On fat acceptance and obesity research; or, lying about the science is just another form of fat bigotry

Obese individuals don't deserve to be bullied. But they don't deserve to be lied to, either.
Obese individuals don’t deserve to be bullied. But they don’t deserve to be lied to, either.

Health media and bloggers are a-buzz over a recent study by Flegal and colleagues of the National Center for Health Statistics at the CDC. The study analysed data published in other papers in an effort to build a comprehensive dataset and determine whether BMI corresponds with mortality. In short, the broad analysis revealed that (not surprisingly) morbid obesity is associated with significantly higher likelihood of death; however, the study also found that the “overweight” BMI category is associated with a significantly lower (~6%) risk of death relative to “normal weight” individuals.

With this single statement in the study’s abstract, the primary investigators stirred a massive controversy. Obesity researchers have called this study “rubbish” while self-identified fat activists have taken to CNN to proclaim this paper a definitive blow to institutionalized fat bias.

Before I dive into this controversy, let me first remind us of a few things. The Flegal study used the BMI — or “Body Mass Index” scale — to categorize subjects based on their body weight. BMI is calculated from a person’s height and weight to generate a number that typically falls between 18 and 40; that number then allows you to plot your index against a table that let’s you classify yourself as “normal weight”, “overweight” or “obese”.

An "overweight" person  (called "slightly overweight" in this table) typically has a BMI that falls between 26-29, and an "obese" person has a BMI greater than 30.
An “overweight” person (called “slightly overweight” in this table) typically has a BMI that falls between 26-29, and an “obese” person has a BMI greater than 30.

Now, there’s a simple explanation for the findings of the Flegal study — the BMI scale is deeply flawed. The BMI scale is, in fact, a horrible tool for predicting an individual’s fitness or health status. The Flegal study is simply pointing out how the BMI scale is both over-used and incorrectly interpreted by primary physicians.

BMI was first created as an epidemiological and clinical tool that allows for rapid screening of a patient’s weight status, as a first-step indicator of whether or not a person may be at higher risk for obesity-related diseases. It’s strength was that the input data — height and weight — could be collected with a simple scale and tape measure, and you could arm patients with BMI as a way of monitoring their own fitness level at home. Further, the reason for its existence is that en masse,  high BMI correlates with a host of chronic diseases you really don’t want.

The problem is that in the non-scientific world, the correlative nature of the BMI scale has been confused with causation. While scientists still understand how BMI is (and is not) correlated with poor health and morbidity, doctors and the average citizen puts too much emphasis on BMI, mistaking the fact that having a high BMI means one may be at risk of also having other obesity-related diseases with the misguided notion that having a high BMI will cause those obesity-related diseases.

On an individual level, your BMI is relatively meaningless, precisely because BMI doesn’t directly test the actual health factors that lead to the development of chronic disease — blood pressure, diet, body fat percentage, and lifestyle choices. The BMI scale fails to consider variations in people’s body compositions (by definition, it assumes everyone’s muscle mass, bone mass, and hydration level are the same) in order to produce a rough guesstimate of whether or not a person might be obese or not.

According to the BMI scale, heavily muscled (and clearly fit) individuals like Arnold Schwarzenegger would be considered morbidly obese.
According to the BMI scale, heavily muscled (and clearly fit) individuals like Arnold Schwarzenegger would be considered morbidly obese.

Given these considerations, it’s not surprising that the Flegal study found that folks who fall into the “overweight” category of the BMI scale might have (en masse) a slightly lower risk of death than “normal weight” individuals: “overweight” individuals will include a large proportion of bonafide “slightly overweight” folks, but this group will also include a “contaminant population” of strength training athletes, people who have better access to food because of better socioeconomic status, folks who may eat a more varied nutrient-dense diet, and perhaps even younger individuals (who tend to have more muscle mass). The Flegal study did not — actually could not — tease these confounding factors out.

The bottom line is this: it is quite possible to be a healthy individual and to appear “fat” on the BMI scale. However, it’s also clear that folks who are both fit and fat typically have unusual body composition and weight distribution and/or have spent at least some of their daily energy focusing on maintaining a fit and active lifestyle that result in better heart health, good nutrition, and high muscle mass; even if they still carry a few extra pounds of fat. You can be fit and fat, but it is not the default.

What troubles me in this debate is, in fact, the opposite end of the spectrum, personified by self-identified fat activist Marilyn Wann is equally as alarming as the popular misconceptions about the BMI scale. Wann write in an editorial on CNN:

After a careful review of all relevant research worldwide, the U.S. government’s leading analyst of weight data just confirmed what I’ve long known: Being fat might not be a death sentence.

