Total Pageviews

Tuesday, December 27, 2016

Diet and poverty: Mind over money?

Diet and poverty: Mind over money?

Obesity is popularly linked with low-income groups and with a low socio-economic status (SES). This relationship is frequently explained by a lack of money, a lack of understanding of healthy nutrition, a general disdain for healthy eating and both isolation from supermarkets and proximity to convenience and fast food outlets. I won’t argue with this except to say that while obesity is more prevalent in low SES, there are a lot of fat judges, mayors, prime ministers, presidents elect, ex-sportsmen, doctors, nurses and others for whom the putative drivers of obesity in poverty are generally absent. New research[1] on obesity and SES has recently been published by scientists from Singapore’s A*STAR and published in the prestigious journal, The Proceedings of the National Academy of Sciences. The key question they asked was thus: Irrespective of your true SES, if you were made to think you were of a low SES, would that influence your food habits?

They started their thinking process with reference to the animal kingdom where in any social organisation, those that are marginalized for whatever reason (too slow, too weak, too fussy), also tend to overeat and gain weight. This might be a precaution for survival in the event that they might one day be in fact ousted from the group. They cite a hamster study where a smallish hamster living in harmony with other small hamsters is placed alone in a cage into which a larger hamster is admitted. The latter will assert itself as dominant within less than a minute and the  “defeated” smaller hamster is left to sulk. The plot of body weight over time comparing those hamsters subject to defeat against those spared this ignominy shows that the defeated hamsters, stigmatized and “psychologically” marginalised, get fatter.

Against that background from the animal kingdom of which the hamster data is merely illustrative of so many such studies, they set out to study if an “imagined” low SES could alter food habits.

The human volunteers were placed in a private room and the study was conducted in what were effectively laboratory conditions. They were shown a ladder with the following instructions:

Think of this ladder as representing where people stand in Singapore. Now, please compare yourself to the people at the very top of the ladder. These are the people who are the best off—those who have the most money, most education, and most respected jobs. In particular, we’d like you to think about how YOU ARE DIFFERENT FROM THESE PEOPLE in terms of your own income, educational history, and job status. Where would you place yourself on this ladder relative to these people at the very top? Please select the number that corresponds to the rung where you think you stand in relation to these people.

 A second group was similarly treated but they compared themselves to people at the bottom of the ladder: poor, less education and rotten jobs. To emphasise the mental induction of subjective higher or lower SES, subjects had to write down in detail how a conversation might go with the group above or below themselves and to suggest how such a conversation might flow. Now, suitably indoctrinated into a feeling of subjective high or low SES, the experiments began and the easiest way to communicate the studies is to simply focus on the results.

  • When asked to select the foods they might be subsequently included in a meal, those who subjectively thought of themselves as lower SES, chose higher energy dense foods and overall indicated an intention to eat more calories than those who were attuned to think of themselves as higher SES
  • When asked to rate foods according to descriptors such as pleasant (e.g., tasty, delicious, wonderful) and unpleasant (e.g., disgusting, nasty, awful) those in the lower SES subjective state chose the higher energy density foods as the most pleasant
  • In a third study, the subjects viewed a short documentary video while freely eating three snacks (potato chips, M&M candies, and California raisins) from separate bowls. Those with the lower SES frame of mind ate more calories and they more or less favoured the chips and M&Ms over the raisins.
  • Finally, the subjects were given access to a meal in the form of noodles and asked to eat to appetite. For completeness, a control group who did not take part in the imagined SES ranking exercise was included. The outcome showed that those who subjectively rated themselves as lower SES ate significantly more then the other two groups.

These studies were carefully designed and the analysis took statistical account of all key variables (age, gender, weight etc.) including the actual SES of the subjects. The results could be interpreted thus: When induced to think they are of a low SES, people overeat because that’s what they think those of low SES actually do. However, there are a few reasons to downplay this potential explanation. First, each experiment involved different volunteers so there could be no sense of training. Secondly, food was not portrayed as a central element of the study but rather the self-imposition of a state of mind. Finally, and most importantly, those induced to imagine themselves as low SES, were comparing themselves to those at the top of the social ladder who were the economic elite.

