Has Obesity Research Failed?

I frequently encounter the argument that obesity research has failed because it hasn't stopped the global increase in obesity rates.  According to this argument, we need to re-think our approach to obesity research because the current approach just isn't working.

Grant funding for obesity research keeps increasing in the US, and the prevalence of obesity also keeps increasing*.  What gives?  Maybe if we just scrapped the whole endeavor we'd be better off.

Let's take a closer look at this argument and see how it holds up.

Why Do Research?

There are two fundamental reasons why we do research:

1. To gather accurate information about the natural world.  This information is intrinsically valuable because we like knowing how the world works, and it may eventually have practical value that's not immediately obvious.
2. Practical applications.  We want to solve problems and improve our lives.

If we want to determine whether or not obesity research has failed, we should evaluate it using those two metrics.

Has Obesity Research Gathered Accurate Information?

By this metric, there can be no question that obesity research has been a smashing success.  Starting as far back as the 1700s**, and accelerating rapidly until today, researchers have uncovered a tremendous amount of information about the human body that relates to obesity-- from digestion, to the thermodynamics of energy balance, to the physiology and endocrinology of energy regulation by the body, to the neurobiology of appetite and body fat regulation, to the psychology of eating behavior.  

We understand many of the hormones and neural signals that regulate the body's use and storage of fats, carbohydrates, and proteins.  We understand many (probably most) of the primary pathways that regulate appetite.  Researchers like Scott Sternson, Brad Lowell, and Richard Palmiter are dissecting the neural circuits of appetite down to the cellular and molecular level-- a feat that would have been inconceivable 50 years ago.  

We understand the impact of a number of food properties on food selection and intake.  We understand how the food environment impacts eating.  We've identified countless differences between obese and lean people, including in genetics, neurobiology, energetics, metabolism, endocrinology, and behavior.

Collectively, the information we've gathered forms a massive, highly buttressed body of knowledge that meshes seamlessly with related fields of science.  Some of the details are probably wrong, but the likelihood that there's something fundamentally wrong with the whole edifice, is, in my opinion, approximately nil.  

The current literature contains more than enough evidence to form a reasonable working model that explains the obesity epidemic.  But explaining isn't the same as doing something about it.

Has Obesity Research Delivered Practical Value?

In an ideal world, obesity researchers would have discovered techniques for easily preventing and reversing obesity.  Similar to how researchers identified the nutritional cause of pellagra (niacin deficiency) and virtually eliminated the epidemic with vitamin fortification in the early 20th century, perhaps we might have identified some agricultural chemical that's behind it all, and eliminated it.  Case closed, great work, high five, let's call it a day.

Clearly, that hasn't happened.  Despite the mountains of information we've accumulated, we still have an obesity epidemic, and no easy cures.

On the other hand, researchers have actually identified a number of effective techniques for preventing and reversing obesity.  The simplest of these is calorie restriction.  A number of tightly controlled studies show that calorie intake has a powerful influence on body fatness.  The problem with restricting calories isn't that it's ineffective-- in fact, it's 100 percent effective.  The problem is that it's difficult to implement and maintain.  The unconscious energy-regulating parts of the brain fight back by increasing hunger and reducing the number of calories expended, and ingrained preferences and habits undermine long-term adherence to the restriction.  Increasing calorie expenditure via exercise has similar effects.

A large proportion of obesity research has been dedicated to trying to figure out ways to get people to eat fewer calories.  Countless diets and behavioral strategies have been tested in randomized controlled trials.  If there's one consistent finding that has emerged from these studies, it's that it's exceedingly difficult to get people to change their behavior sustainably, particularly when the change involves a diet.  We know people lose fat when they eat less, but it's difficult to get them to do so, regardless of the strategy.  That being said, some strategies are clearly more effective than others.

Another effective technique is bariatric (weight loss) surgery.  The Roux-en-Y and sleeve gastrectomy procedures cause remarkable fat loss and resolution of type 2 diabetes-- far superior to any other method that has ever been tested in free-living people.  This isn't because people can't fit as much food in their stomachs, and it also isn't because more calories end up in the toilet.  It seems to relate to (poorly understood) alterations in the communication lines between the gut and the brain.  The problem with bariatric surgery isn't that it doesn't work, and it isn't that it's unsafe***-- the problem is that most people don't want to have major surgery on their digestive tract, and it's only available for extreme cases.  Researchers are currently trying to understand the mechanisms of bariatric surgery so we can mimic it without surgery, and they've made some encouraging progress.

Research has also uncovered a number of effective anti-obesity drugs.  One of the earliest was 2,4-dinitrophenol, which causes the mitochondria to pump out heat instead of chemical energy, wasting calories.  It's effective drug but it caused too many deaths due to overheating.  Apparently some bodybuilders still use it, illegally.

