Total Pageviews

Tuesday, March 17, 2015

Getting old and fat ~ no problem!

Growing old and fat ~ no problem


In today’s obesocentric (a new word I’ve coined) world, there are certain things that are given. These are the immutable facts about obesity. One of them is that the link between obesity, as measured by Body Mass Index (BMI = kg/m2), is U shaped. Over the range corresponding to desirable weight (BMI =20 to 24.99), there is no relationship between BMI and mortality or morbidity. Below 20, there is a rise in the risk of mortality as one gets skinnier. Above 24.99, mortality rises as BMI increases and it soars when obesity exceeds 30. No matter where you look, this is a given. It’s on the WHO website. It’s in your pharmacy window. It’s in schools, in textbooks and its like the boiling point of water, a given, never to be challenged.

Recently, in writing my new book on obesity, I re-discovered a set of data that was 30 years old. The data were published in a Working Party Report of the Royal College of Physicians of London in 1983[1]. It looked at the association between mortality and BMI over decades of age. Among 20 and 30 year olds, the rise in mortality with increasing BMI was quite dramatic above25. So far so good. Among 40 and 50 year olds, we begin to see the emergence of a U shaped curve, rising at both low and high BMI values. However, among 60 year olds, whereas BMI cause rising mortality at very low values, there was no relationship between BMI and mortality above a BMI value of 20. Of course, death rates rose with age but the pattern of mortality with BMI was very much influenced by age.

The subject has recently been revisited through a systematic review and a meta-analysis[2]. A total of 564 publications were identified in the literature based on key words of which 394 were rejected because of a focus on specific patient groups or non-human studies. Of the remaining 170 papers, a further 150 were excluded because they did not provide two or more age comparisons. That left 20 papers of which 13 were US based, 2 were Finland and Taiwan based and 1 each were from Germany, India and Japan. The increase in the risk of mortality with a BMI greater than 24.9 kg/m2   was assessed for decades of age. Averaging the values for men and women, the increased mortality risk with elevated BMI was 60% among those aged less than 35. That figure the fell progressively: 40% among thev35-45 year olds, 355 for 45-55 year olds, 28% among 55-65 year olds, 20% among the 65 to 75 year olds and a mere 11% in those aged above 75 years. The studies all controlled for the possibility of smoking or a pre-existing illness (e.g. high blood pressure or diabetes) being statistically confounding factors. However, the paper reports that this statistical control had no effect on the outcome. Basically, each increasing decade of life above 35 years of age reduced BMI related mortality by 10%.

Writing as a pensioner, a grandfather and a 67 year old, I say YIPPEE. But my GP doesn’t know this nor does my friends cardiologist and so we grey-heads are demonized unnecessarily by medics into being of adequate BMI. Of course, we do benefit from cardiovascular fitness so a good walk during the day when all the younger folk who kindly earn our pension, work away in a stress-filled environment increasing their BMI mortality risk, is a good idea.

Why does this obesocentric world choose to ignore such data? The answer, it seems to me, is that the simpler the message, the easier is the public health communication. So, they don’t complicate it by absolving older folk from the curse of BMI related mortality. And of course there is the other issue of the obesity paradox[3]. It gets even messier but I love it, the truth, that is!!!



[1] Royal College of Physicians Working Group on Obesity (1983) Journal of the Royal College of Physicians London 17:3-58
[2] Wang Z (2015) Obesity Research & Clinical Practice, 9:1-11
[3] BMI, Obesity & mortality: three grand challenges. Gibneyonfood, December 15th, 2014

Thursday, March 5, 2015

Fats, facts and baloney

In the last year or so, we have seen a number of scientific papers reviewing sets of older scientific publications linking saturated fats to heart disease and calling for a reversal of prevailing wisdom that a change in the composition of dietary fats (less saturates and more unsaturates) would help reduce the public health burden of heart disease. The spin-doctors from the food industry have whipped up the hype and last summer the front cover of Time magazine headlined: “Eat butter. The scientists labeled fat the enemy. Why they were wrong”.
Before I take at look at the detailed issues, I would make two important points. The first is that no matter how unpalatable a critical paper might be to the custodians of some aspect of conventional scientific wisdom, there is an absolute need for continued critical analysis of existing theories. As I have often argued, dissent is the oxygen of science. The second point I would make is that the advent of any one paper is never in itself sufficient to unravel prevailing wisdom. Philosophers of science have the luxury of arguing that the just one black swan demolishes the “All swans are white” theory. Well its not so black and white in an area as complex as public health nutrition.

