Biology Online is a Biology blog and dictionary site that provides up to date articles on the latest developments in biological science. The Biology Online Dictionary is a completely free and open dictionary with over 60,000 biology terms. It uses the wiki concept, so that anyone can make a contribution.

The Real Causes of Obesity?

A very warm welcome to noted human performance expert and Mayo Clinic physician, Dr. Michael Joyner

Is Obesity Research a Dead End?

Obesity is a big public health problem. Currently about two-thirds of American adults are either overweight or obese. While this rate appears to be stabilizing, it is also likely to have a big impact on early mortality and be an overall drag on public health statistics [MJJ1] in the United States.  Additionally, the rest of the world appears to be “catching up” as more and more people adopt a lifestyle that includes low levels of physical activity and easy availability of cheap calorie-dense food. With the exception of surgical approaches, the medical world does not have much to offer in terms of treatment for obesity.  There are no good or widely used drugs that work.  So the question is what does biomedical research have to offer the obesity problem?

 Human Studies: The Wrong Focus?

The big thing in human research is what kind of diet works best.  Low carb, high carb, high protein, you name it.  A lot of this is about the search for a magic bullet diet with a macronutrient composition that somehow makes it easy for people to lose weight and keep it off.   Much of this focus is also driven by diet industry and media hype.  In terms of what diet works best, the short answer is both none of the above and all of the above.  The key for any diet (and also exercise program) is adherence to a few fundamental principles[MJJ2] .  So perhaps we need a truce in the diet wars [MJJ3] and more focus on adherence and also the ecology of the obesogenic food and physical activity environment that surrounds us.

Animal Studies: Wrong Models?

A lot of obesity research in animal models uses either rats or mice.  There are mice and rats that have either been bred or genetically modified to develop all sorts of metabolic disorders associated with obesity.  Typically, these animals have some sort of very dramatic phenotype and one or more pathways can be identified that are possible “targets” for intervention.  When these targets are evaluated in humans, sometimes there are natural and rare human “experiments in nature” where a gene defect that causes a clearly inherited phenotype is identified.  More frequently a gene variant in the pathway of interest is identified that might be associated with a minimal increase risk in the population.  Typically, it has been difficult to translate these responses to medically actionable interventions.  Why?  The simple answer is that there are three main problems with rodent models in obesity research:

 1)    One of the factors that dominate basal metabolic rate is the ratio of body surface area to body volume.   Animals lose heat through their surface area and the smaller the animal the higher the surface area to volume ratio.  Thus, the smaller the animal the faster the heat loss to the environment[MJJ4] .  Along these lines, rodents have essentially two-log differences surface area to body weight ratio in comparison to humans.  This means that they need much higher basal metabolic rates normalized for body weight just to maintain core body temperature at a level similar to ours.  It also means that thermoneutral conditions vary dramatically.  This means that the standard animal quarters (~20-22°C at most institutions) are in fact cold for rats and mice.  Thermoneutral for a rodent is typically ~30°.  So the first thing we have to recognize is that most rodent models include unwitting chronic cold stress [MJJ5] which causes a number of physiological and behavioral adaptations that might confound the results and interpretation of studies on obesity and metabolism. 

2)    The next thing we need to recognize is that most animal models are cage-reared and do not have access to much physical activity.  This is not about specific exercise training.  It is about what mice and rats do at night, which is essentially run as much as they can.  When rodent models of obesity and other metabolic disease are given access to voluntary running wheels, frequently the phenotype of interest does not appear[MJJ6] .  So what is being studied obesity or unusual adaptations to physical inactivity amplified by whatever genetic modification the animal strain of interest has?

3)    With the exception of a few rare syndromes, garden-variety human obesity and type 2 diabetes do not have a clearly identifiable genetic footprint[MJJ7] .  So, while animal models of obesity might identify targets for drug therapy, there are a lot of “redundant” mechanisms contributing to obesity, making me skeptical that a “magic drug bullet” focused on a discrete target will ever be identified.


Obesity studies in humans are too often focused on macronutrient composition and not the behavioral, environmental, and cultural factors associated with successful weight loss.   Animal studies frequently use small, cold and inactive animals which might limit that applicability of any findings to humans.  Both human and animal studies are overly focused on the genetics of obesity and not the obesogenic environment that leads to it.

More from Dr. Michael Joyner can be found at: