Friday, 16 March 2012

Red Meat and Rubbish Science

I'm a little late to pass any comment on this, in fact I was going to avoid it because Zoe Harcombe and Gary Taubes have done such an excellent job, but I can't overcome the urge any longer....

So you have likely heard all the fuss in the mainstsream media regarding the greater likelihood of death from consuming red-meat (or meat in general). All of this stems from a recent journal article which I am more than happy to link to and I  genuinely hope you will go away and read because ultimately it proves some of the points I will be making.

The Media
Firstly people in the media aren't scientists, if they were they'd be in lab coats doing experiments somewhere, not writing or sitting behind a desk/in front of a news camera. I read recently the perfect quotation for media interpretation of science. I think you'll get the point.

SCIENTIST: "When taken out of context my results mean nothing!"
MEDIA: "Scientist reports 'results mean nothing' !"

Scientific Process
So within science there are different kinds of study; there are observational studies where we watch things for however long we deem appropriate and then we use this to come up with a hypothesis and to what they were doing that caused these things. And their are clinical trials or intervention studies, where we test said hypothesis. The first is an observation. It's an observation. It doesn't necessarily mean anything, and the point is that no-one can say that it really doesn't mean anything in the same way that no-one can say that it does. It is used to develop a hypothesis plain and simple.

I've blogged about this before and have shown a Tom Naughton video which discusses all of this in detail. An excellent example he gives was the observation that nurses supplementing with Oestrogen were less likely to suffer a cardiac condition (heart disease, etc.). Everyone jumped up and down and said A (supplement) prevents B (cardiac condition). Cause and effect. Right? Wrong? When the clinical studies were done they actually found that Oestrogen supplements were actually more likely to cause a cardiac condition!! So how did they even think that A caused B. Well imagine the kinds of people who take supplement, they look after their body, possibly exercise, possibly eat 'right', and so forth.

Proof or lack thereof...
So coming back to the present discussion.....well the Harvard School of Public Health guys who've come out with this decided not to do the second part. They did the observation bit, and then they skipped straight in to conclusion. What they're presenting to you is a hypothesis. They're suggesting that red-meat is bad for you. But this is NOT scientific evidence. I could claim that when I sing it rains, thus my singing must make it rain, but you wouldn't just believe that observation without some proof, would you?

Track Record
There's a final point I want to raise here. I want to look at Harvard School of Public Health's track record with hypothesis from observational studies, that are then proven correct or disproven after all. Now then, we can let the stereotypes sink back in here, this is Harvard after all....what percentage do you think? Go on have a guess....

Well...cited from Taubes' blog earlier (who has done the research - whereas I am just re-citing it)...Up to 2007, their average was....(drum roll please). 0.000. Yes, they got them ALL wrong. Not wrong most of the time (like me), or wrong just a bit of the time (like some people I know), or never wrong (like most women I know). They were wrong EVERY. SINGLE. TIME.

Now I could run some fancy statistics on all of this, but really; who needs it. Forget who likely meat are to kill you - how likely are they to be wrong over this!?!?

Be Well

JF

No comments:

Post a Comment