Last year, a paper was published in the Annals of Internal Medicine[1], which looked at previously published studies that sought to establish links between different types of dietary fats and heart disease. In response to this publication, the journal received many letters most of which were critical of the study. I won’t re-hash their arguments but will offer my own. Back over half a decade ago, Professor Ancel Keys conducted a study across 7 countries in which he showed (a) that measures of the composition of dietary fat could help predict blood cholesterol levels and (b) that the national statistics in these countries showed a link between rising blood cholesterol and higher rates of heart disease. His work is now often referred to as being flawed and certainly, this study, the first of its kind, wasn’t exactly top drawer in today’s standards no more than the falling Newtonian apple can rank with today’s Hadron Collider in the study of gravity. It is not the 7 countries study that mark out the great contribution of Keys to modern nutrition but to the subsequent experiments he conducted on humans to study how precise control over the intake of saturated, monounsaturated and polyunsaturated fatty acids would influence blood cholesterol levels. In this he was joined by two other Doyens of that great nutrition era, Paco Grande and Scott Grundy. They developed equations, which would accurately predict the change in blood cholesterol that would occur with changes in the composition of dietary fats. These were real experiments with human subjects, not some fancy statistical model from the world of epidemiology. What that world delivered, and delivered in spades, was epidemiological data showing that rising population levels of blood cholesterol were associated with rising levels of coronary heart disease.  So here we had dietary data showing an unequivocal link between the compositions of dietary fat and plasma cholesterol and powerful evidence linking elevated plasma cholesterol levels with heart disease. So, it made sense to recommend a reduction in the intake of saturates and their partial replacement with unsaturated fats to minimize population levels of blood cholesterol BUT NOT TO BE THE MAIN DRIVER IN ANY MOVE TO REDUCE OVERALL HEART DISEASE.  Why not? Because heart disease was also partially caused by smoking, by high blood pressure, by the advent of the menopause in women, by blood clotting potential, by bodyweight, by fitness, by family history and by many other factors, some of which were easy to measure on a population basis and some of which were difficult to measure and thus were rarely entered into complex statistical models (platelet aggregation, post-prandial lipemia, cardio-respiratory fitness, family history of heart disease, common known gene variants etc).

So the biggest error in looking at dietary fat and heart disease is that nobody ever said that there was a direct link between heart disease and the composition of dietary fat. All that was ever said and in two very clear stages was (a) that dietary fat for certain, influences blood cholesterol and (b), blood cholesterol along with a wide range of other factors contribute to heart disease. A more recent paper[2] makes the same mistake of attempting to link dietary fat intakes to the incidence of heart disease when the human intervention studies with a focus on fats had blood cholesterol as the end point.

To hammer home this point, consider the generally accepted public health nutrition success that is attributed to the reduction if not elimination of hydrogenated trans fats from the human diets. It was a triumph for NYC Mayor Michael Bloomberg in engineering a citywide movement for their elimination.  What constituted the risk associated with trans fats? It wasn’t mortality from heart disease. It was a series of human experiments conducted by the Dutch researchers Martijn Katan and Ronald Mensink in showing an unequivocal effect of trans fats on adverse levels of blood cholesterol. The epidemiologists contributed both the only actual human studies were on blood cholesterol and trans fats intake[3].

Very recently, the industry spin-doctors debunking the dietary fats heart disease putative link (the true link is dietary fats and blood cholesterol) have acquired a new string to their bow, the 2015 US Dietary Guidelines. The 2015 guidelines make the point that “available evidence shows no appreciable relationship between consumption of dietary cholesterol and serum [blood] cholesterol”. This is simply the conservative policy catching up with the science since it has been know for decades that the main source of cholesterol in our blood is manufactured by our guts and livers and the dietary contribution from ingesting cholesterol is minimal. The report goes on to argue that: “The DGAC encourages the consumption of healthy dietary patterns that are low in saturated fat, added sugars, and sodium. The goals for the general population are: less than 2,300 mg dietary sodium per day (or age-appropriate Dietary Reference Intake amount), less than 10 percent of total calories from saturated fat per day, and a maximum of 10 percent of total calories from added sugars per day”. For many, this is taken to be a major turn around for dietary advice on restriction of sugar intake but I’m afraid its no more than the US catching up with the rest of the world who have held the 10%target for added sugars over several decades now.

Despite the minor adjustments and catch up of the US to dietary guidelines and notwithstanding the odd paper suggesting otherwise, its business as usual for nutritional advisory committees globally. The press will report on dramatic somersaults and loop-the loops and the public will continue to be confused. In effect, nothing has changed.




[1] Chowdhury et al (2014) Ann Intern med, 160, 398-406
[2] Harcombe Z et al (2015) Open Heart, 2, e000196 (BMJ)
[3] Brouwer IA (2010) PLOS one (5) 10, e 9434