A recent posting on Jo Nova’s blog: “There goes another consensus. Crash diets solve diabetes in 3 weeks” has caused me to call into question the motivations of some questioners, or bare minimum, their actual understanding of science.
At question: a “new” study that shows 11 Type 2 diabetics who ate 600 calories per day for 8 weeks “cured” type 2 diabetes. In researching this, I found the following websites to be interesting:
The first was from July 2013. The second from June 2011 and third from 2013/2014. All three seem to originate from the same university. All had a very tiny number of subjects. The study from 2011 is very nearly the same as 2013, so much so that I cannot tell if it’s actually separate research or just a new release.
In all cases, promoting this as “consensus busting” is very, very bad science. Both studies were tiny, both done at the same university. It appears to be the same physician, Dr. Mark Hyman, who also sells books and has a newsletter. (I am not saying that this means he is a bad physician. I am just curious how a physician gets time to write books, do research, write newsletters, see patients, do interviews and still sleep.) Of course, even if Dr. Hyman were a complete quack, he might have hit on the truth. It’s just that there is no evidence that he has. From what I can find, there was no randomization, 11 subjects are not nearly enough to generalize from, and it’s not clear if there was a control group. In other words, this is junk science. Being generous, this would be a good place to actually start a study.
The original inspiration for the study seems to have been remission in diabetics after bariatric surgery (which Dr. Hyman does not approve of, it appears). Only 27% actually were “cured” of their type 2 diabetes:
“Publication of a new, albeit retrospective, study has shown that almost a third of obese patients with type 2 diabetes undergoing gastric bypass were effectively “cured” of their diabetes, being in complete remission as per the strictest definition possible that was maintained for 6 years after the surgery. ” (http://www.medscape.com/viewarticle/811284) This study appears to be actually done with proper research techniques and cautions against too much optimism.
Why am I writing on this? Because using a very poorly done study that cannot be generalized and claiming consensus is wrong is a frightening thing to see on a “science” blog. It’s also indicative that the author either desperately wants to prove consensus can be wrong and hopes no one will notice the bad science, or the author really did not understand how bad the science was. Either way, I find the entire posting deeply disturbing. The comments were equally disturbing–how many people hate “Big Pharma” and will jump on anything that might help prove how bad drugs are is quite frightening. Worse, none seem to care if the science is good. While most of the time, I am loathe call things “confirmation bias”, some things just seem to be glaring examples of said phenomena. The same people who decry the lack of science in climate change research jump on any tiny little thread of information to condemn pharmaceutical companies.
While proving consensus can be wrong (and has been) may be a valuable thing, using junk science to do it just makes the questioners as unscientific as the advocates. It’s very sad when both sides veer off from science in an attempt to “win”. No one wins and science loses.