This blog from the What The F&$^ file is thanks to an e-mail from a reader named Sabrina. Thanks Sabrina!
I love getting reader e-mail so feel free to send me interesting stuff at dances with fat at yahoo dot com.
Sabrina sent me a link to the article: “Twinkie Diet Helps Nutrition Professor Lose 27 Pounds”
The story highlights are as follows (let’s pay particular attention to the third one, shall we):
- Nutrition professor’s “convenience store diet” helped him shed 27 pounds
- Haub limited himself to 1,800 calories and two-thirds come [sic] from junk food
- Haub said it’s too early to draw any conclusions about diet
To sum up: For reasons somewhat passing understanding this guy decided to see what would happen if he ate like crap for 10 weeks, but only 1800 calories of crap per day.
According to the article, he ate a snack cake every three hours in lieu of meals, but at the table with his kids (so as not to set a poor example) he ate vegetables. To quote the article:
“Two-thirds of his total intake came from junk food. He also took a multivitamin pill and drank a protein shake daily. And he ate vegetables, typically a can of green beans or three to four celery stalks.”
Wait, so do his kids think that a can of green beans or 3-4 celery stalks is an appropriate meal?
During the 10 weeks he lost 27 pounds and his health markers improved: his “bad” cholesterol went down, his “good” cholesterol went up, and his triglycerides went down.
What does he say about the experiment?
“I wish I could say the outcomes are unhealthy. I wish I could say it’s healthy. I’m not confident enough in doing that. That frustrates a lot of people. One side says it’s irresponsible. It is unhealthy, but the data doesn’t say that.”
He’s exactly right. We can’t draw conclusions for a bunch of reasons. Here are some of those:
- He didn’t properly track his eating before he started the diet to create any kind of baseline
- It looks like he went from eating a few large meals a day to many small ones which can have an effect on the metabolism, at least in the short term
- His study had no control group
- 1 dude for 10 weeks does not statistical significance make
- The body corrects for weight over time – in long-term weight loss we like to see 5 year success so I’d like to hear from him in 4 years and 42 weeks
- His results have not been replicated (and I wouldn’t hold your breath for someone to try)
He says over and over not to draw conclusions based on this “experiment”. I’m not mad at him, he understands the limitations of his study.
But then the article quotes Dawn Jackson Blatner.
She is a dietitian based in Atlanta, Georgia and spokeperson for the American Dietetic Association (who I personally think should be stripped of both titles for being epically bad at science) who says:
“Being overweight is the central problem that leads to complications like high blood pressure, diabetes and high cholesterol, she said. When you lose weight, regardless of how you’re doing it — even if it’s with packaged foods, generally you will see these markers improve when weight loss has improved,” she said.
Really Dawn? Science is hard, let’s go shopping.
If you read this blog, oh I don’t know…EVER, you’ll know that I disagree with this based on a TON of research that I’ve done. I’ve not seen any good research to back up anything that she said here, and I’ve seen a lot of research (of course, she doesn’t cite any). She doesn’t know the difference between correlation and causation which activates my eye-roll reflex – studies show that these problems and being “overweight” happen at the same time, they haven’t proven that one leads to the other. Caloric deficit weight loss methods (eating less calories than you burn) have an abysmally low success rate, and over 95% of people end up LESS healthy than they started within 5 years. Mostly though, I just want to ask her – I’m trying to choose between a diet based on cocaine or crystal meth. Which would she recommend? I mean, whatever makes you thin, right? Right…?
In a not-that-redeeming slightly lucid moment she did say: “There are things we can’t measure. How much does that affect the risk for cancer? We can’t measure how diet changes affect our health.” I find a dietitian saying “We can’t measure how diet changes affect our health” pretty suspect, but I refuse to waste any more time on this woman. What I don’t understand is why, in an article about a nutrition professor, they thought it was necessary to bring this woman in for an “expert opinion”? My guess is that they wanted a conclusion and, as a scientist with ethics, he couldn’t give them what they wanted. Enter Dawn Jackson Blatner.
Bottom line: I could start eating 2/3 junk food in the hopes of getting to a BMI of which Ms. Blanton would likely approve, but if I didn’t lose weight long term (and I suspect I wouldn’t) and my health failed (and I suspect it would), I’ll bet Ms. Blanton wouldn’t be there backing up my method, confused as to why it didn’t work, and singing the praises of the Twinkie diet.
It’s not called “Healthy Skepticism” for nothing y’all.