So, once again, we see what we refer to as the “mind” playing a driving role in one of nature’s most powerful biological drives, the drive to eat[2].

[1] Cheon BK & Hong Y-Y (2016) Mere experience of low subjective socioeconomic status stimulates appetite and food intake Proceedings of the National Academy of Sciences (Early edition)
[2] Mind over matter: Perceived time as opposed to real time on blood glucose in type 2 diabetes: Blogged October 14th 2016

Monday, November 28, 2016

Dieting and the appetite battle

I have previously written that if slimming were a drug, it wouldn’t be allowed for sale given that after 5 years of treatment, less than 5% would be successful in their weight loss. This is worse than the cure rate for the worst cancer.[1] To many in public health nutrition, this relapse tendency is utterly ignored as some arcane piece of physiology, which quite simply is inconvenient to their strategy and, anyway, not as important as is made out to be. Indeed “Look at all the people who have lost weight” they cry, while all the time ignoring the far greater numbers who weren’t so lucky. A new paper from the NIH National Institute for Diabetes and Digestive and Kidney diseases, led by Prof Kevin Hall who is doing stalwart work in this field, throws strong light on this issue of relapse and is well worth reflecting on.

When we organise experimental weight loss programmes, we use dietary advice on weight loss and comparable expert advice on physical activity. There can be no placebo. We can have a control group who receive little structured and certainly not personalised or customised advice, but until now, the experimental advantages of a placebo controlled diet is almost impossible to achieve, Until now that is. A drug, with the silly name Canagliflozin, which, as the lawyers say, hereinafter is known as the “drug treatment” has been used in this study.[2] It is presently available for the management of type 2 diabetes and leads to a sizeable excretion of glucose each day. That means that the subjects pee out a lot of untapped calories and therefore this drug leads to some weight loss.  The key question the researchers set out to address was that in a placebo controlled blind trial using this drug, how did subjects respond in terms of compensating energy intake to the weight loss anticipated from the drug.

As is seen from the attached figure, the drug did lead to weight loss but it reached a plateau. The daily drug-induced loss of untapped calories from urinary glucose remained for the 54 weeks but the initial weight loss was checked by a compensatory rise in energy intake. Normally energy intake is measured using dietary assessment methods but in this case the change in energy intake was related back to changes in bodyweight using a complex equation that had been previously shown to be valid[3] against very robust biomarkers (doubly labeled stable isotope water along with multiple dual-energy X-ray absorptiometry scans).  The authors estimated that the energy intake compensation was 100 kcal per day per kg of weight loss. Adaptations in terms of changes in energy expenditure were minimal.

In this study, half the subjects (153 Type 2 diabetics) received the drug and half the placebo and there was no advice on weight loss. So there was no psychological effect of the intervention on weight loss or on appetite. As weight was shed, appetite responded to a strong and lasting effect, which minimised the process of weight loss. As the authors point out, those who successfully diet and who achieve significant and prolonged weight loss, battle by ‘heroic and vigilant efforts’ against the powerful effect of this compensatory rise in appetite. Maybe there is a threshold of weight change below which the body choses to ignore in which case we can win limited success in mass weight loss programmes. But if as this study shows, weight loss is not minimal (>3kg average), then we will need to have a much better understanding of how to motivate people for that long term battle against compensatory weight loss. May I end this blog by once again citing the Sir David King FRS formerly Chief Scientific Adviser to HM Government, and Head of the Government Office for Science who in the UK Foresight Report on Obesity wrote: “What quickly becomes apparent to anyone who examines the body of evidence from several different disciplinary sources is that the answers are neither straightforward nor, as is popularly supposed, necessarily known. Although a great deal of research has been done into the problem, much of the evidence is not integrated”.  What is the threshold of weight loss for energy intake compensation? What aspects of appetite adaptation are most affected by the most and least successful dieters? What are likely to be the most helpful strategies to those who want to lose weight and must battle this enhanced appetite phenomenon? There are no simple answers.

