Tuesday, 20 July 2021

Obesity and physical inactivity - the evidence

First published by the Spectator in 2015

Part One

You may have heard the news that the nation’s doctors have had a change of heart about physical activity and no longer believe it to be a sensible way of staying slim. Don’t be too quick to put your feet up. All is not as it seems.

The doctors responsible (or, arguably, irresponsible) for this claim are Aseem Malhotra, Tim Noakes, and Stephen Phinney. Malhotra is a Croydon-based cardiologist who rose without trace several years ago, first attacking junk food and then climbing aboard the anti-sugar bandwagon. Now the scientific director of the wacky pressure group Action on Sugar, he explicitly tells people to eat more saturated fat and implicitly tells people not to bother exercising – unusual advice from someone who looks after people’s hearts for a living. Last year, he wrote an article for the British Medical Journal which was investigated and corrected after it made insupportable claims about the safety of statins. It turned out that Malhotra had brushed aside concerns raised by one of the peer reviewers. His Action on Sugar briefing papers have also contained very questionable assertions.

Malhotra’s co-author, Tim Noakes, is a South African paleolithic diet advocate currently promoting a new diet book. He disputes the evidence that high cholesterol is a risk factor for heart disease and is currently being investigated by the Health Professions Council of South Africa for ‘disgraceful conduct on social media’ after telling a mother on Twitter to wean her infant on to a low-carb diet. (Noakes appears to be taking this in his stride.) Stephen Phinney, the other co-author, is a scientific advisory board member of Atkins – as in the Atkins Diet – and has written several books extolling the low-carb way of life.

They are not, in short, typical doctors. Together they wrote a short editorial for the niche British Journal of Sports Medicine, in which they made the striking and unambiguous claim that ‘physical activity does not promote weight loss’. They then praised the alleged virtues of fat and called for legislation to clamp down on carbohydrates, especially sugar.

Curiously, the authors’ poorly referenced, 1,000-word opinion piece became a national news story. The op-ed was (wrongly) described as a ‘scientific study’ by the Independent and the trio were (questionably) described as ‘international experts’ by the BBC. Indeed, the Beeb’s claim that ‘Physical activity has little role in tackling obesity – and instead public health messages should squarely focus on unhealthy eating, doctors say’ made it sound as if this was the consensus view of the medical profession rather than the eccentric opinion of three Atkins evangelists, one of whom is a lobbyist for a pressure group.

The idea that burning off calories does not help prevent weight gain is idiosyncratic to say the least, but the media haven’t misrepresented them. That is what they say in their article and there is no reason to think they don’t believe it. Malhotra has previously boasted of having convinced the Minister for Public Health that ‘physical inactivity is not linked to obesity’. Their evidence for this claim seems to boil down to the fact that obesity has risen in recent decades despite levels of physical activity remaining the same. Therefore obesity must have risen because people are eating more calories.

The trouble is, it is not a fact. As Public Health England noted in a major report last year, ‘People in the UK today are 24 per cent less active than in 1961’. British Heart Foundation figures show that British adults are walking less (from 255 miles per year in 1976 to 181 miles in 2012) and the proportion of British children who walk to school has dropped from 70 per cent in 1980 to less than 50 per cent today.

At work, we are less physically active than ever. Jobs in agriculture declined from 11 per cent to two per cent of employment in the 20th century while manufacturing jobs declined from 28 to 14 per cent. Less than one in five adults report doing any moderate or vigorous physical activity at work. Outside of work, 53 per cent of us take part in no sports or exercise at all.

You don’t need to be a social historian to see that Britons are leading increasingly sedentary lifestyles. Only a minority of households owned a car in 1965. Today, only a quarter do not. In Britain, as in all western countries, there have been what the World Health Organisation describes as ‘decreased physical activity levels due to the increasingly sedentary nature of many forms of recreation time, changing modes of transportation, and increasing urbanisation’.

If Malhotra et al’s theory seems counter-intuitive it is for the simple reason that it doesn’t stand up. They provide no new research in their ‘study’ and only cite one article in support of their claim about physical activity. That, too, is an opinion piece from the fringes of the scientific debate and has been widely criticised. The phrase ‘overwhelming evidence’ can be overused, but it can certainly be applied to the countless studies showing that physical activity helps prevent weight gain – and to the data showing that rich westerners burn fewer calories than their grandparents.

Why claim something that can so easily be challenged by reference to laboratory, animal and observational trials? Why fly in the face of empirical evidence and lived experience?

The answer, I think, lies in political pragmatism. Governments are not yet ready to pass laws forcing people to exercise, so it makes sense for the likes of Action on Sugar to focus on the food supply where regulation is more likely. A soda tax is more realistic than a sofa tax. If politicians view obesity as a cultural symptom of greater wealth and structural changes in the labour market, they will be less likely to support taxes, advertising bans, graphic warnings and all the other interventionist policies that you would expect from a campaign group which claims that ‘sugar is the new tobacco’.

