Friday, February 11, 2011

WTF Friday: Duke Finds Obese People Are To Blame for Rising Health Care Prices

I was thumbing through the most recent issue of the Duke University Alumni magazine as I sat on the john the other day, about the only time I get to read undisturbed and in all truthfulness, my short sessions are all the time I care to devote to Duke's publication because quite frankly, I get fed up with their WE ARE DUKE...WE ARE GENETICALLY SUPERIOR TO ALL LIFE FORMS attitude, when an article caught my eye and made me clench up a bit. Not a good thing, considering where I was sitting.

It was a study (and God knows how Duke University loves itself a study proving just about anything it wants to because they've got the money and the backing) which concluded that obese people in the workplace tend to be less effective at their jobs and cost a company's health insurance more money because of their illnesses and how expensive it is to treat them.

You know, because obese people always have diabetes, high blood pressure, and heart disease on account of all the doughnuts they sit around scarfing, washed down by all the sugar-laden coffees consumed on company time. Apparently after such a gorging, fat people waddle off to their desks where they sit staring blankly at their computers for 20 minutes before they pick up the phone and inform their bosses they're taking the day off because they're too fat to work and don't feel well.

At least, that's the image that Eric Finkelstein, author and propaganda minister of this study would have you believe.

The article in the magazine was a bit abbreviated, so I dug a little deeper to find the specifics. And then I started poking some holes which let in a whole lotta light.

Finkelstein's yard stick for measuring obesity was based on the widely discredited BMI scale which straight away sent up red flags in my head. When a study begins to lose credibility within the first paragraph, that's when I cease to be upset by it and begin to be amused, the same way you watch someone who's so clearly guilty of a crime flounder and protest their innocence even as the bailiff is clapping the handcuffs around their wrists.

Just for tickles and damp panties, I looked up "BMI", and I found something really interesting. Even though it's gained popularity in the last 20 years or so, take a guess as to when the BMI scale was invented.

If you said mid-1800s, you're either very smart or incredibly lucky.

Because that's right. Body Mass Index was invented between 1830 and 1850 by the Belgian polymath Adolphe Quetelet. It's a heuristic method, meaning it's not exact and gains its data by trial-and-error and experiment. Which means it's nowhere near accurate enough to be used as a certified diagnostic tool. For it to be an accepted scientific principle, it would have to be true every time, in every instance, tested reliably with roughly the same result each time in a variety of conditions. And it's not.

Because you're talking about people's body mass composition, the end product as opposed to the cause. Shall I point out that people are heavy for a variety of reasons, or am I preaching to a saturated and dozing choir? For BMI to be a reliable scientific method of determining what makes up a person's body, it would take a LOT more than simply weighing someone, then consulting a chart for a random number for your ranking. By that reckoning, a body builder, who is a solid wall of muscle, would be labeled obese on a BMI chart purely based on weight and height.

Would you tell a guy who could bench press you and your best friend at the same time that he's in the same health risk category as a guy who eats food from buckets, guzzles soda and beer, and gets no exercise is? Or that because of his BMI score, he doesn't qualify for insurance?

(I'll be waiting for what's left of you right over here if you do...)

Why do doctors still use the BMI? Quite bluntly, it's cheap, requires no medical forethought or evaluation or even one of the most accurate diagnostic tools in a doctor's arsenal--TALKING with a patient and asking intelligent questions. It needs no special training or instruments. And insurance underwriters love it.

From the Wikipedia article:

"Most private health insurance providers will use a particular high BMI as a cut-off point in order to raise insurance rates for or deny insurance to higher-risk patients, thereby reducing the cost of insurance coverage to all other subscribers in a 'normal' BMI range."

However, it's loaded with shortcomings. BMI does not take the following things into account: frame size, bone structure, masculinity, fat, cartilege, water weight, and even more. Plus its standards slide around depending on the country. You could be merely overweight in the US but Obese Level II in Japan. On top of that, the BMI scale slides up and down even more for professional athletes, the elderly, and children with adjustments made all over the place, rather like Papal Indulgences ("Well, you're technically overweight according to the BMI chart, but you're Asian, so that'll earn you a few extra points downward...and you're female and elderly...so let's add that into the calculation...")