[1] Ever seen a fat fox ~ human obesity explored (2016). UCD Press My new book on obesity
[2] Polidori D et al (2016) Obesity, 24, 2288
[3] Sanghvi A et al (2015) Am J Clin Nutr 102, 353

Monday, November 21, 2016

Fat calories count most in obesity ~ new study

Conventional wisdom, at least from the media and the high priests of public health nutrition, tells us that sugar causes obesity and all its ailments to the point where it is likened to tobacco in its sinister provenance and its foul health implications. Sugar is to be taxed and restricted in every way and politicians of all persuasions win popular support for their call for yet more extreme measures to decry the putative effects of sugar. And all the time, this conventional wisdom is promoted by the media and by celebrity chefs and diet gurus.. So it should come as no surprise that when a study from a high quality group using high quality data, publish a high quality paper in a high quality journal, which flies in the face of conventional wisdom, that this fine data, is simply ignored.

The paper[1] is based on data from the UK Biobank database that tracks individuals over time (2011 and 2012). It uses direct measures of body weight and height and classified the 132,479 subjects into underweight, normal weight, overweight and obese with obesity further divided into grades I, II and III obesity. Data were also obtained for each subject on % body fat and on waist circumference. Dietary data was collected using four 24-hour recalls over 2011 and 2012. Subjects were excluded from the study if there was any possibility that their diet or weight might be compromised in some way. For example, smokers, who tend to be under-weight, were excluded, as were those with a reported energy intake deemed by international norms to indicate under-reporting of calorie intake.

The bottom line in the findings was that moving across increasing grades of either body weight or body fat, energy intake, as in calories per day, also increased. As I have previously paraphrased US President Bill Clinton in his election campaign: “It’s the calories, stupid!” Compared with normal BMI, obese participants had 11.5% higher total energy intake and 15%, 14%, 10% and 5% higher intake from fat, protein, starch and sugar, respectively. Hence, the proportion of energy derived from fat was higher (34.3 v. 33.4) but from sugar was lower (22.0 v. 23.4).

This simply tells us there is a direct positive association between calorie intake and rising levels of overweight. A key question remains as to the contribution of fat and carbohydrate (with sugar specifically in mind) to obesity when we statistically control for the calorie intake. The authors thus transformed the data to control for the following: sex, ethnicity, physical activity and calorie intake.  This allows us look at data where fat intake and sugar intake are examined for their links to obesity where caloric intake is controlled to a common level. Now we see that when sugar intake and starch intake (all carbohydrates) are compared across rising levels of obesity, they actually fall. In contrast, fat rises. This isn’t anything new because when energy intake is fixed at a constant point, as one contributor to caloric intakes rise (e.g. fat) others, (carbohydrates including sugar) falls, a fact recorded so many times (the sugar-fat see-saw) and so repeatedly ignored by the anti-sugar lobby.  The figures below are taken from the journal where quintiles if nutrient intake are compared to BMI.

The authors conclude as follows from their research:

 Key Messages
• Adiposity is associated with higher intake of sugar; but the association is stronger for fat intake and strongest for total energy intake.
• Fat is the largest contributor to overall energy intake.
• There is only a weak correlation between absolute energy derived from sugar and from fat. Therefore, targeting high sugar consumers will not necessarily target high consumers of fat and overall energy.
• Focusing public health messages on sugar consumption may mislead the public on the need to reduce fat intake and overall energy intake.
This was a study funded by a University of Glasgow internal grant and was not funded by the sugar lobby nor have the authors any conflict of interest to report. But it doesn’t conform to conventional wisdom so it will largely remain ignored by the media and senior policy makers will find it awkward reading, if at all, because they have already followed the herd of the anti-sugar lobby.

