Obesity Crankery – A growing problem

Recently, it seems there has been a backlash against medicine and the current knowledge of the relationship between diet, weight and overall health. I don’t actually believe this is directly the fault of scientists or doctors, who react to the trashy mainstream reporting of science with little more than the occasional raised eyebrow. However, many people in response to all these silly health pronouncements, which seemingly come from on high but really are from press coverage of often minor reports in the medical literature, have lost their trust in what science has to offer as a solution to what Michael Pollan refers to as “the Omnivore’s Dilemma”. That is, what should we be eating?

The result of this confusion is a mixture of distrust, cynicism, and receptivity to crankery and lies about diet. After all, if science ostensibly can’t keep their message straight, who knows what to believe?

The fact is, science knows many things about the relationship between diet, obesity, and health with great confidence and it hasn’t changed nearly so much as the popular press would have you believe. The failure to state clear messages about nutrition is a reflection on the haphazard way in which nutritional health is reported, the often confusing nature of epidemiologic science, and the various parties that are interested in cashing in the confusion by promoting their own nonsensical ideas about diet.

Take, for example, Sandy Szwarc. Sandy doesn’t believe obesity or any food choices are actually bad for you. To help spread this nonsense she dismisses valid sources of information like WebMD (which has quite good information) based on the rather silly conspiracy that they have designed their entire website and health enterprise around misleading people into using their products – especially weight-loss products. Because, you know, it’s impossible for a corporation to offer free health advice as a public service without conspiring to grab you buy the ankles and shake the money from your pockets. But it doesn’t end there. We see rest of the standard denialist tactics of course!

Case in point, in a recent article she makes the astonishing assertion that her mortal enemy – bariatric or gastric-bypass surgeons – have admitted that obesity makes you healthier!

Today brought another unbelievable example of ad-hoc reasoning, as well as a remarkable admission that the war on obesity is without scientific merit. It appeared in a paper published in the journal for the American Society for Bariatric Surgery (now calling itself the American Society for Metabolic and Bariatric Surgery), which is edited by the Society’s president, Dr. Harvey Sugerman, M.D. FACS.

The article, “Do current body mass index criteria for obesity surgery reflect cardiovascular risk?” was “work presented at the 2005 American Society for Bariatric Surgery Meeting in poster form.” The authors, led by Edward H. Livingston M.D. at the University of Texas Southwestern School of Medicine, reported that the conventional risk factors for cardiovascular disease “decreased with increasing degrees of obesity.”

Yes, you read that correctly, decreased.

“Therefore,” the authors argued …

“the criteria for obesity surgery should be changed to lower BMIs than are currently used.”

Now, boys and girls, what is the very first thing you do when a suspected denialist feeds you some nonsense in quotes? Check the source! Always, always, always, check the source. Let’s expand those six words that Sandy lifted out of the abstract and see what else the authors had to say:

Obesity is an underlying cause of the development of cardiovascular disease (CVD). Obesity itself does not result in CVD, rather, it acts through intermediate-risk factors. [ed – note this, it becomes important later] Most, but not all, studies examining the obesity-CVD relationship have found them to be correlated. We hypothesized that the inconsistencies among the studies of the obesity-cardiovascular relationship were attributable to an irregular relationship between obesity and the presence of risk factors for CVD.

Of the conventional CVD risk factors, blood pressure, serum glucose, and waist circumference increased linearly with adiposity. The reverse was observed for high-density lipoprotein. Insulin, C-peptide, apolipoprotein B, non-high-density lipoprotein cholesterol, low-density lipoprotein, and serum triglycerides all peaked in the body mass index range of 30-40 kg/m2 and then decreased with increasing degrees of obesity. [ed – this is all that Sandy took out of this abstract!]


Cardiovascular risk factors are markedly increased for many individuals with a body mass index >30 kg/m2. Massively obese individuals might have better CVD risk profiles than less obese individuals. The discrepancy is related, in part, to body conformation. The criteria for obesity surgery should be changed to lower BMIs than are currently used if patients have risk factors for CVD.

Sandy goes on to suggest that the conclusions were a “twist of logic” etc. Well, yeah, if you take 6 words of an abstract out of context, ignore that risk factors peaked with BMIs between 30 and 40, and take the interesting finding that some of these risk factors (but by no means all) are decreased in the > 40 group completely out of context, yes you could possibly find a twist of logic. There is no other way to describe her cherry-picking of this article as anything but gross mendacity.

Sandy makes a lot of hay out of articles which show, as the authors alluded in this abstract, a weaker link between obesity and mortality/cardiovascular disease. Her secret is of course to ignore that first bolded sentence from this article, which is absolutely true. I’ll quote it again, “Obesity itself does not result in CVD, rather, it acts through intermediate-risk factors”. This is absolutely right, a very clear and concise description of the problem. Obesity itself does not cause death! Obesity causes high blood pressure, diabetes and insulin resistance. There are a host of other problems it causes or worsens, but most importantly, diabetes and insulin resistance which then cause atherosclerosis. Diabetes also does not result directly in mortality – few people die of diabetic ketoacidosis or hypoglycemic shock – however, 80% of people with diabetes will die of cardiovascular disease.

