And now for something a little different —

Okay, I haven’t in general said much about the nonfiction books I read, and I certainly haven’t done a long post about any of ’em, but this one deserves it. So —

The book: WHY WE GET FAT by Gary Taubes

Taube’s argument: Obesity is not a moral penalty for the twin sins of gluttony and sloth. Instead, obesity is caused by a problem with insulin regulation driven by an overabundance of carbohydrates in the diet. People will lose weight and improve their health regardless of their caloric intake and/or exercise regime if they remove carbohydrates from their diet and allow their insulin metabolism to correct itself.

I read this book because I kept seeing people mention Taubes and I finally decided to see what he was saying. My bias before I read the book was solidly against the low-carb idea, partly because I thought it sounded like one of the innumerable stupid modern diet fads that’s supposed to work by magic and partly because I personally love carbohydrates in all their forms and would rather remove meat from my diet than bread, pasta, rice, and desserts. I mean, I have nine kinds of rice in my pantry! (In case you’re interested, these are: ordinary Carolina long grain rice, brown rice, dhura dun basmati rice, jasmine rice, Thai red cargo rice, Arborio rice, Calrose rice (a short grain rice I use for Chinese cooking and sushi), glutinous rice, and wild rice.) I also have many kinds of lentils for Indian cooking, and many kinds of Chinese, Japanese, and Thai noodles. I am definitely not someone a low-carb diet would appeal to.

But I have to admit, Taubes provided a lot of evidence that a carb-heavy diet causes obesity. You may well want to read the book and get all the details, but allow me to present this evidence, briefly. (Well, not all that briefly! This is a very long post! If you want to skip it, but are interested in my personal low-carb experiment, skip to the end. But I hope you will stay with me and let the experiment be the punchline, okay?)

The fact: When insulin levels rise, we accumulate fat in our fat tissue; when insulin levels fall, we release fat from our fat tissue. Insulin is secreted in response to carbohydrates and particularly in response to sweet, easily digested carbohydrates. This is evidently not at all medically controversial. Taubes goes into a lot of detail about insulin metabolism.

The evidence:

* The modern obesity problem can’t possibly be due to fast food or the sudden popularity of high-fructose corn syrup or the modern sedentary lifestyle, because the problem is not actually all that modern. It was already very noticeable in the 1930s. Not only was that way before fast food etc, it was also the era of soup kitchens and bread lines – not the conditions under which one would reasonably expect an explosion in the rate of obesity.

* Physicians in the 1930s were already trying to help obese children slim down, but found the children just couldn’t.

* The Pima Indians have about the highest obesity and diabetes rates in America. They didn’t always, but their problems started way, way back – before 1900. In fact, their problems started when they went from being one of the wealthiest Indian groups to one of the poorest. In the 1850s, the Pima were hunters and farmers, enjoyed an abundance of food, and were not fat. By the 1900s, they had gone through years of starvation and were living on government rations that were calorically and nutritionally inadequate but that included a lot of white flour and sugar. Malnutrition and obesity coexisted in this and other very poor Indian tribes living on government rations. A lot more women than men were fat despite the fact that women did nearly all the hard labor, while men were much more sedentary. Nearby tribes, such as the Pueblos, who did not live on government rations, were not fat.

* Other very poor populations have shown simultaneous malnutrition and obesity, including the Italians of Naples in the 1950s, who ate a diet of pasta, bread, salads, olive oil, and wine. Taube cites 14 other populations which simultaneously experienced poverty and very high rates of obesity. In all cases, the rates of obesity were much higher in women than men and especially high in women past middle age.

* Taubes cites a British specialist on Jamaican diabetes, who in the 1970s declared inadequate nutrition accounted for almost 25% of all pediatric admissions, that inadequate nutrition continued through adolescence, that obesity began to manifest itself in the female population as the women reached their mid-twenties, and that two-thirds of the women then became very obese (and 10% of the men). From birth through their early teens, Jamaican children were very thin, lacking enough food and also nutritious food (Taubes does not report on the actual diet in this population). They then became obese as they reached maturity, especially the women.

* In the 1990s, the National Institute of Health enrolled 50,000 women, the great majority of them overweight, in a study. They randomly selected 20,000 of these women to eat a low-fat diet emphasizing vegetables and high-fiber foods. These women also consumed 360 calories less per day than they had prior to enrolling in the study, a total of 1/5 fewer calories than what public-health agencies advise today. After eight years, these women had lost an average of only two pounds, while their average waist circumference increased, implying that they had lost lean muscle mass, not fat.