That this study reported in the Journal of the American Medical Association seems at all shocking is a measure of the intensity and pervasiveness of weight prejudice in our society and in our sciences.

In short, Wann weighs the (questionable) findings of this single JAMA article against an entire body of obesity research literature, and finds the thousands of published papers showing a clear connection between obesity and poor overall health to be lacking; worse yet, she lobs a veiled accusation against the scientific community of, in essence, falsifying data based on internalized “weight prejudice”.

This is, in every facet, equivalent to climate change denialists touting the findings of a single, potentially flawed, study with “favourable” findings against reams of evidence that support global warming.

Despite the headline of her article — “You Can be Fit and Fat” — Wann fails to make a single compelling argument that in favour of her overall thesis that obese individuals are just as healthy as non-obese individuals. Instead, she spends the bulk of her piece in CNN speaking about weight prejudice and fat tolerance, telling anecdotal tales of individuals whose primary health advice from medical professionals is to lose weight. Wann goes so far as to suggest that obesity is not, itself, a risk factor for disease, but rather that fat individuals have poorer health because hospitals aren’t equipped with blood pressure cuffs that fit around larger arms:

One woman called in the middle of the night, hoping I knew of an MRI she could use for an important test. The machine at her local hospital, which she’d used before, was being guarded by a technician who strictly enforced the weight limits. The tray that slides in and out of the machine could break. Instead, she was denied potentially life-saving information in a crisis. How many of the deaths blamed on weight are actually caused by medical equipment — everything from blood pressure cuffs to surgical instruments — that fails to accommodate fat people when we need it most?

I’ll never forget the teenage girl who was told by a nurse practitioner that her complaint would go away once she lost weight. Luckily, she had the nerve and the parental backup to get another appointment and the prescription necessary to treat her condition. How many of the deaths blamed on fat actually happen when people are diagnosed as fat instead of being diagnosed and treated for an illness?

Wann’s argument is that obesity is not, in any way, related to fitness. She notes a study that showed that age, not obesity, was the primary cause of death in patients over the age of 65; she forgets that obesity tends to cause early death before the age of 65, and that cardiovascular disease (which arises in part from obesity) is the leading cause of death in America.

Let’s be absolutely clear. Obesity is strongly associated with metabolic disorders (with the latter likely causative of the former). Metabolic disorder is likely the principal cause of cardiovascular disease. Fat individuals are dying because — without being too hyperbolic — obesity kills.

The mouse on the left is the famous leptin mouse, a classic model of obesity. He basically suffers from metabolic disorder leading to profound obesity. The leptin mouse is a notoriously sick mouse, suffering from hypertension, diabetes, acute inflammation, arthritis, and signs of rapid aging. This mouse is not sick because we don't have access to a scale big enough to weigh it.
The mouse on the left is the famous leptin mouse, a classic model of obesity. He basically suffers from metabolic disorder leading to profound obesity. The leptin mouse is a notoriously sick mouse, suffering from hypertension, diabetes, acute inflammation, arthritis, and rapid aging. This mouse is not sick because we don’t have access to a scale big enough to weigh it.

Here’s the thing: Wann is completely on the mark when it comes to wanting to fight against fat intolerance. Obese individuals should not be targets of society’s derision and ridicule. Obese individuals need access to adequate healthcare, insurance, and resources for healthy weight loss. Obese individuals shouldn’t be made to feel shame or self-hatred for their obesity, or to be associated with such negative terms as “lazy”, “unintelligent” or “ugly”. This is about as productive, and as charming, as throwing rocks at the fat kid in the playground until he breaks down and cries. It’s bullying, plain and simple, and it’s nothing I want to be teaching my kids.

But, isn’t it demeaning, patronizing, or — dare I say it — even intolerant to advocate that obese individuals should be treated to a selective interpretation of the science because they don’t deserve to hear the truth that obesity is, demonstrably, a risk factor for one’s health? Doesn’t is suggest that obese individuals can’t, or shouldn’t be, given all the facts about their own medical status because their self-esteem can’t handle it? Doesn’t it imply that we should lie to obese individuals because it’s mean to suggest they need to lose weight? And, when the consequences of these lies is a proven reduction in lifespan and overall quality of life for obese patients, aren’t we at risk of institutionalizing a form of fat oppression by way of deliberately limiting an obese person’s health outcome?