This, admittedly, requires a certain amount of self-delusion, since the cold facts show that physical activity is not the only thing that has declined since 1980 – sugar consumption has too. But what are activists to do? Bore people with all of the science? Leave them alone? Of course not. Better to trust the media to be suitably deferential to them on account of their PhDs. The media did not let them down last week. The closest Malhotra came to being challenged was when the National Institute for Clinical Excellence said that an obesity strategy that ignored physical activity would be ‘idiotic’.

The multi-million-pound diet industry is, largely if not wholly, based on the conceit that there is more to weight loss than calories in and calories out. Supposed diet gurus will be ignored by right-thinking people, but the line between these people and the medical establishment is becoming increasingly blurred.

It’s difficult to say which is more depressing – three medics making a highly doubtful claim which, if acted upon, would probably harm people’s health, or the media taking a one-page editorial from an obscure journal and reporting it as news. Since most people don’t take health reporting too seriously, the damage to the public is likely to be negligible. It is more likely to be a blow to the reputation of the ‘public health’ lobby, but given the say-anything, do-anything mentality of some self-styled public health experts, it is a blow that is well deserved.

Part Two

Last week I mentioned a widely reported article in the British Journal of Sports Medicine which claimed that ‘physical activity does not promote weight loss’. The article was taken down by the journal last week due to ‘an expression of concern’. It remains offline as I write this, but the controversy rumbles on. At the risk of further upsetting the low-carb community (who seem particularly antagonistic to the doctrine of ‘calories in, calories out’), I am returning to it today.

Let’s start by looking at a series of blog posts by Jason Fung of Intensive Diet Management that have been doing the rounds on social media. He, too, argues that ‘there is no measurable association between obesity and physical activity’. In his first post, he argues that people are exercising more than ever and yet are becoming fatter and fatter. The positive correlation between obesity and exercise, he says, shows that physical activity really doesn’t make much of a difference.

I once met somebody who wondered whether artificial sweeteners caused obesity, based on the fact that their use had risen more or less in line with obesity rates in recent decades. It seems pretty obvious that this is a case of reverse causation. Artificial sweeteners are a response to obesity, not a cause of obesity, just as the rise of jogging and gym membership is a response, not a cause.

To be fair, Fung is not claiming that exercise causes obesity, only that it is not a solution. He cites figures showing that Brits are exercising more than they used to – or, more precisely (and this is crucial), that more Brits are exercising than they used to. Forty-two per cent of British men met the government’s physical activity recommendations in 2008, up from 32 per cent in 1997. At the same time, the male obesity rate rose from 17 per cent to 24 per cent.

But, leaving aside the question of whether people accurately report how much they exercise, why would we expect the minority who exercise to have an effect on the majority who don’t? There is nothing incongruous about obesity rising even if one in ten men are exercising more than they used to. Moreover, there is a conflation between physical activity and leisure-time exercise that is often made by those who downplay the benefits of physical activity.

Office jobs, computers, cars and gadgets have created a more sedentary society which requires fewer calories to be burned. People can offset their less active working and domestic environments by eating fewer calories or burning more calories in their leisure time, but not everybody does. Hence the rise in obesity in recent decades. It cannot be assumed that those who engage in leisure-time exercise are more physically active than previous generations, and it certainly cannot be assumed that the leisure-time exercise of a minority makes the sedentary majority less likely to become obese.

The evidence is quite clear that, on average, calorie expenditure has declined over the years in developed countries. Public Health England (who say that the ‘link between physical inactivity and obesity is well established’) report that ‘People in the UK today are 24% less active than in 1961’. The World Health Organisation has remarked on the ‘trend towards decreased physical activity levels due to the increasingly sedentary nature of many forms of recreation time, changing modes of transportation, and increasing urbanization.’ Harvard School of Public Health says that ‘Physical activity levels are declining’ and that ‘this decline in physical activity is a key contributor to the global obesity epidemic’.

Even this demonstrable fact is disputed by the revisionists. Fung cites a study of hunter-gatherers as evidence that modern man burns the same number of calories as our ancestors. There are a number of problems with this interpretation (see here for a few) and other studies of hunter-gatherers have found significantly higher energy expenditure, but it is not necessary to speculate about prehistoric man for us to see that calorie consumption has fallen over time.

In his 1946 essay, The Politics of Starvation, George Orwell noted that the average Briton was eating ‘about 2,800 or 2,900 calories a day’ despite rationing and a shortage of food that was on the verge of leading to civil unrest. This would be enough to fatten up most Britons today, which is why we are advised to eat just 2,000-2,500 calories a day.

This was not journalistic licence on Orwell’s part. Two years later, the British Medical Journal published a study which found that the average Briton lost weight if he consumed fewer than 2,900 calories. Unless you believe that human metabolism has evolved dramatically in the last 70 years, the only explanation for our grandparents eating more yet staying slimmer is that they burned more energy in their daily lives.

Few people joined a gym in the 1940s. They didn’t need to. Today, there are lots of gyms and lots of obese people, but this is not proof that physical activity is useless in preventing weight gain. Asking whether leisure-time exercise leads to weight loss is a much narrower question than asking whether physical activity is linked to obesity, but to answer it we need to look at the people who are exercising, not at the whole (largely sedentary) population. In the next post, we will do just that.