* Jeffrey Flier, Dean of Harvard’s medical school and a specialist on obesity, wrote the chapter on obesity for the 2005 edition of Joslin’s Diabetes Mellitus. In this chapter, he states that “reduction of caloric intake” is “the cornerstone of any therapy for obesity.” But after examining all kinds of reduced calorie diets, including very extreme starvation diets, he also concluded that “none of these approaches has any proven merit.”

* Taubes cites one study (among many others) that enrolled 13,000 runners and compared their weekly mileage with their weight from year to year. Those that ran the most tended to weigh the least, but all the runners gained weight over the years, even those who averaged 40 miles running per week. Taubes spends a whole chapter knocking to pieces the idea that an active vs a sedentary lifestyle matters significantly when considering weight gain. Rather than going into exhaustive detail, let me just say that he’s pretty convincing.

* Okay, just one more detail, because wow: one study in Denmark got 18 men and 9 women who had been leading sedentary lives to train for marathons. (Seriously, wow. They would never have got me to enroll in this study, let me tell you!) After eighteen months of training, the men had lost an average of 5 lbs of body fat. The women hadn’t lost any weight.

* You can sum up this part of the book by noting that people have been trying for decades to prove that exercise can reduce weight or maintain weight loss, but have not been able to produce compelling evidence that this is so. After this many studies, it seems pretty safe to conclude that the hypothesis that exercise is important in weight loss sounds reasonable but is not true.

* We can see very plainly that genetics play a large role in fat accumulation, but let’s think for a moment about just how big a role that is. Take beef cattle vs dairy cattle. Beef cattle are fat. Dairy cows are bony. Their hip bones stick out. Their ribs show. These are animals in excellent condition. They certainly are not eating different low-fat grass or running marathons through their pastures. They are genetically selected to put the calories they consume toward milk production rather than fat and muscle accumulation.

* Ground squirrels double their weight in the fall, all of the extra weight being fat. They do this even in captivity and even if they are placed on a restricted-calorie diet. Unless they are actually starved nearly to death, they gain a ton of weight in the fall regardless of diet.

* Also, as anybody can see, boys and girls put on weight differently. It obviously isn’t girls eating more than boys that causes girls to put on a lot more body fat during puberty.

* Also, there’s a rare genetic disorder called progressive lipodystrophy, in which the affected person, almost always a woman, accumulates a ton of fat and becomes obese – but only below the waist. Above the waist, she loses essentially all her subcutaneous fat. (Yes, this looks strange.) It is obviously stupid to suggest that these women undereat above the waist but overeat below the waist.

* Rats that are spayed – ovaries removed – eat a lot, quickly gain a lot of weight and become seriously obese. They gain just as much weight if placed on a restricted-calorie diet – to compensate for the lack of calories, they become extremely sedentary. In other words, the rats become sedentary because they are becoming fat and thus have less energy available for activity; they are not becoming fat because they are sedentary. Both gluttony and sloth are caused by the fat tissue sequestering calories rather than allowing those calories to be used for activity.

* Spaying the rats removes estrogen from their systems, estrogen regulates lipoprotein lipase (LPL), in the absence of estrogen, LPL causes a lot more fat to enter fat cells for storage than muscle cells for use. Thus the fat tissue sequesters the available calories, so the rats have to eat a lot more if they want to stay active. If they can’t eat more, they must become sedentary.

* Isn’t that interesting? Getting fat is driving increased appetite and sedentary behavior; the customary view that gluttony and sloth drive fat accumulation gets the causal relationship backwards.

* Zucker rats, bred to be obese, become obese even if they are put on severe calorie-restricted diets from the time they are weaned. On restricted diets, they actually wind up fatter than unrestricted siblings. But the rats on the calorie-restricted diets also have muscles, organs, and brains that are significantly reduced in size compared to their unrestricted siblings. The nutrients they needed for normal organ and muscle development got sequestered in the fat tissue.

* If those rats are actually starved to death, they die with a lot of their fat tissue intact. They compromise their crucial muscles and organs and die rather than burn fat tissue.

* Re-read that last point.

* This implies that thin people have fat cells programmed not to take up calories, or muscle cells programed to burn calories. If they are more active than fat people, it is because they are driven to use the calories they take in for physical activity. They are not thin because they are active, they are active because their fat tissue is not hogging the calories needed by their muscles.

* I can certainly confirm that my mother has great trouble gaining weight. She is right at the low end of the normal weight range for women, and in fact she is probably significantly underweight for her age. I can’t get her to eat enough things like cheesecake, but Taubes’ book implies it wouldn’t help anyway.

