Sunday, July 6, 2014

How do you "prove" something?

God all this "calorie myth" stuff that blew up during the week is doing my head in. Nutrition must be the only field where literally anyone can decide they're an expert based on reading a couple of articles on the internet and maybe one of the books they refer to, also all written by self appointed experts with no qualification on the subject.

Actually climate science would be the other one, apparently. All the qualified people who have made meteorology their life's work have got it wrong, and the deniers are the ones who really know what's going on. It's all a conspiracy, innit?

SO... god help me. Everyone has their own idea or their own opinion. I suppose that's fair enough. For nutrition, everyone is entitled to make the decision to follow whatever strategy or ideal they feel is best for them. However, that doesn't give them the right to insist that their way is the "only" way that's healthy, and present their opinions as "proven facts" to convince other people. It would be fine to say "well I think this is quite sensible and it has worked out nicely for me, so I do recommend that you consider trying it".

That's never what anyone says though, is it?

So to PROVE something, what's required seems to vary on the setting.

In a court of law, if you're the prosecutor / Queen's counsel or... you know... depending on what country you're in I think they're called different titles. If you're the law talkin' guy trying to put an alleged crook away, you can't just stand out there and say "Isn't it obvious? He's responsible. Everyone knows that, the Counsel for Defence probably still thinks the Earth is flat too! lololol"

Nope that aint going to cut it. You have to present all the evidence, and then there's a highly educated expert on the law who's job is to poke holes in all of your evidence. You have to prove "beyond reasonable doubt" in order to get a conviction. The standard of proof is quite high.

In real science... I'm happy to have a real scientist correct me on any of this if I'm mistaken but my understanding is, to prove something is correct, what you actually have to do is set out to prove that it is incorrect.

For example... we have a theory that if we do AB&C, the result will be XYZ. We test that theory and as expected, the result is XYZ. Does this prove that AB&C (and only AB&C) causes XYZ? Not necessarily. To prove that this is the case, we'd have to try to re-create the same result in other ways, without success. So if you get to the point where you can say "we tried literally everything, and AB&C was the only thing that produced the result XYZ...." then what? Is it proven?

Not yet. My observation of reading a few studies is that the conclusion might be something like "the evidence suggests that XYZ is indeed caused by AB&C", which is still a safe distance from a statement more like "this is now proven and anyone who thinks otherwise is a deluded idiot". Right?

But we're still not done. Scientific method then requires a "peer review" process, where... I believe it is three other, independent scientists will review the work and see if there is anything wrong with the way the study was conducted, that would cast doubt over the results. Again, the standard of proof is really very high indeed.

Now let's get back to talking about nutrition. There is so much confusion in ideas about diet and nutrition especially related to weight loss, and it is because people are so eager to inflict their opinions on others as "indisputable facts" that only an idiot wouldn't know already. Their standard of proof is much lower.

For example "I cut out bread and lost weight. Therefore bread makes you fat, and you can't lose weight without cutting out bread". For this to be true, you'd have to ask "well, has anyone ever lost weight while still eating bread?" and of course the answer is yes. Finding more people who've cut out bread and lost weight doesn't strengthen the position. Substitute bread with whatever other supposed single culprit people like to blame, be it grains, GMOs, sugars or whatever.

The bottom line here is total calories. There are people who have lost weight by cutting out a particular food or food type, but it is not proof that this particular food choice is to blame for all cases of overweight or that cutting it out is required in order for weight loss to be successful. If removing a particular choice of foods from your daily eating habits means that you are now no longer consuming an excessive amount of calories, you'll lose weight.

Appropriate vs inappropriate total calories will always be the primary determining factor in body weight. How you arrive at an appropriate intake is up to you, although obviously I'd suggest learning your requirements and planning to meet them with your choice of foods is a better strategy than trial and error through restriction of food choices.

Whichever way you do it is up to you, and is fine. But it's not fine to present this approach as a proven "only" way of doing things, which disproves all other approaches and ideas, and it is far from OK to claim that this is "backed by science".

Another disturbing trend I (and others) have noticed recently is the claim that something is "proven by science", supported by links to research which has actually reached the opposite conclusion. You have to worry then... are these people just mistaken and lacking in some comprehension skill, or are they deliberately setting out to confuse and deceive the public?