Monday, September 14, 2009

So Is Obesity Dangerous, or What?

Some of my cohort may remember this:


Well, ok, you may have to enter "Confuse A Cat" in the search line for the Monty Python skit. Go ahead, it's fun. That's my last link for this post. You'll have to take the rest on faith. Only so much putting off of actual productive work is allowed.

I have a tendency to stay up too late reading weight loss articles on the internet. Show me a chubby person with an inquiring mind and I will show you someone who's absorbed a lot of conflicting weight loss research. Most of us are more or less familiar with the 2005 CDC study which was confirmed by a more recent Canadian Study (Must. Not. Link. Time consuming!) Using actual recorded heights and weights and thousands of folks, it was found that being overweight is better for longevity than being of normal weight, as defined by BMI in the usual ways. Even being obese was not particularly dangerous. It wasn't until the point of morbid obesity that there were increased risks of death associated, and those were actually not as bad as the risk of being underweight.

These two studies reach opposite conclusions from another large, well-regarded study, the Harvard Nurses' Health Study, which found what I understood to be a somewhat increased risk of death for a BMI over 25 (relative risk 1.28) which climbed to a relative risk of 2.something for BMIs over 30. Gaining more than 30 pounds over the age of 18 (gulp) seemed to be a bit of a cliff so far as disease incidence. There was a study of Swedish men which reached a similar conclusion. Twice the risk of death is not negligible, eh?

So how did four large, carefully designed studies conducted by scientists with lots of letters after their names and a firing squad of educated scrutiny arrive at such vastly different conclusions?

I gather that the Harvard Nurses (well, no, their lead researchers, including the well-regarded Walter Willett of Eat, Drink and Be Healthy fame) accuse the CDC of not excluding smokers and the sick from their numbers. We all get that people can get thin as part of getting sick and in fact getting thin from sickness usually indicates a pretty serious problem. The CDC's answer? They ran the numbers both ways-- both including and excluding those people-- and the differences were not significant.

Meanwhile, the CDC claims that Willett's group used a mostly homogenous population (mostly white and educated). More importantly, they claimed that Willet's group cherry picked their cases, in some cases excluding 90% of deaths by theorizing that those nurses had some underlying ailment which would confound results.

I read Dr. Willet's book and he struck me as a straight up guy. It bothers me to think that he would perform sleight of hand with data in order to demonstrate that being fat is bad for your health. Moreover, I haven't seen any studies that convince me that the association of obesity with Type 2 diabetes, certain types of cancer, gallstones, and heart disease is null. It's a stretch for me to think that, yes, getting fat makes you more prone to Type 2 diabetes, but that won't affect your longevity. The head of the CDC distanced the agency somewhat from their own conclusions-- This doesn't make it okay to be fat, despite what our study says! was the take-home message.

On the other hand. The CDC and Canadian studies were large studies. My doubts were somewhat soothed by their insistance the numbers were run both ways. Moreover, scientific and social pressure is decidedly in favor of proving that obesity is bad, so I would imagine they were pretty sure of their stuff before they stepped forward. I would like to see a good explanation from the Harvard group of which deaths they chose to exclude and why. If possible, it would be nice if data from both studies could be run the same way, although I am sure there are important statistical reasons why this can't be done. And I would really like to read the studies which are driving the weight loss surgery industry (ok, I'll provide a link: this site has some eye-popping things to say on the subject, though I am not completely sold). We often hear that someone has been told they will die in a year or two years if they don't lose weight, causing them to undergo weight loss surgery which is not exactly minor stuff. We worry about our loved ones in the 300 plus weight class. We measure our waists and now our thighs with a hint of anxiety-- are we okay? A scientifically-based answer seems to be as far away as ever.

Even more to the point-- and more difficult to study-- is whether losing weight actually improves mortality? As far as I know, this has never been shown (or not shown, for that matter).

For those of us worried about diabetes, the debate is kind of academic. I haven't seen anything remotely convincing that fat is unrelated to Type 2. There is good encouraging evidence that losing weight and keeping fit can postpone diabetes. Moreover, unless I have been brainwashed by the diet industry, my personal study group of one has found that reducing the amount of sugar and refined foods in my diet coupled with weight training and hiking (don't forget Zumba!) has:

1. Improved my energy and mobility. Watch me run in kickball.
2. Improved my blood pressure.
3. Reduced cravings.
4. Dramatically increased the number of middle-aged guys who check me out in the supermarket.

It remains to be seen whether I'll see improvements in blood sugar. And I don't have a twin who will agree to stay at 195 pounds and fairly sedentary in order to compare what would have happened. I have to do the best I can with the information I have (dang it-- AGAIN.)

I welcome your own thoughts/late night researches on the subject.


  1. Wow, you're really making me think that I'm underthinking this whole deal. Eat more healthy food, eat less food, exercise more... that's the three pages in my game plan.

  2. Hmmmmm! Many points to ponder. I can't help but think that healthy habits are a good thing, so I guess I'll simply continue my attempts to improve mine.

    And, yes, the comment in BB&B was said in a snotty, snarky way. Otherwise I would never have made the comment I did.

  3. @VR, I'm sorry! How maddening! Some days it just seems like wearing earplugs would make life easier.

  4. Thanks for outlining all of this information. I haven't been investigating the health stuff too much and, like Jack, am now rethinking if I should educate myself further.

  5. Wow, there is a lot to this post. You bring up a really interesting point though (actually several of them). The researchers can make the studies say what they want to. You have to analyze the parameters that led them to their conclusions if you really want to understand what is going on. Unfortunately, that is far more than most people are willing to do.

  6. I no longer have faith in most of the published studies. Everyone has an agenda.

    I *do* find valid your study of one, with its convincing evidence of improved life with healthier diet and exercise. I've had the same experience. :)

  7. Thanks for stopping by! You have exactly the same number of baby-weight pounds I did, I have 3 kiddies and love Zumba and cherry protein shakes too :-)

    I have read about the Factor 5 plan and that's one of the few plans that has really interested me yet I haven't tried.

    Going to add you to my Reader right now.

  8. I tend to have more faith in Studies With Large Numbers of Subjects, but even then it's important to read the fine print about what was studied and how it was studied.

    I've grown more cynical the more I've read, but not all studies are bad. Though when I hear the words "questionnaire" and "phone research" I tend to throw my hands up in horror and stop reading right then.

    unmjdmev - hmmn. Clearly, you're using a Word Verification program designed by someone in the former Czechoslovakian region.