Loose thoughts on various diets
Nov. 10th, 2011 02:55 pmSince I decided, with Jaimie’s help, to try to get on a healthier diet, prompted by my recent decision to try BJJ competition but a good idea anyway for obvious reasons, I’ve spent some time reading about healthy diets, rather more time listening to Jaimie talk about it (as she’s done a lot more reading than I have energy for), and some time pondering it. This post isn’t likely to contain any great revelations, but is meant more as a way to organise my own thoughts (correct or not). It may well be that I am making glaring omissions because I’m not that well-read on the subject, but I can’t very well figure out what I need to read up on until I sort out my own thoughts in advance.
Let me hasten to assure you, by the way, that I am not planning to obsess unhealthily over this until I can play my rippling washboard abs like a xylophone, nor to start morally judge people over their adiposity. Still it happens to be true that the reason for my own personal lovehandles is less genetics than a long lack of interest in healthy food, and whatever the fat positive crowd may say (and I morally agree with them), I just can’t put any stock in claims that overweight is without health consequences.
The world is full of preposterous fad diets (Gerson, Ornish, cabbage soup, morning banana), but there are a number of prima facie reasonable, but contradictory diet ideas out there as well. Diets that have any shred of evidence for effectiveness in healthy weight control seem to focus on reducing either fat intake, or carbohydrate intake, or total energy intake by controlling both. Pick any basic type of diet—low carb, low fat, low calorie, vegan, paleolothic, macrobiotic, you name it—and unless it’s a bonkers variety like cabbage soup, you will no doubt easily dig up a mountain of evidence in its favour, making much reference to the dangers of trans, saturated, or polyunsaterated fats; the benefits of ketosis (sustained or periodic); glycemic indices; and any number of biochemical processes.
Faced with any such evidence load, I must admit myself defeated: I’m not a biochemist, and I don’t understand it. I might be able to read up enough to recite the party line, but I wouldn’t get it. I don’t even know the Krebs cycle, for heaven’s sake! I can at best follow and roughly understand what you’re saying, but I’d need at least another university degree to be in a position to spot any errors. And when one source claims that fat is the root of all evil, and another would stick that label on carbohydrates, or some fats but not others, or what have you, what is a poor non-biochem-postgraduate to do?
Personally, I suspect that lots of these diets work, but I’ve yet to see anything that convinces me that the specific rationale advanced for any one of them is really to credit. A funny thing that emerges when you follow any paleolithic, low-carb, low-fat, or energy-restricted diet on the market is that you tend to end up consuming less energy—in effect, it seems that all of them contain “reduced calories” as one component. One immediately turns to ask whether there’s any more to it. In some cases it seems fairly probable that it isn’t so. For instance, a 2002 Cochrane review compared a low-fat and a general low-calorie diet and found them to be equivalent. I think if you lined them all up, they’d all agree on some basics:
- Eat a reasonable amount of food and a reasonable amount of energy (calories). It might not say this up front, but do the food-math. I think that low-fat, low-carb, and low-calorie diets in name are all low-calorie diets in fact.
- Don’t eat so much heavily processed stuff full of refined sugars and ground-down food products.
- Keep track of what you eat. I’ve seen some pretty convincing research establishing that people who keep track of what they eat tend to eat less and eat more healthily even if they don’t otherwise have much of a plan. Presumably this is a kind of Hawthorne effect.
And of course if you go on a diet, or several diets, consciously trying to lose weight or BFP or feel more energy or what have you, then if at least you succeed, you will naturally tend to attribute your success to what you were doing at the time (confusing correlation with causation; damned hard to avoid with a sample size of one!). Maybe it was because of diet X that you finally got in shape, but maybe diet Y would have worked just as well, even though it failed last time, because this time you’re more determined, your metabolism has changed, your lifestyle has altered, you just have less of an appetite these days, you follow diet plans more strictly… The possibilities are nigh endless, and this is why no one can really know that a particular diet is what works for them. (You can fairly say that “diet X works for me” if you’re on it and get healthier, but you just can’t know that diet Y wouldn’t have worked in its place.)
What of all the rationales, then, all the reasoned and detailed arguments about where polyunsaturated fats go to die and how glycemic indices change depending on whether you eat 26 grams of carbohydrates a day or only 22.5 grams? —Well, what of them? It’s very often true of basic medical research that translating from chemical or cellular scales to gross physiological ones is difficult. The human body is a system of some ten trillion human cells living in community with some hundred trillion bacteria; it’s not a scaled up test-tube! Lots of things can be demonstrated admirably on a cellular level but fail entirely to happen in the human body because it will synthesise, regular, deposit, or excrete things you attempt to add or withhold, use alternate pathways, and what have you.
To take a concrete example, when faced with the evidence linking glycemic index to insulin responses and fat deposits I can only nod and agree that as far as I can tell, it makes sense; but at the same time I know that I’m not actually qualified to tell whether it does. The one question I feel qualified to ask is whether or not this line of evidence has, or has not, put the cart before the horse. Before investigating how carb restriction is a superior method of healthy weight loss to fat restriction, we must establish whether it is. Curiously this seems to be discussed more rarely (though as I said, my reading has been pretty cursory yet). If it turns out that a low-carb diet is superior to an energy-equivalent low-fat diet, then and only then are the mechanisms important to study.
I’m not dismissing basic research here, by the way, or siding with people who misguidedly sneer at research less than Phase III trials. Discovering basic chemical mechanisms in vitro is a perfectly good way to arrive at a hypothesis to test in vivo. Just make sure that once your hypothesis is generated, it’s tested before it’s refined and elaborated. (I don’t mind if you obtain a cart before you obtain a horse, in other words, so long as you hitch them up in the proper order before you try to take me for a ride.) Nor does approaching a problem from that direction mean that the conclusion is wrong. It just means that I need to see the in vivo study before the in vitro study is of the faintest interest to me as a consumer; properly the in vitro stuff should be of real interest chiefly to other researchers—because, to reiterate, the in vivo effects of in vitro findings are difficult and error prone even for experts to predict, and as I am not an expert, it’s pretty hopeless.
My own personal suspicion is that much of the differentiation here is basically product differentiation: If (as I suspect) a large array of diets all work, then the way to launch a successful new product in the diet marketplace is to come up with a variation that sticks within what works but is sufficiently distinct to be marketable on its own, and provide rationalisations for why this particular thing is so superior. I use the language of economics here, but it might not be money per se. Maybe the people selling a diet are indeed selling literal products, but they could also be selling books, or social capital as valued members of an online community. And their product really does work, after all…
With my very tentative working assumption that this whole array of non-insane diets all work and that the differences are far less important than the similarities, my equally tentative conclusion is that the most important feature of a calorie-restricted diet is the psychological factor: How easy is it to stick with? Here, I suspect that the low-carb diets have an edge over low-fat diets simply due to relative satiety. You can eat yourself unhealthy by eating too much sugar and starch, and you can eat yourself unhealthy by eating too much fat; but it’s easier to do it on the former because the latter make you feel fuller.
In other words, my current stance is: Look at all the mainstream healthy diets. Pick the one you think will be easiest to stick to. Stick to it. It’ll probably be good for you if you do. And you might still not lose weight, because (1) it can be very difficult to stick to restricted diets and (2) some people are genetically predisposed to carry more fat (but going on that healthy diet is a good idea anyway, as long as you don’t mistake starvation for health).
Further reading I should do:
- Are there horse-before-cart studies comparing low fat, low carb, Atkins, paleo, &c. diets? I.e. studies that attempt to perform longitudinal comparisons at controlled calorie consumption levels?
- …I was going to make a list, but the above item really looks like the necessary first step.