So...what you have in a sliding scale that's whimsically dependent on situation, age, race, and insurance underwriters which is in itself a hit-and-miss methodology, the statistical equivalent ofthrowing a dart at a revolving dartboard whilst blindfolded, predicting where it will strike, recording the results, and calling it science. In fact, "Mathematician Keith Devlin and a restaurant industry association The Center for Consumer Freedom argue that the error in the BMI is significant and so pervasive that it is not generally useful in evaluation of health" (from "Do You Believe in Fairies, Unicorns, or the BMI?". Mathematical Association of America. 2009-05-01. http://www.maa.org/devlin/devlin_05_09.html. Retrieved 2009-05-22 and "Is obesity such a big, fat threat?". Cox News Service. 2004-08-30. Archived from the original on 2007-08-04. http://web.archive.org/web/20070804113535/http://www.rockymounttelegram.com/featr/content/shared/health/stories/BMI_INDEX_0830_COX.html. Retrieved 2007-07-08).

THIS is what less enlightened doctors are using these days to determine your span on this earth and whether or not you are worth treating or saving. Or hiring.

Once more from the Wiki article, then I'll leave it alone because I think you take my point:

In an analysis of 40 studies involving 250,000 people, patients with coronary artery disease with "normal" BMIs were at higher risk of death from cardiovascular disease than people whose BMIs put them in the "overweight" range (BMI 25–29.9).[22] (emphasis mine). In the "overweight", or intermediate, range of BMI (25–29.9), the study found that BMI failed to discriminate between bodyfat percentage and lean mass. The study concluded that "the accuracy of BMI in diagnosing obesity is limited, particularly for individuals in the intermediate BMI ranges, in men and in the elderly. ... These results may help to explain the unexpected better survival in overweight/mild obese patients.
(once again, my emphasis).


This, my friends, is the shaky framework on which the esteemed Herr Finkelstein has based his entire well-funded study and its conclusions. I call shenanigans. You begin to see why I've posted this under WTF Fridays. Bad science is still bad science even when it's got a lot of funding and a well-respected name behind it.



(Is this a good time to bring up that Duke University also has a highly esteemed and highly profitable obesity treatment clinic whose numbers will no doubt increase now that this study has been made public? No? Okay, I'll wait till later. Forget I said anything, but I'd be willing to bet that the funding for this little bit of Obesity Epidemic Flame Fanning came straight from the Obesity Clinic itself).

From the Finkelstein article:

With a burgeoning obese population in the U.S., the study has important implications for employers, as they are faced with increasing costs to insure full-time workers.

“Our study provides evidence of yet another cost of obesity,” said Finkelstein. “Employers should consider both the medical and productivity costs of obesity when thinking about investments in weight management or other wellness programs.”


Perhaps Finkelstein should familiarise himself with the CDC's redaction of the "400,000 people a year die from obesity!!" statement with a more revised and scientifically solid "less than 26,000 deaths per year can be blamed on obesity". You know what kills more people per year than being fat? TOBACCO. That's right, ladies and germs, smoking fags kills far more people faster and more effectively than the extra girth around your gut.

In fact, according to the CDC's website, the top ten killers of Americans are...

* Heart disease: 616,067
* Cancer: 562,875
* Stroke (cerebrovascular diseases): 135,952
* Chronic lower respiratory diseases: 127,924
* Accidents (unintentional injuries): 123,706
* Alzheimer's disease: 74,632
* Diabetes: 71,382
* Influenza and Pneumonia: 52,717
* Nephritis, nephrotic syndrome, and nephrosis: 46,448
* Septicemia: 34,828

"But but but!" I can hear Finkelstein's supporters bleating. "Aren't those things caused by obesity? Doesn't it raise the risk factors of those killers?"

*Sorry, I had to pick myself up from the floor, I was laughing so hard*

And I'm glad you piped up with the term RISK FACTOR. Remember, folks, these are statistical possibilities, not science, big Maybes, not fact. The fact remains that you cannot crawl inside someone's bloodstream and DNA at a glance and determine inconclusively whether or not there's a genetic pre-disposition for things like stroke and cancer. If you dad had colon cancer, yep, you PROBABLY have an increased risk for it. If your mom died of a stroke, yeah, MAYBE your odds are increased. But maybe they aren't. Any scientist worth his salt will tell you environmental factors are just as relevant in determining risk factors.