Part Three

Now let’s look at the effect of exercise on individuals. Fung – who coined the term ‘Calorie Reducation as Primary’, or CRaP, to describe ‘current obesity thinking’ – is unequivocal. In a series of blog posts entitled ‘The Myth about about Exercise’, he writes: ‘There are many benefits to regular exercise. Weight loss, though, is not one of the benefits‘ (italics in the original).

He cites three studies which found that burning off a certain number of calories did not result in a commensurate loss of body weight. He rightly attributes this to a degree of compensatory eating. In other words, exercise creates appetite which can lead to more calories being consumed.

But, in making this point, he downplays the conclusions of the studies themselves, all of which also make it clear that the participants who exercised lost a significant amount of weight.

The first of these studies concluded that ‘physical activity expressed as energy expended per week is positively related to reductions in total adiposity’. The second found that people who exercised most intensively did not lose more weight than people who trained less intensively, but the crucial fact remains that all the subjects who trained lost more weight than those who didn’t. All exercise groups had a significant reduction in waist circumference. Similarly, the third study concluded that ‘supervised exercise, with equivalent energy expenditure, results in clinically significant weight loss’.

By the time he gets to his third ‘Myth of Exercise’ post, Fung has almost given up on his claim that ‘there is no measurable association between obesity and physical activity’ and is instead arguing that exercise is merely less effective than dietary change.

He cites a 2007 study involving people who exercised for around 45 minutes a day. After a year, their body mass index (BMI) had dropped by 0.5 and 0.6 (for men and women respectively) while the BMI of the control group either rose or stayed the same. Moreover, those who exercised the most lost the most weight.

Fung views this amount of weight loss as trivial, saying: ‘Colour me unimpressed. Exercise is just not that effective for weight loss.’ That’s a matter of opinion, but it’s rather different to claiming that weight loss is not one of the benefits of exercise.

As the killer blow, Fung discusses marathon running, which he evidently sees as the ultimate test of the exercise/weight loss hypothesis. He relishes the chance to talk about a 1989 study of people who were training for one:
Endless chub rub (chafing between the inner thighs). Miles upon miles on the dreadmill.  But so worth it, right?  Average body fat loss for men …  5 pounds.  Average weight loss for women … zero.

What Fung doesn’t mention is that this study was not aimed at achieving weight loss and there is no evidence that the participants wanted or needed to lose weight.

The men had a healthy average BMI of 23.4 before they started training and the women were a slender 21.1. Both groups consumed significantly more calories while training, mainly from carbohydrates, and both groups actually did lose weight. The men lost an average of 2.7 kg and the women lost 0.9 kg, though the latter was not statistically significant. Both groups also lost fat as a percentage of body weight (from 16.6 to 13.4 per cent for men and from 24.9 to 23.6 per cent for women) although there were were too few participants for these results to reach statistical significance.

The literature on physical activity is very large and Fung is entitled to select any part of it to make his case, but if these studies debunk the claim that exercise helps people lose weight then you can only wonder what the rest of the literature says.

Unsurprisingly, other studies make the case for physical activity even more convincingly. Listing a few in chronological order:

• A randomised control trial published in 2000 found that ‘weight loss induced by increased daily physical activity without caloric restriction substantially reduces obesity’.

• A 2003 study from the US found that ‘moderate-intensity exercise sustained for 16 months is effective for weight management in young adults.’

• A 2004 study of overweight women from Singapore found that an eight-week exercise programme ‘significantly reduced body weight, body mass index, percentage body fat and waist circumference’.

• A 2005 review concluded that ‘Regular exercise can markedly reduce body weight and fat mass without dietary caloric restriction in overweight individuals.’

• A 2009 study of middle-aged women found that ‘body mass, body composition, waist circumference, and high-density lipoprotein cholesterol changed favorably’ after a 30 minute, five days a week exercise routine.

• A 2009 study of younger women found that physical activity was ‘associated with a reduction in long-term weight gain, and greater duration is associated with less weight gain’. Moreover, it found that ‘sedentary behavior independently predicted weight gain.’

• A 2012 study found that a moderate-intensity exercise programme reduced BMI by 2.4 per cent amongst post-menopausal women, rising to 10.8 per cent if combined with a reduced calorie, low-fat diet.

• A 2013 study from the US concluded that ‘supervised exercise, with equivalent energy expenditure, results in clinically significant weight loss’.

I could cite many more studies (and have before) but there is no point in labouring a point that should be obvious: if you burn off calories they cannot turn into body fat.

The simple, unavoidable fact of human physiology is that you can’t lose weight without creating a calorie deficit. Whether you do this by eating less, moving more or a combination of the two is a matter of preference. Some people may find it too difficult to change their diet while others may find it too difficult, or too time consuming, to start exercising. Some find it easy to cut out alcohol or sugar, while others find it easier to play more sport.

Since nobody denies that physical activity has health benefits that extend beyond weight management, it could be argued that a calorie burned is better than a calorie foregone, but the crucial thing is to eat fewer calories than you burn. If your preference is for radical dietary change then by all means make your case – but don’t let your interest in ‘calories in’ blind you to the importance of ‘calories out’.

No comments: