registry was to study genetic influence in schizophrenia but he used the data collected to study the effect of adoption on body weight. He found almost no correlation between the weight of adopted children (now adults) and their adoptive parents despite the shared environment. Fortunately, the registry had recorded the weights and heights of the biological parents at the time of adoption and Stunkard discovered a close relationship between the weights of children and their biological parents. Eighty per cent of those who had two obese biological parents were also obese – it was as if the rearing environment had no effect at all (Stunkard et al 1986).
Our obesogenic society
Stunkard’s dismissal of environmental effect was understandable in 1986 when the prevalence of obesity was still low. But the sharp rise in obesity prevalence since then can only be attributed to environmental change. My take on this is that if you look around a room full of people, the difference between those who are fat and those who are thin is their genes. But the number of people in the room who are fat compared to 30 years ago is the effect of the environment. Whether we call it the thrifty genotype or something else it makes sense that people with active food-seeking behaviours, people who have appetites that allow them to eat all the food in front of them at one sitting in case food is subsequently scarce, will have had a survival advantage in evolutionary terms. It is only in modern times where food is ubiquitous and energy-dense that the food seeking genes have become maladaptive. Understanding the genetic basis of appetite explains why some of us eat more in response to the enticements of advertising, two for one deals, and increased portion sizes while the lucky few remain unaffected. We don’t have to do genotyping to see which we are, we only need to look at our parents.
Collusion between libertarian governments, food and drink manufacturers, supermarkets and advertisers and the weight loss industry
Of course it is more complex than that; there are the effects of poverty, stress, shiftwork, culture. By the time you add long commutes, sedentary jobs and cities that are too scary to walk in as contributors to inactivity, you see that free will as a determinant of our weight plays a very tiny part indeed. Why is it then that all of our efforts to turn the obesity epidemic around are focused on persuading people to change their behaviour? I see it as collusion between libertarian governments, food and drink manufacturers, supermarkets and advertisers who don’t want to restrict commerce and the weight loss industry which makes money out of individuals’ doomed attempts to lose weight.
Not only do I not ask people to lose weight I tell them that despite their best efforts, my expectation is that their weight will not change
A gentler approach to lifestyle change
So what do we do? We start off by being honest. Not only do I not ask people to lose weight I tell them that despite their best efforts, my expectation is that their weight will not change. We therefore have to concentrate on keeping them as healthy as possible by other means. Nutrition remains critically important and my focus is on getting people to eat more fruit and vegetables. The only thing I suggest they prohibit is sweetened drinks. If they are addicted to the sugar/caffeine combination I recommend artificially sweetened drinks as an alternative. Exercise too is extremely beneficial independent of any effect on weight (negligible). I’m interested in the data that shows sedentary behaviour is particularly harmful. I explain this to my patients and describe how some of the physiological processes that drive the metabolic syndrome are switched off by interrupting spells of sitting with short bursts of activity. (Whenever I chair a session at a conference I get everyone to stand and walk around after an hour). Otherwise I recommend incidental exercise over recreational activities. Best of all is travelling to work by public transport as this provides non-discretionary exercise walking to and from the bus or train etc. Attendance at gyms or participation in group sports is all good of course but it tends not to be sustained. Best of all is to have a job that is physically active but these are now rare. For individuals with morbid obesity the only effective treatment is obesity surgery, preferably the laparoscopic Roux-en-Y gastric bypass although newer techniques may supplant this.
A whole-of-society approach
For the population we need public health measures that reduce the consumption of energy-dense, nutrient-poor food and drinks and increase physical activity. This requires a whole society approach and at government level there needs to be a food policy that encompasses transport, treasury, agriculture, trade, education and health. Anti-obesity initiatives need to be evaluated for their potential for stigmatisation. Governments tend to favour behavioural approaches such as requiring schools or doctors to measure BMI and provide education but these continue to attribute obesity to personal choice and have the potential for harassment – especially at school. Environmental approaches such as changing zoning laws to prohibit the building of fast food outlets near schools and fiscal measures like taxation on sugar-sweetened drinks have low potential for stigmatisation. There is no need to remove choice or the opportunity for highly motivated individuals to overcome genetic odds and remain slim. We do however need to understand that it is for reasons outside our control that most of us are now fat and becoming more so with time. We need to change the default so that it is easy to eat healthily and to exercise, not the other way around.
- Ayyad C, Andersen T (2012) Long-term efficacy of dietary treatment o f obesity: a systematic review of studies published between 1931 and 1999. Obesity Reviews 1(2) pp 113–19.
- Llewellyn C, van Jaarsveld CH, Plomin R, Fisher A, Wardle J (2012) Inherited behavioural susceptibility to adiposity in infancy: a multi – variate genetic analysis of appetite and weight in the Gemini birth cohort. American Journal of Clinical Nutrition 95(3) pp 633–39.
- Stunkard AJ, Sørensen TI, Hanis C, Teasdale TW, Chakraborty R, Schull WJ, Schulsinger F (1986) An adoption study of human obesity. New England Journal of Medicine 314(4) pp 193–98.
- Webber L, Hill C, Saxton J, van Jaarsveld CHM, Wardle J (2008) Eating behaviour and weight in children. International Journal of Obesity 33(1) pp 21–28.