* Also, I can certainly confirm that some, but by no means all, Cavalier puppies are VERY thin as teenagers and feeding them more just doesn’t help them gain weight. (They refuse to eat as much as I offer). They will gain weight when they are about 18 months old and there’s just no point fussing about it before that. I don’t plan to show Folly till next year! Right now she is all legs and ribs and hip bones! If she is like her mother, she will suddenly fill out when she is about 16 months old. I certainly believe teenage puppy weights are governed by genetics!

The counter-evidence Taubes should have addressed, but didn’t, adequately:

“Uncooked, rice is called mai; cooked it is fan. Once cooked rice was traditionally taken as food at least three times each day, first for jo chan, or early meal, either as congee or, if the weather was cool, cooked and served with a spoonful of liquid lard, soy sauce, and an egg. To eat rice is to sik fan, and there is, in addition to those morning preparations, n’fan, or “afternoon rice,” and mon fan, or “evening rice.” There is even a custom called siu yeh, which translates literally as “cooked midnight” and means rice eaten as a late evening snack. No time of any day in China is without its rice.” Eileen Yin-Fei Lo, The Chinese Kitchen (I just happened across this quote at the same time I was writing this post; Taubes didn’t quote it.)

Taubes doesn’t completely ignore China. On p. 138, he very briefly mentions that the Chinese and Japanese didn’t get fat even though they ate so much rice, an easily digested carb. He ascribes the non-fatness of the East Asian peoples to the lack of sugar in their traditional diets. This is perfectly plausible, especially as Taubes spends quite a lot of time describing the ins and outs of insulin and why sugar might reasonably be a more potent fattening agent than other carbohydrates. On the other hand, Taubes constantly blames white flour for disarranging insulin metabolism, too, and it’s not clear to me why white flour would be worse than white rice.

Also, in India, where the diet of a lot of people is VERY carb heavy, there is a solid tradition of extremely sweet milk-based desserts. I have no idea how frequently those desserts were consumed by most people. And the reason I don’t have any idea is that Taubes didn’t address the populations of India at all.

The evidence Taubes should have included but didn’t:

A rundown of the typical diets of the fifteen poverty-stricken-and-obese populations he mentions specifically, compared to historical and modern American diets, the historical diets of some Asian populations, and the historical diets of some East Indian populations – in each case with notes on rates of obesity.

Tracking all that information down would of course be tedious, but honestly, that’s kind of what you let yourself in for when you write this kind of book, isn’t it? I get why Taubes barely mentioned China and all but ignored India: he wanted to deliver a clean, straightforward, consistent message. But it does weaken his argument to leave out the kind of data that would let us really estimate the kind of dietary and weight variance that has existed between populations that have historically emphasized carbohydrates in their diets.

Taubes does, however, trace out the history of the idea that carbohydrates are uniquely fattening – hardly a new idea, and well-tested experimentally in the early part of the 1900s, but thoroughly driven out of fashion in the sixties and seventies by historical contingency and the sheer refusal of medical and public health organizations to believe the evidence. The take-home message of this entire section is that obesity somehow always seems to be incurable when the diets prescribed to treat it are the exact diets that cause it.

And Taubes also has a pretty extensive section refuting the idea that high-fat diets are bad for heart health or for your health in general.

* Experimental evidence does not support the idea that saturated fat is bad for health. (Taubes acknowledges that this will seem hard to believe given how vehemently health authorities insist on this point, but declares that “what we’ve been told and what the evidence actually supports parted ways in 1984, when the National Heart, Lung and Blood Institute launched its massive health campaign.”).

* The NHLBI spent $115 million on a huge clinical trial and found that a reduction in saturated fat did not have any effect on heart attack rates. Despite this, they continued to claim that reducing saturated fat in the diet would reduce heart attacks.

* An international organization was formed in the 1990s to do an unbiased literature survey on the issue. This was the Cochrane Collaboration. Its judgment is highly regarded when it comes to deciding whether a diet, a surgical procedure, or a diagnostic technique actually does what it is supposed to do. The CC assessed the benefits of reducing dietary saturated fat in 2001 and concluded that the evidence that such a reduction is beneficial is at best limited and inconclusive.

* Since then, a huge trial focused on women, the Women’s Health Initiative, tested the benefits and risks of eating less fat and less saturated fat. The study was the one mentioned above, that enrolled nearly 50,000 middle-aged women. Not only did the 20,000 women assigned to the low-fat diets (diets emphasizing less meat, more vegetables, more fruit, and more whole grains) fail to lose weight, they also showed absolutely no reduction in heart disease, stroke, or breast or colon cancer compared to the women who ate whatever they wanted.

