A lot has been written by various experts, journalists, and researchers about the role of exercise in treating obesity and for weight loss in general. Some of it is close to the reality, while increasingly more and more of it is spurious, unfounded, non-evidenced, magic fairy dust rubbish created all too often to sell products, pills, supplements, or training methodologies. In several articles exercise has even become the pariah of weight loss with some outlandish claims being made about all sorts of negative effects, again without any kind of substantive evidence to back these up. Read these sensationalist headlines with great care and skepticism. However, it isn’t just the Daily Mail behind this, even eminent science writer and author of Diet Delusion Gary Taubes has been quick to dismiss exercise for it’s effect on food consumption, asserting that exercise simply makes people eat more. An idea that is far less straightforward as it turns out when the evidence is reviewed. Taubes isn’t the only one, John Cloud’s provocative piece in Time magazine rubbished the benefits of exercise (mainly due to his own inability to resist eating a donut afterwards), and Emma John (Observer) and Sophie Morris (Independent) have both written pieces with a similar message. 

Researching exercise* and its effect on weight loss and then applying that research to the general population is a serious challenge. Studies that rely on self-reported data are immediately vulnerable to the sad truth that people tend to grossly overestimate their activity levels, while similarly underestimating their food intake. Monitoring food intake, energy balance, NEPA (Non-Exercise Physical Activity), and other lifestyle factors makes getting truly meaningful research very difficult. This skews results horribly and may well be a reason that the role of exercise in weight loss has been so understated over recent years. Lab studies have shown a bit more promise but leave us with another problem when it comes to prescription, in that without supervision most people simply will not stick to such rigourous routines and protocols. It was only a few years ago that the UK fitness industry jumped on the ‘Tabata Training’ bandwagon in an attempt to market a new form of training as being the magic bullet for, well just about everything including weight loss. Of course, Izumi Tabata never even measured body composition in his study, it was a study aimed at improving aerobic fitness using a 4-minute anaerobic training format that involved cycling to exhaustion for 8 x 20 second bouts. This didn’t stop all kinds of ‘Tabata fat-loss’ classes cropping up everywhere replete with misleading and incorrect information that tapped into that human desire to want to achieve more for less. As we will see high-intensity interval training, while it may be a more time efficient way for trained people to improve fitness, is of little use to the obese person. I’d venture that many of those who advocate these protocols have little or no experience of working with the obese population. 

* Exercise is a rather all-encompassing term, but the predominance of research has been done on aerobic training protocols. There is very limited data on resistance training specifically for weight loss and next to none on the obese population. 

Contrary to the journalistic despair at the sheer pointlessness of it all, exercise can in fact positively affect risk factors for health irrespective of weight loss; blood pressure, insulin sensitivity, triglycerides, cholesterol, bone density, mood, anxiety, depression, stress reduction, self esteem, recovery from injury and illness, occupational performance, shorter labour times and improved pregnancy measures, fasting glucose levels, and reduction of chronic pain, have all been shown to be positively affected by the introduction of exercise. A sentiment further echoed by the Cochrane Collaboration report on Exercise for Overweight and Obesity (http://summaries.cochrane.org/CD003817/exercise-for-overweight-or-obesity)  that concluded that “The results of this review support the use of exercise as a weight loss intervention, particularly when combined with dietary change. Exercise is associated with improved cardiovascular disease risk factors even if no weight is lost.”

As we have already discovered in Parts 1 and 2, this subject is far more complex than the media and many of the pundits on the topic would like you to think, and I am about to illustrate that. 

Exercise seems to have wildly different responses in different people. Unfortunately this information is often lost by people reporting (at worst) based on the abstract of a study rather than the entire study or (at best) a reporting of the mean data and not the individual results.

Luckily (and you might have just seen this coming) someone spotted this and decided to look a bit closer at the data and what actually happens. Here is a link to the complete study, the full text of which I have sitting in front of me (courtesy of the author himself, Prof. John Blundell). It is well worth a read:


Now, firstly this is a fascinating study with an excellent and very accessible discussion section that I encourage you to read in full yourself. In this study they took a group of 35 overweight men and women and got them doing 5 sessions a week of supervised exercise (in this case various formats of ‘cardio’ ranging from steppers to rowing and treadmills). Sessions were measured individually to produce an expenditure of 500kcal per session. According to energy expenditure vs intake calculations measured at week 0 it was anticipated that a weight loss of around 3.5kg could be expected over the 12 weeks.

Guess what, after 12 weeks the mean* weight loss was 3.7kg (+/- 3.6kg) so about as expected. However, when the results were looked at on an individual level things got a lot more interesting. At one end of the spectrum subjects lost as much as 14.7kg of weight, while at the other end several subjects managed to gain weight (albeit very small amounts). Overall though, just about everyone (with the exception of four individuals) lost weight, some more and some less so than others…however, even those who gained weight tended to lose fat and mostly gain lean mass as the below chart from the paper shows, only three subjects in the entire study gained bodyfat. 