So what happens when you study an obese population for mortality or cardiovascular disease risk? It becomes difficult – especially in more recent studies – to identify the obesity-disease relationship. Why? Because when you do a study or a survey on obesity, you don’t force the obese group to stop taking their BP meds, their cholesterol meds, their diabetes meds etc., while you are studying them to see if obesity is risky. What we’ve learned since the first major national study of health (NHANES I) is that when risk factors like blood pressure, cholesterol and diabetes are controlled, the risk of death decreases dramatically from being overweight. See this figure from the most recent NHANES trial:


As you can see, we’re getting better at keeping people alive generally. The data from NHANES I and II weren’t wrong, in the last 20 years there has been a great emphasis put on treating the risk factors of the obese rather than just constantly yammering at them to lose weight. Sandy is correct, it is very difficult to make people lose weight, and attempts to design a scientificaly proven effective weight loss trial have been largely futile. That doesn’t mean it’s a bad idea to lose weight, that it is impossible to lose weight, or that people shouldn’t be encouraged to do so. You wouldn’t discourage a smoker from quitting just because it’s difficult to make people quit smoking. Some will, and they will benefit from the change.

The other issue is that she keeps on alleging a protective effect from obesity because all those skinny people in trials of heart medications keep on dying faster. For someone who complains about correlation not being causation, it’s funny that she suggests obesity is therefore protective.

If these bariatric researchers are concerned about heart disease deaths, they only have to look at the body of evidence showing a protective effect of obesity among heart patients, even those with the highest rates of diabetes, high cholesterol and high blood pressure. The huge randomized INVEST trial of people with heart disease and hypertension, for example, demonstrated that the obese, BMIs 30 – <35, had the lowest risks for all-cause mortality, heart attacks and strokes -- nearly half that of "normal" weight patients. Only at the very highest BMIs did the risks begin to creep up but they were still less than the overweight and most notably less than the 'normal' weight patients.

As has been made clear, again and again, obesity doesn’t get you, it’s the secondary effects. What happens when, like in these trials like INVEST, the blood pressure, diabetes, and cholesterol are controlled with drugs (remember – they have higher rates but in a trial these conditions are being managed with drugs!)? Well, they’re going to live just as long! And in cohorts of older folks, skinnyness is a bad sign. What happens to old folks before they die? They shrink. You think I’m kidding, work in a nursing home. Weight loss and skinniness in the elderly is an indication of impending doom, not good health. A higher BMI group, therefore, selects for a group that is less likely to have chronic illness or is experiencing wasting and cachexia. It selects for a healthier group in the elderly! A similar obesity protective effect is often shown with smokers and is just as specious.

Further, I can’t figure out why she cites the INVEST trial in this particular instance. INVEST was a huge trial and has resulted in multiple publications based on post-hoc analysis of its data. I’m sure one of the analyses provided some result for her to cherry-pick and read out of context. It’s bad for Sandy, but the predominant emphasis of some of these analyses was that control of diabetes (with higher risk of DMII in obese cohorts) was one of the most important factors in controlling cardiovascular disease. And how do we avoid diabetes? Maintaining a reasonable weight!

She ends with a little conspiracy suggestion in this article too – I love it! Such denialism.

*** Readers may not know that the University of Texas Southwestern Medical Center, where a lot of recent bariatric and weight loss promotional papers have been originating, was awarded a $22 million grant this past September for its Task Force for Obesity Research.

What pisses me off most about Sandy is that some people think that she’s a real skeptic. Despite the illusions of persecution, the David and Goliath complex, the cherry-picking, the alleged conspiracies, not to mention the prominent link to Steve Milloy’s Junkscience from her homepage, they can’t figure out that this is garden variety denialism. She gets regularly featured in our very own Skeptic’s Circle, and shouldn’t we be the ones best at smelling a rat? Clearly not. Part of the problem is she doesn’t always submit the really cranky stuff on obesity to the circle, and even a broken clock is right twice a day. The other problem is the hosts are rarely able to investigate each blog that sends a nomination.

All the same, I’m afraid I’m going to have to put my foot down and say, if she is linked from a future Skeptic’s Circle, I’m not going to link it. Don’t lend this person any more credibility than she deserves by calling her a skeptic, unless you want to also include HIV/AIDS skeptics, Global Warming Skeptics, and all the other skeptic wannabe’s out there.

Finally, as far as the original question of what should we be eating, the answer, I think is pretty straightfoward and hasn’t changed much. You should eat a nutritionally-balanced diet, low in fat, and your caloric intake should not exceed your caloric output. I like Michael Pollan’s advice. Roughly, eat what your parents did, not too much (take in consideration the change in physical activity), and avoid fads. Ultimately, the worst problem with all the fad diets, stupid reporting, and nonsense from people like Sandy who say exercise causes weight gain, is that they’re at least tangentially suggesting the human body violates the laws of physics. While the efficiency of metabolism might vary slightly between individuals, it’s not going to vary by a huge amount, and people are always exaggerating the amount they excercise versus the amount they eat to justify their “slow metabolism” (read: violation of fundamental laws of physics). If you eat more than your baseline metabolic requirement plus your daily caloric output, you will gain weight. If you eat less than your baseline metabolic requirements plus your daily caloric output, you will lose weight. Anything else, and you’re suggesting energy is being created or destroyed, and I just don’t have time for such nonsense.


P.S. I recently acquired Gary Taub’s new book attacking nutritional epidemiology (not a difficult task), and plan to evaluate it for crankery. While I know I disagree with Taub based on what I’ve heard him say so far, I haven’t heard him use the tactics. Generally, nitpicking is his major fault, but that in itself isn’t enough to justify reproach.