Animal experiments are useful tools to help elucidate complex physiological systems. For example, forcing animals to exercise can cause changes in energy balance. Animals are useful in this regard because their diets can be controlled and they can be forced to exercise on treadmills and in swim tests against their will. Humans studies investigating the effects of exercise on energy balance are less clear, and this likely relates to the inability to force food restriction or exercise on humans. In addition, controlling a free living human is more difficult that a rat in a cage. Epidemiology suggests that there is an association between body weight and physical activity in humans, such that those who exercise more tend to be leaner. However, the cause and effect of this relationship cannot be ascribed because while it is possible that exercise causes weight loss, it is just as likely that being lean facilitates participation in physical activity. Care must therefore be taken when trying to interpret human data regarding the cause and effect of weight loss and physical activity.
Another problem with measuring energy balance in humans arises from the fact that while direct calorimeters are the most accurate way of assessing energy expenditure, they are unsuitable for long term use. Therefore most studies rely on other measurements such as body weight, lean mass or body fat percentage, but these methods are not as reliable. Long term studies are also problematic because while it is possible to accurately record physical activity during exercise periods, it is less easy to report physical activity during non-exercise periods. Concerns have also been raised about the accuracy of self-reporting questionnaires and their ability to accurately relay information regarding food intake or physical activity. When changes in body weight occur, there is also a question as to whether the change is real or simply a reflection of natural fluctuation. The accuracy of measurements in this regard can also be questioned because unless the body composition change is large, meaningful measurements may not be possible.
Exercise intensity is pivotal in determining energy balance changes in humans. However, traditional models assume that all exercise can cause such changes, and that the intensity is simply a determinant of the rate of change seen. However, as exercise intensity increases, the physiological systems stressed during that exercise period change, and as such physiological reactions to the exercise can be quite different at different exercise intensities. For example, intense exercise causes depletion of glycogen, and this may have beneficial effects on insulin sensitivity. This is an important concept to understand because it is likely that insulin resistance is a pivotal factor in the development of obesity, and improvements in insulin sensitivity may therefore be beneficial at causing meaningful weight loss. However, lower intensity exercise that does not cause glycogen depletion may have no significant effect on weight loss. It should also be questioned as to whether men and women respond differently to exercise because of their hormonal differences.
It is also apparent that lean subject and obese subject respond differently to different kinds of exercise and different exercise intensities. While lean subjects respond to exercise with reductions in subcutaneous fat, this is not always apparent in obese subjects. This may relate to hormonal difference because it is known that obesity is associated with both insulin and leptin resistance and that other hormones may be in a state of dysfunction in the obese individual. Because abdominal fat can only be reduced through improvements in the quality of nutrition, and is not affected by exercise, obese individuals with large amounts of visceral fat may not be expected to undergo the same responses to exercise as lean individuals with mainly subcutaneous fat. Lean individuals also appear to undergo only very slight changes in body weight during exercise periods, and this may relate to the increased energy intake that often accompanies periods of physical activity. This is a normal regulatory mechanism to maintain a stable body weight through hypothalamic regulation.
The inconsistent results from experiments assessing the effects of exercise on body composition in obese individuals is therefore problematic. It suggests that exercise is not effective at reversing the obesity disease, and this may relate to the fact that it is not overeating, but eating a low quality diet, that is the cause of excessive weight gain. This is highlighted by the fact that while energy intake appears to be linked to physical activity levels in normal weight people, there appears to be a disconnect between the energy expenditure and energy intake in the obese. Therefore it cannot be predicted that exercise will cause a specific result because the normal regulatory mechanisms that govern the ability to maintain body weight are absent or diminished. That energy intake likely increases with increased physical activity in the obese, as with normal weight individuals, is also problematic because it shows that attempts to force the hypothalamus to accept a lower energy balance is not possible, and exercise may actually lead to weight gain in the long term.
Another major flaw in the science of weight loss is the experimental bias inherent in all studies investigating weight loss in humans. Drop out rates from studies are often high, but only those who successfully complete the programme have their results measured. In the real world setting a high dropout rate for a particular exercise regimen would signify failure based on the poor adherence. However, this is often not considered when measuring the parameters of those determined enough to complete the protocol. Bias in the selection process is also evident as often only those committed to losing weight may volunteer for the trial and this produces a problematic methodological dilemma that cannot be completely ignore when looking at the results. The lack of adequate control and true placebo groups in weight loss trials is also evident. Comparing a diet group with a diet plus exercise group is not greatly useful without a control group as it could be argued that too many confounding variables interfere with the observations that can be measured.
Another problem associated with studies investigating weight loss is the short duration of the experimental phases in some intervention studies. Generally ‘weight loss’ is possible on most exercise regimens in the short term. However, this weight loss often comprises mainly of lean tissue, and in the long term such subjects often gain back any lost weight. However, by curtailing the studies before this occurs it can be made to look like an intervention is successful. Often such studies do not report lean mass changes or body fat percentages which gives the impression that weight loss is beneficial when in fact it may be lean tissue, losses of which are detrimental to metabolic rate and long term metabolic function. It is also interesting that most research investigating ‘weight loss’ chooses to use aerobic style low intensity exercise, avoiding more intense forms of training such as resistance training and sprinting. These more intense forms of exercise train different physiological systems to aerobic style exercise, and as such may be more beneficial at causing fat loss while maintaining lean mass.
RdB