The study authors go on to discuss some interesting theories and findings. One finding in particular was that those Non-compensators (those who tended not to increase energy intake in response to energy expenditure) tended to be heavier and fatter at the outset compared to the compensators (those who lost less weight than expected and tended to increase energy intake). One theory being that when exercise threatens lean mass (as it does in those only carrying a few extra pounds…) it drives the need to increase energy intake more than in those with an abundance of excess calories in body fat. While this notion is an appealing explanation I am not sure it is only the mechanism behind this, although it does make intuitive sense. We do also know that in some people there is a tendency to ‘reward’ exercise by being less active throughout the rest of the day, which can reduce potential weight loss results. This reward mentality does seem prevalent and is a strong driver for the argument that psychology is a prime factor in obesity. Studies looking at eating behaviour, such as those by Jane Ogden at Surrey University have shown how certain environments and situations can influence eating, often unrelated to our actual hunger. Simply eating in front of the television has been demonstrated to increase overeating irrespective of actual reported hunger levels. 

What this study does confirm is that in some people doing exercise and losing weight can serve to increase hunger, while in others this response is far less pronounced and in some non-existent. However, this does not mean that exercise makes you fat or even makes you eat more, far from it in fact. Firstly, even in the group who were compensators in this study, most of them still lost weight. Secondly, and perhaps most importantly, these subjects were allowed to eat as they pleased during the 12 weeks. Had their food intake been as strictly controlled and supervised as their exercise routine then I am very confident that the results would have been different (not to mention further proof that exercise does in fact work for weight loss when eating is controlled). Perhaps if this had been taken a step further and the study participants had been counseled on behaviours to avoid overeating, such as not eating in front of the television, then the results may have been even more pronounced. What if their sleep habits had been studied alongside the exercise? Would we have seen further improvements by attempting to improve this given the very clear relationship between poor sleep habits and weight gain? What the studies are clear on, is that exercise is a key factor in the maintenance of any weight lost and that the addition of it to any dietary regime seems to be an overwhelmingly positive addition given its effects on mood, sleep, confidence, depression, fitness, health risk factors, bone density, metabolic rate, and stimulation of lean muscle mass. 

What is capable versus what is optimal?

When it comes to working with the obese population and looking at exercise strategies it is really a case of what people will actually do versus what might be ideal. Given that the biggest benefits to the health of the population come from getting the sedentary folks moving and not from getting the already fit ones even fitter, we should be targeting the obese, in particular those who are totally sedentary, with far different strategies to what we advise an established trainee. 

If our target is to address the public health aspect of obesity then a reduction of the continual emphasis on exercise as solely a weight loss tool would be wise and instead more efforts to promote the overall health benefits would be a better option. It seems to me that there are many who are relatively apathetic about their actual weight, but are concerned about their overall mental and physical health and approaches exclusively aimed at weight loss will alienate these rather than include them. Health related goals are also far more tangible and should not be viewed as a by-product of weight loss, but rather the other way around. 

In terms of actual protocols, it seems conclusive that a combination of aerobic exercise, resistance training, and diet seems to have the widest reach in terms of overall health benefits and weight loss. Frequency of actual exercise is a key factor in predicting positive outcomes and has a strong inverse relationship with exercise intensity. If advising someone to exercise who is unsupervised and currently sedentary then simple and achievable is the way forward, there is little need for complex interval training routines or resistance training formats. The less the participant perceives their own ability to complete the routine the greater the chance that they will give up and return to their former weight. In contrast as someone begins to exercise and experiences benefits their capabilities will increase and their perceptions will alter, then the training can be progressed accordingly. It is all too often forgotten that the sedentary obese person will rarely have the joint and muscle integrity to withstand intense training at the outset anyway and any over-zealous routine could well be injurous, potentially setting the person back further than when they began. Despite this, many shredded fitness professionals (you know the types with their topless pics as avatars and disparaging attitudes towards anyone overweight) like to criticise simple approaches and continue to advise routines online that are inappropriate and show a poor level of both empathy and understanding. Expertise in bodybuilding does not constitute expertise at training obese people. 

Although interactions at a biochemical level are complex, exercise has clearly been demonstrated to have a positive effect on those seeking to lose weight. Some types of exercise seem more effective than others, swimming for example has been shown to raise appetite and hunger substantially compared to running. This effect though may be small in those who fail to give attention to other aspects of their lifestyle, while in others it may be more pronounced, albeit still below what people may be expecting for the time invested. Individual responses to exercise seem widely variable, and while it may cause some to eat more this view simply isn’t born out in the evidence as Professor John Blundell elegantly states in a meta-analysis on the topic  “Only 19% of interventionist studies report an increase in energy intake after exercise; 65% show no change and 16% show a decrease in appetite. Of the correlational studies, approximately half show no relationship between energy expenditure and intake. These data indicate a rather loose coupling between energy expenditure and intake. A common sense view is that exercise is futile as a form of weight control because the energy deficit drives a compensatory increase in food intake. However, evidence shows that this is not generally true.” 

As part of any nationwide intervention to address the obesity epidemic, exercise is a valuable and in my opinion, indispensable aid. By encouraging more sport and play in our young children we can help set a platform by which a healthful way of life becomes normal, without it always being connected to the images of ‘health’ in the media that tend to be only driven by the next skinny model or celebrity diet. In adults exercise can bring people together within communities and with the right prescription it can offer everyone a platform to succeed in improving their health and experiencing a better quality of life. It can have a massive effect on improving disease markers and can help with weight loss, although weight loss is not a pre-requisite for the health benefits. The actual what and when of the exercise aspect are very simple, it is getting people to do it and keep doing it that remains the complex part of the equation.