In recent decades, a number of drugs were developed that act on the brain pathways that regulate appetite.  You may have heard of fen-phen (fenfluramine/phentermine), an effective weight loss combo that increases monoamine neurotransmitter (dopamine, serotonin, norepinephrine) levels in the brain.  Fen-phen was taken off the US market in 1997 due to negative side effects-- primarily its ability to promote heart valve disease.

Rimonabant is basically "reverse marijuana", acting to block the same receptor that gives people the marijuana munchies.  This pathway regulates appetite and pleasure.  It was an effective weight loss drug, but perhaps unsurprisingly, it had adverse psychological effects including increased suicide rates.

Currently, we have three obesity drugs approved for the US market, and more in the pipeline.  Lorcaserin acts on an appetite circuit in the brain, and seems fairly innocuous, but its effect is pretty modest.  Phentermine/topiramate (Qsymia) acts on appetite and reward centers and is a fairly effective drug, but it has significant negative side effects.  Orlistat blocks about 1/3 of fat digestion so that some of it ends up in the toilet.  It's relatively safe but the results are modest.

Liraglutide is a drug that has done well in clinical trials of obesity treatment.  Novo Nordisk has filed for FDA approval of the drug for obesity treatment purposes.  It's already approved for the treatment of type 2 diabetes.  It mimics a natural satiety (and insulin-secretion) factor called GLP-1.  It's relatively safe as far as we know, and some of its known side effects are actually positive, but unfortunately it's expensive and it has to be injected daily.  It may be available for obesity therapy relatively soon.

Contrave (bupropion/naltrexone) is another drug that may gain FDA approval soon.  It acts on appetite and reward centers in the brain, and its effectiveness seems to be somewhere between lorcaserin and phentermine/topiramate.

It's clear that obesity research has uncovered a number of effective obesity therapies.  At a minimum, the effectiveness of these treatments demonstrates that our current scientific models of obesity are correct.  Yet all of them have major drawbacks, either related to limited effectiveness or negative side effects.  None are the super easy, 100 percent effective magic bullet we would like.  Why not?

The reason we don't have a magic bullet is that obesity is a difficult problem.  Preventing and treating obesity means fighting the natural tendency of the human body and mind in the context of our current culture.  You can tell people to eat less sugar, white flour, added fats, and processed foods in general, but only a minority of people will actually alter their behavior significantly as a result.  This is because people don't eat junk food for its health benefits-- they eat it because they like it, it's cheap, and it's readily available (obesity isn't caused by eating junk food in all individuals, but it is a major cause on a population level).  Obesity is much more challenging than a simple infectious agent or nutritional deficiency that can be readily treated.

Since obesity is largely related to brain activity, we typically attempt to treat it using drugs targeted to the brain.  These approaches work, but they're limited by the fact that the brain is extremely complex, it's protected by the blood-brain barrier, and drugs are typically not specific enough to hit the relevant circuits without collateral damage.  Bathing the entire brain in a drug is a sledgehammer approach that tends to be too non-specific for therapeutic purposes, although there are occasional exceptions****.

Personally I think it's inevitable that we'll develop better obesity therapies, but these will have to swim upstream against our increasingly fattening culture.

Has Obesity Research Failed?

I hope it's clear that obesity research has not failed-- it has produced huge amounts of scientifically robust information, and a number of effective therapies.  None of these therapies are the magic bullet we wish they were, yet there's no reason to believe this is because our understanding of obesity is fundamentally flawed.

The reason we don't have a magic bullet is that obesity is a tough problem to solve.  Researchers can identify causes, but they can't force people to change their lives.  They can develop effective treatments, but these tend to cost a lot of money and/or have serious drawbacks.

Measuring the success of obesity research in terms of its ability to prevent or reverse the obesity epidemic is setting an impossibly high standard.  In a sense, I believe this is the point of the argument.  By setting an impossibly high standard and rejecting the field as a whole, this allows people to excuse themselves from the time and effort required to understand the subject.


* Although NIH funding in general, including for obesity research, has dropped in recent years.  Hopefully that trend will reverse as the economy bounces back.

** Antoine de Lavoisier was one of the first researchers to contribute meaningfully to the modern understanding of obesity.

*** There is a significant risk of surgical complications and death from bariatric surgery, however the complications and death risk of remaining extremely obese are much higher.  Therefore, relative to remaining extremely obese, bariatric surgery is very safe.

**** There are many therapeutic drugs that act successfully in the brain.  I use the word "occasionally" because for every drug that is currently in therapeutic use in humans, there are probably hundreds that didn't make the cut.  Among those that work, most are blunt tools.