[1] Andersen JJ (2016) Adiposity among 132 479 UK Biobank participants; contribution of sugar intake vs other macronutrients. International Journal of Epidemiology, 2016, 1–10

Friday, October 14, 2016

Mind over matter: Perceived time as opposed to real time on blood glucose in type 2 diabetes

To most experimental human biologists, among whose company I include myself, life, in its most basic form, is a series of biological steps, relentless biological steps but beautifully masterminded by genes, orchestrated by hormones and managed by enzymes. These biological steps are shaped by our biological clock, our age, our sex, and by a myriad of inherited factors. They are also shaped by the food we eat and the drugs we take which together, dominate human biology experimental studies.  From time to time, we face assertions that biological processes can be mastered by mental power. We accept that relaxation causes biological effects but do so with the same thought process that tells us that physical activity causes biological effects. We even accept that relaxation can be trained to quite high levels through techniques of, for example yoga or other forms of meditation. The neurological systems are very complex and thus those not expert in the field of neurobiology simply accept such findings on trust.

Now data is beginning to emerge that one of the most measured aspects of life, blood glucose, might be somehow subject to forces other than food, exercise, drugs, disease and other such tangible matters. We are now beginning to see mental forces enter this once biological citadel. This summer, a paper[1],[2] appeared in the highly rated scientific journal, “The proceedings of the National Academy of Sciences (PNAS)” which opens up a whole new vista on mind and biology.

Volunteers suffering from type-2 diabetes (the type associated with obesity as a causative factor) were recruited. All were receiving dietary management therapy and also they were taking metformin as a medical treatment to manage their diabetes. Prior to the study, they were asked to monitor their blood glucose levels before and after every meal and to create a diary of blood glucose fluctuations. Thus they were trained to fully understand their own blood glucose measurements. On the morning of the study, they were asked to arrive having fasted for at least 8 hours. They were asked to hand in phones, watches and anything else that could be used to monitor time such as fit-bits, bleepers, computers, etc. They were also asked to hand in all medication. Now they were ready for the study, which quite frankly was straightforward: sit, relax and watch videos, changing the video every 15 minutes (the actual changing of the video was done by the research staff so the volunteers simply sat and relaxed).

Having taken away all time-related devices, the subjects were left with just one indicator of time, a wall clock. And herein lies the rub. The true time spent by each group was 90 minutes. However, two groups were tricked. For one third of the subjects, the clock was set perfectly accurate and when the allotted 90 minutes was up, they could see so on the clock. For the other two groups, the clock was rigged. For half, it ran twice as fast as it should in reality and for the other, it ran half as slow as reality. Thus one group spent the true allotted 90 minutes in the test. For another group, the test lasted just 45 minutes and for the third group, the time taken was 180 minutes. Blood samples were taken before and after the test and the drop in blood glucose noted. At the conclusion of the test, the subjects were asked to state how long the study lasted and they all more or less agreed with the clock, even though all of them had just spent 90 minutes under the the test conditions. They did not differ in terms of perceived stress but hunger ratings did differ. Those with the fast clock showed the highest level of hunger compared to those rigged to imagine a longer time period. The critically important finding was that the blood glucose levels were dramatically altered by the covert manipulation of time.

Perceived time
Real time
Blood glucose
Decrease over the study (milligrams per deciliter)



The authors conclude thus: “Our findings that the mind can actually adjust the body’s glucose suggests new avenues for treatment and prevention”.  The study was well conducted by experts in psychology from Harvard and Milan. But at the end of the day, experimental biologists should really not be surprised. We use placebo treatments in our study and expect always to see some placebo effect albeit very differently fro the true treatment effect. But it does reinforce the need to have much stronger links between the biological and behavioural sciences.

Note: Almost 60 years ago, in a review in Science[3], the effect of tricking people on time as regards appetite were reported by the great Stanley Schachter at Columbia and these effects are covered in a chapter in my new book on obesity. The dictum that “an hour in the library saves days in the lab” still rings true!

My new popular science book on obesity: "Ever seen a fat fox ~ Human obesity explored" is outlined
on my blog of Tuesday, May 17th, 2016 and at UCD Press

[1] Park C et al (2106) PNAS, 113, 8169-70
[2] I am grateful to Professor Bobby Cheon of Singapore’s A*STAR nutrition group for pointing out this paper to me and I look forward to blogging on his forthcoming studies

[3] Schachter S (1968) “Obesity and eating: the internal and external cues differentially affect the eating behaviour of obese and normal subjects” Science, 161: 751-6