So if you eat crap, don't exercise, and pretty much live the sweet blob life, you most certainly will not be healthy (erm...duuuuuh) and will probably develop health problems. But if you are genetically destined to be heavy, like I am, you actually increase your risk of developing problems if you try to force your body to drop below its threshold weight. And in fact, folks who routinely weight cycle (shed and regain 20-40 pounds regularly) or folks who lose 20% of their body mass are actually at a greater risk of developing all the Horrible Things that are supposed to happen to you if you are obese (and oh yes, there's science, not theory, to back that one up).

And can someone tell me please how the things listed up there BEFORE obesity can be blamed on being a fatty? Chronic lower respiratory disease, unintentional accidents, and Alzheimer's disease, all stemming from obesity? I'm supposed to believe that?

Now take a look at how prospective employers would interpret this information. Two people, equally qualified, apply for a job. One of them is fat, the other is a smoker. Given a choice between these two, whom do you think the employer will choose? (Consider carefully: at a glance, it's not possible to tell if someone's a smoker, but it IS possible to tell someone's obese. Oh, and remember a lot of work places will give smokers two fag breaks a day plus 20 minutes every 6 hours to do with what people wish).

Perhaps that's taking it a step too far, but a study backed by the highly esteemed Duke University packs more of a punch for an employer who's really struggling to make ends meet insurance-wise. If one were adopt that mentality based on Finkelstein's findings without really thinking about it through, denying an obese person a job based solely on what health problems they might develop is the same discriminatory thinking that employers used to apply toward not hiring women because they might get pregnant and demand that blasted maternity leave.

If an obese person has diabetes, it might've been the result of his weight. But it might not be. Look at the risk factors for diabetes on any website. The top three (depending on type) are heredity factors, diet, and age. THEN obesity. And in some cases, if your mom had gestational diabetes when she was pregnant with you, that's an increased risk too (and for the record, GD hits randomly, think and thick alike).

Correlation is not causation, Mr Finkelstein, something a scientist of your learning should know by heart and adopt as his mantra. Publishing a study as flawed as this one and signing Duke's name to it is a discredit, and I'm willing to wager that in a few years, this study will be overturned just like the past studies that swore up and down that low-fat dieting was the way to go to lose weight. Or any of the ones that say dieting works (when 90% of people who diet gain back all their weight and then some) and thin is the same thing as healthy (ask an anorexic if they believe that one).

Shame on you, Mr Finkelstein. Bad researcher. Go to your mat now and stay there and think about what you've done.

2 comments:

  1. "Bad science is still bad science even when it's got a lot of funding and a well-respected name behind it." Yes, it is!

    While I don't think even the "unhealthy" person you've listed above who DOES perform all the stereotypical actions attributed to the Obese Populations is any less worthy of respect and decent health care than someone else; I think the main point you're pulling out is very clear and valid: basing a study of this supposed level of import on faulty metrics (which were never meant as individual markers of health mind you) in a large study funded by a company with a clear interest in the results coming up a certain way, does not lead to a very trusting result by the end for me...

    ReplyDelete
  2. I dunno, April, I tend to have less sympathy for people who do nothing to help themselves get and stay healthy, then expect the doctor to treat them the same way as the folks who take good care of themselves. It's what wrong with so many social systems these days, from big banks to health care. I'm NOT talking about seeing a fat person and thinking, Clearly this is a person who just troughs it...they're all the same. I'm 99% vegetarian, I eat really well, I exercise, don't smoke or even drink, etc...and it's one of the first things I tell any new doc when I meet them before their brains can even think, Oh my, she's a fatty. Haven't had a problem since I've been here in Germany...it was more of a problem in England.

    But it seems like Herr Finkelstein is taking the exception, not the norm, with obese people and trying to make it the rule of thumb. This kind of study is so dangerous because any employer can look at the results, then look at at his work force and see heavy people not as folks who may be doing their job quite well but suddenly as time bombs that will ruin the company's insurance policy. And given that you can't look at someone and gauge their weight, it might even lead to profiling based on size...and that's where I start swinging my Constitutional Rights around.

    ReplyDelete

Whatchyu talkin' bout, Willis?