* Triglycerides levels in the blood definitely do constitute a risk factor for heart attacks. This is not controversial. Triglycerides rise when we eat a diet high in carbohydrates. If you replace bacon and eggs for breakfast with yogurt and bananas, both your HDL and LDL cholesterol will go down, but your triglycerides will go up. And though lowering LDL cholesterol might possibly reduce your risk of heart attacks, lowering your HDL cholesterol definitely and substantially increases your risk.

* No, really, low HDL cholesterol levels is definitely a huge risk factor, especially for women, and a high carb / low fat diet definitely lowers HDL cholesterol levels. Absolutely nobody argues with these two facts.

* Clinical trials of the Atkins diet, trials that specifically examined the effects of eating a diet high in saturated fat diet on weight and heart disease and diabetes, have consistently shown that people who eat a high fat / low carb diet lost more weight than people instructed to limit calories and avoid fat and saturated fat. The high fat / low carb diet also causes levels of HDL cholesterol to rise, triglycerides to fall dramatically, blood pressure to fall, LDL cholesterol levels to rise slightly, and the risk of heart attacks to decrease significantly.

* Yes, really. There’s one particular Stanford study Taubes cites in detail. There’s just no question but that the people on the Atkins diet lost more weight and had much better health indicators in every important category than the people on a low fat / exercise regime. Yes, really.

I must admit that this whole section has made me really doubt that we want government agencies deciding what’s best for us and going on huge national campaigns to persuade us to change our diets to suit their misconceived ideas about nutrition. Because when they’re wrong, whoa, the harm that they cause just boggles the mind. And you evidently had better not trust health organizations to actually pay attention to any kind of experimental evidence, even from huge well-designed studies they organize themselves, when the results conflict with their pet theories. Because once they get an idea in their heads, they just cling to it like grim death. I mean, Taubes sure convinced me to ignore all official health advice for the rest of my life.

The take home message: Just how completely you have to cut carbs in order to lose weight or maintain weight loss, and how long it will be before you see the effects of a low carb diet, depends on your genetics and how screwed up your personal insulin metabolism is. Women, older people, and people who are already seriously obese are more likely to fail to lose weight even on a carb-restricted diet, either because they give up too soon, because they don’t restrict carbs enough, or because they have already reached the point of no return.

The experimental results:

Okay, after all that, I was curious to see what a low-carb diet would do in action. I only had one easily available test subject, of course.

As it happens, at the beginning of September, I was just about exactly at the top of the official normal weight for a woman of my height and age. I had also been trying to lose five or so vanity pounds, in the standard counting-calories way. (I would have been happy to lose more than five pounds, but had just given up on that, on the grounds that life’s too short to renounce food.) I had been more or less seriously trying to lose those five pounds for about two months, with zero success. In the past, I have lost up to 15 pounds at a time on a semi-starvation diet. I do not enjoy this kind of diet any more than the next person, and this time around it didn’t seem to be working anyway.

Did I actually go zero carb?

No. I meant to, but I never quite out all carbs. I didn’t drink anything but water, but in fact I never do drink anything but water, so that wasn’t a difference. For the first two weeks, I had a lot of omelets with bacon and cheddar, cheese and sausage for lunches, and baked dishes made with eggs, cottage cheese, spinach, and ham. I also made deviled eggs with bacon and avocado (very good; I made them twice). I cooked pork in foil with squash and onions, then did the same thing with kielbasa. I made chipotle burgers and ate them without the bun. I made “Rubin dip” with corned beef and sauerkraut and cheese and mayonnaise – I spread the dip on slices of raw yellow squash. I sprinkled cauliflower with oil, cumin and salt and roasted it at high heat. Salads with steak, I did that several times. It was sort of fun going for all this high-calorie meat-heavy stuff I normally avoid. I didn’t really eat that many vegetables as vegetables, but I had eggplant several times in various dishes because I have a lot of eggplant in the freezer – eggplant did well for me this year and it freezes surprisingly well after it’s cooked.

But I also made kasha dishes several times (I have a lot of kasha in the freezer). Right at first I ate the rest of the frozen grapes I had in the freezer (don’t knock frozen grapes till you try them!). I added corn to the steak salads. I ate carrots with my cheese and sausage. I had chocolate a couple of times, because, hey, chocolate. I had a doughnut at work when someone brought in a box, because I only have an average amount of willpower and I didn’t figure one doughnut would make much difference.