Obese individuals can be smart, beautiful, competent people. But, obesity is also the single greatest health epidemic facing America today, with a full 30% of children now being classified as overweight, and with some developing type II diabetes (traditionally an adult onset disease) as young as age 10.

I get that the fat acceptance movement comes from a place of wanting to encouraging healthier relationships between people and their own bodies. But, the movement against weight bias should be about addressing the institutional bigotry against fat people that aims to shame obese individuals into losing weight, rather than supporting them in that process. The movement against weight bias should be highlighting the mental health and self-esteem issues that obese children face when they are ridiculed by classmates and teachers, instead of offered in-school training and access to proper nutrition. The movement against weight bias should be working hand-in-hand with federal health initiatives that are promoting better eating and more activity by promoting the message that obese individuals should lose weight because they love their bodies, not because they hate them.

Instead, the movement takes a step backward, in my mind, when it highlights fat activists like Wann who would rather fall back on science denial to support her argument. It takes a step backwards when its most high-profile issues involve an extra fee for airline seats rather than actual public policy debates that could genuinely improve the quality of life of obese patients. It takes a step backwards when it focuses on self-described anecdata. The movement simply can’t be about promoting the misguided notion that fat people are healthy because doctors are intolerant bigots; this is just not a rational argument supported by the scientific facts.

The medical community really needs to improve how we assess and treat obesity. We need to throw out the BMI scale, and popularize measures of body fat percentage as our primary assessment tool. We need to be encouraging all patients, including those that are normal weight as well as obese, to take the stairs, to eat better, and to spend more time on our feet. We need to be issuing pedometers to every patient who walks through the door’s of a doctor’s office, we need to be putting nutritional information on ever restaurant menu, and we need to bring back physical education and recess in public schools.

And yes, we need to hear about how our society is intolerant and abusive of obese individuals. We need to challenge our current beauty ideals, which portray skinny women (and some men) who are no more healthy than a morbidly obese individual.


At 110 lbs and 6'2", model Jodie Kidd is medically underweight, by BMI.
At 110 lbs and 6’2″, model Jodie Kidd is medically underweight, by BMI.

We need to get Americans comfortable with being active for its own sake — because it’s fun and it’s healthy and it’s better for the environment.

What we don’t need to be doing is give any more airtime to obesity denialists.

Is Diabetes More Deadly for Asian Americans?

Despite their reduced numbers in the American population, non-White minorities have significantly higher incidences of Type II diabetes compared to Whites (image and data courtesy of the CDC).

This past week, the CDC’s National Center for Health Statistics — a division of the CDC responsible for maintaining and publicizing national health trends for various diseases — held its annual National Conference for Health Statistics (NCHS). Members of the Asian Pacific Islander American Health Forum (APIAHF), an Asian American organization whose mission is to advocate public policy changes regarding major health concerns specifically affecting the Asian American population, was at the NCHS and tweeted several interesting facts regarding health trends in the Asian American population.

In particular, two tweets, posted back-to-back, caught my eye:

The incidence of diabetes doubled among #AAs from 1997-2008#NHIS, the highest increase among all racial groups. @kathylimko#NCHSconference


#HeartDisease and Stroke are the leading causes of death among#AsianAmericans after #Cancer@kathylimko #NCHSconference

The increased prevalence of diabetes and heart disease in the Asian American population is not well-known or well-publicized. In part, it may be that these diseases — rightly or wrongly — have been characterized in the popular eye as “fat people diseases”. They are seen as a complication arising from morbid obesity, which itself results from the nutritionally poor diet and low activity levels common in the Western world. Indeed for the most part science backs this up: metabolic disorders and reduced physical activity are well-recognized precursors that greatly increase risks for subsequent development of coronary artery disease, atherosclerosis and Type II diabetes.

By contrast, Asian Americans have reduced (but rising) prevalence of obesity compared to the overall U.S. population (although this is in conjunction with lower levels of physical activity as well). This perpetuates the stereotype that Asian Americans don’t suffer from health factors associated with obesity; consequently, we are virtually ignored when it comes to affecting public policy changes in this country that would address obesity and (perceived) obesity-related health problems.

And yet, the numbers don’t lie: diabetes is more prevalent amongst Asian Americans (and other non-White minorities) compared to in the White American population. Further, the statistics presented at NCHS this year (and reported by APIAHF) indicate that diabetes (and coronary artery disease) are are a growing health concern for the Asian American population. Yet, the impact of these diseases  on the Asian American population is virtually ignored by makers of public health policy, perhaps because of the general perception that Asian Americans simply don’t get diagnosed with obesity-related illnesses.

But, not only are these diseases a growing problem for Asian Americans, a new study published this week suggests that, at least when it comes to diabetes, Asian Americans may suffer a unique, more potent form of this disease that dramatically increases mortality. A study published in the Journal of the American Medical Association (JAMA) (full text, may require institutional access) collected data from over 2000 patients diagnosed with Type II diabetes and compared the mortality of these patients based on their BMI at the time of diagnosis (i.e. “normal weight” vs “overweight” patients). The authors found a surprising trend: “normal weight” patients were more likely to die during the study period than heavier patients:

Survival rate (vertical y-axis) vs. Time (horizontal x-axis) in "normal weight" and "overweight" patients diagnosed with Type II diabetes, from Carnethon et al 2012. As you can see, "normal weight" patients have lower survival rates over time compared to their "Overweight" counterparts.

The reduced survival rate of “normal weight” Type II diabetic patients remained evident even when the authors adjusted their data to take into account smoking status, socioeconomic status, and race (White vs. Non-White; no more stratified race information was available).

Now, clearly, these data shouldn’t be used to argue that we should all attempt to become more obese. Obesity is a clear, established risk factor for diabetes, so doing this will only increase one’s likelihood of developing Type II diabetes in the first place. And, further, I don’t want to overstate the data — there’s a lot of interpretations that could be made of the author’s findings. However, these data do suggest a couple of possibilities:

1) BMI may be a terrible indicator of the health risks of obesity in Type II diabetic patients. The authors noted that a higher waist-to-hip ratio resulted in higher rates of patient death. Waist-to-hip ratio is a measure that compares a patient’s waist circumference to their hip circumference; a higher ratio means that you have higher fat deposits in your abdominal region and is a better measure of both adiposity and fat localization (which matters when it comes to how devastating that fat can be to a patient’s health). This study may indicate that it’s time for health professionals to begin the long overdue breakup with the BMI table in favour of better measures that can more accurately assess a patient’s body fat levels.

2) The data presented in the study may suggest that the BMI table is biased in its exclusion of racial information. Specifically, Asian Americans with a BMI that the BMI table would categorize as “normal weight” may actually be suffering from obesity-related health complications; in short, Asian Americans should be classified as “obese” at a lower BMI than other groups. This is the point of view suggested by some researchers at the Joslin Diabetes Center’s Asian American Diabetes Initiative.

3) The data presented in this study may suggest that Type II diabetes is a convergent disease that can arise from a combination of multiple factors. While obesity may be a driving factor for the development of disease in “overweight” patients, “normal weight” patients that nonetheless develop Type II diabetes may be developing it for different reasons. Additionally, they may be suffering from other health complications that exacerbate the progression of the disease. Either way, the form of Type II diabetes found in “normal weight” patients may actually be slightly different, and perhaps either more inherently deadly and/or less resistant to existing treatment protocols (which were largely developed through work with “overweight” diabetic patients).

And further, it’s possible that Asian American patients may be predisposed for “normal weight” diabetes due to genetic and/or lifestyle differences. CNN reports:

Overall, about 85% of people with diabetes are heavy. Gaining too much weight is a major contributor to Type 2 diabetes, since excess fat cells can affect the way the body breaks down glucose and produces insulin, but some normal weight individuals can develop the disease as well.

The elderly and people of Asian descent are more likely to be at normal weight when diagnosed, for example.

In the end, we really don’t know enough about Type II diabetes and how it affects the Asian American population. All of these data really raise more questions than answers, and ultimately demand additional attention devoted to the study of Type II diabetes in “unconventional” (i.e. non-White and/or non-obese) populations.

To date, the vast majority of clinical studies have been performed on White populations, which significantly biases our understanding of many diseases. Further, most of the statistical health data collected by scientists and the federal government fail to adequately record racial data, despite ample evidence suggesting that race and socioeconomic status both affect patient outcome in several diseases. Thus, what meager statistics we do have regarding the health status of the Asian American community suffers from low sample sizes and inadequate stratification by factors including geography, socioeconomic status, health insurance coverage, and ethnicity.

Basically, we need more and better studies of how race affects diabetes and other diseases. Even the Carnethon study that I cite in this post doesn’t specifically address how different racial backgrounds impacts the progression of Type II diabetes. We need to know how the rising rates of Type II diabetes in the Asian American population can be treated (or not) by our existing knowledge regarding Type II diabetes in other populations.

And, most importantly, we need better federal funding of science, so that those studies can be conducted.