In the first two weeks, I lost eight pounds. This was on a much higher calorie diet than I usually eat. I was never hungry, believe me. In fact, it was amazing, how I never got hungry. I usually do get hungry before lunchtime when I have pancakes or biscuits for breakfast. The lack of hunger was a huge difference between normal life and this diet. I was sort of hoping to also find myself with plenty of extra physical energy from all those extra calories, but I have to say, unfortunately, no. Well, you can’t have everything.

In the second two weeks, I had bread several times, half a cup of flour at a time as I made single-serving recipes of various things. I had cereal for breakfast twice, with sugar. I ate 1000 calories at one meal the day I was showing my girls and skipped lunch. That was a fast-food burger and fries and I enjoyed it thoroughly, thank you. I had rice several times, in a rice/coleslaw/chicken salad with Thai peanut sauce (it wasn’t as good as it should have been). I made chili with black beans (a carb), sweet potatoes (ditto), and chorizo. And served that over rice, too, because it turned out too spicy to eat plain. I did not watch portion sizes at all. In fact, I specifically had more cereal at a time than I usually would. (I honestly hadn’t realized how constantly I count calories in my normal life, until this experiment when I specifically ignored portion-size advice.) I didn’t have desserts, though, except for chocolate several times, twice quite a lot of chocolate. (This was Sarah Rees Brennen’s fault, because I do love chocolate when I’m reading a good book.) In this second two-week period, I lost two pounds.

I have no idea how much more weight I’d lose if I kept up the low-carb/ low sugar thing. Maybe not much, since I’m now down right in the middle of the normal weight range for my age and height? Anyway, my experiment is over, and I assure you I have every intention of adding desserts back to my diet. For breakfast today I had pumpkin chocolate-chip bread, for example. And I don’t plan to worry about it. Because at this point, I plan to simply plug a two-week zero-carb diet period into my life every three months or so.

Also, in a different but related experiment, I switched one of my girls to a grain-free food. Unlike my other dogs, she’s had trouble losing that extra little bit of weight on a restricted-calorie diet, and she sure hasn’t appreciated my feeding her less, and what with Taubes and his obese rats that compromised their organs rather than lose weight, I thought, well, let’s just try the nearest thing to a low-carb diet there is for dogs. (The food has potatoes and sweet potatoes in it.) Kenya is now eating more calories than before on this high-protein, high-fat food. She hasn’t lost weight, but she hasn’t gained significantly either. I think her activity level is up a bit – she was very sedentary before. I mean, I know she is more active now, but this could be the cooler weather, so that’s hard to know for sure. But I do know for sure that she is happy to have more food in her bowl!

If you want to experiment:

Taubes does not provide detailed dietary suggestions in this book, but he suggests that a seriously overweight person might try a diet including all the meat, poultry, fish, and eggs you want, at least 2 cups of leafy greens per day, and any of the non-sweet vegetables, including the brassicas – broccoli and cauliflower and that stuff – squash, eggplant, green beans, celery, and mushrooms. Also peas, tomatoes, and rhubarb, which surprised me. Of course I can’t imagine eating rhubarb without adding sugar, so not sure why he bothered putting that on there.

He suggested clear broth for sodium replenishment, which strikes me as highly peculiar, since you are not likely to suffer a sodium shortage on a high-meat diet. For heaven’s sake, salt your steak and quit worrying about sodium.

Taubbes suggests four oz of cheese per day, and up to four tbsp. of cream and mayonnaise per day. And up to half an avocado. I made absolutely no effort to limit the cheese I consumed and certainly had a lot more than four oz a day. I have no idea why he would want you to limit cheese. It’s not like it has lactose in it.

What you don’t eat: you don’t get any sugar, flour products, creamed soups, milk or flavored yogurt, rice or other grains, starchy beans, potatoes or sweet potatoes, carrots or parsnips, or fruit. You can use Splenda and Truvia and stuff like that if you really want to. Me, I hold out for real desserts made with real sugar, but your mileage may vary.

It’s very important you not try to maximize protein and minimize fat, because a high-protein low-fat diet is uniquely toxic. (It’s sometimes called fat starvation, I happen to know, and the Louis and Clark expedition nearly died because of it, so don’t do that!)

Then, after you give your metabolism a chance to readjust your insulin metabolism and see what happens on this diet, you add back in various items and see how your body responds. If it were me, I’d add in whole grains first and then try whole wheat flour and then other carbs.

I am not, of course, actually suggesting anybody try this or any other diet. But if you happen to, I’d be interested in knowing what results you get!