Wednesday, July 30, 2008

Exercise and Weight Loss

A friend sent me a link to this just published article about weight loss and exercise. According to the abstract of the actual scientific paper (I couldn’t access the full text), overweight and obese women (201 subjects, aged 21–45 years with BMIs from 27–40) were assigned to 1 of 4 behavioral weight loss intervention groups. They were randomly assigned to groups based on physical activity energy expenditure (1000 vs. 2000 kcal/wk) and intensity (moderate vs. vigorous). Participants also were told to reduce food intake to 1200 to 1500 kcal/d. (Doesn’t sound to me like an easy caloric intake to achieve on any of the usually recommended weight loss diets.)

According to the abstract, weight loss did not differ among the randomized groups at 6 months (8–10% of initial body weight) or 24 months (5% of initial body weight) follow-up. (In other words, calorie restriction on whatever diet they were using was not sustainable; or starvation is not fun!) Post-hoc analysis showed that individuals sustaining a loss of 10% or more of initial body weight at 24 months reported performing more physical activity (1835 kcal/wk or 275 min/wk) compared with those sustaining a weight loss of less than 10% of initial body weight (P < .001).

The authors concluded that the addition of 275 mins/wk of physical activity, in combination with a reduction in energy intake, is important in allowing overweight women to sustain a weight loss of more than 10%. They also concluded that interventions to facilitate this level of physical activity are needed. In other words, the current recommendations for exercise to achieve and maintain weight loss are not nearly enough—at least not as currently practiced.

We couldn’t resist posting our own observations about weight loss and exercise. From the female perspective:

Back when I was a wee lass (in college), I recall weighing myself in the locker room after a tough workout for the crew team, and weighing in at 120 lbs. My usual weight was probably more like 125 lbs. (It sure would be easier to sprint up the stadium steps at 125 lbs than my current weight of 150 lbs!) I don’t know for sure whether there was any weight change as a result of crew training, but I remember eating enormous amounts during spring break when we worked out three times a day! Crew workouts can easily spend energy at the rate of 1000 kcal/hr (though few could keep up that pace for very long).

Then in my mid-twenties, I jogged a lot and ran several marathons. I didn’t train excessively—I probably logged 35+ miles per week, with occasional longer weeks when preparing for longer distance events. It helped me deal with stress and improved my confidence and self-image. Overall, jogging was a good thing for me, though I was never fast. (I almost qualified for the Boston marathon once, but got injured and had to slow down-- I could never be a world class athlete). I was thin then, but not skinny. I don’t recall even having a scale then, so I have no data on my actual weight (though I still have some clothes from that era, which currently do not fit). I remember, in particular, a friend asking if I was having any menstrual cycle disturbances from all this running, and I replied no. He felt my arm and said something like “of course not” in a derogatory voice, meaning I was way too fat for that. Did I say friend? I meant obnoxious jerk!

At any event, I was slim, healthy and strong (relatively), looked better than ever, and ate plenty of carbs (although I recall eating mayonnaise straight out of the jar once after a long run I was so starving!) I was also single, and that I think turns out to be the real kicker here. At least for me, I am less likely to eat well when eating alone. Conversely, I tend to overeat at times when eating with friends and family. I suspect this may be true for many others as well. So I can’t say for sure that my relative slenderness as a young runner and marathoner was only due to the exercise.

Twenty odd years later, running doesn’t do the same thing for me. I have a family—we cook great meals in our house, enjoy desserts, and generally relish living. Is it the inevitable middle-aged spread? Is it due to reductions in hormones related to the aging process? Not exercising enough? Is it due to excess calories in general? Or fat or carbs in particular?

I started using an ankle brace regularly on trail runs (see for example Monte Bello OSP or the Bay to Ridge Trail) about 2 years ago to prevent the perennial ankle sprains I am susceptible to (it has worked well!). I was able to train more or less consistently for longer periods of time. My weight stabilized, but I couldn’t seem to lose any weight, even with ever increasing mileage. An hour or so of jogging or uphill walking most days was typical, with occasional 3–4 hour outings, frequently covering a few thousand feet in elevation change. You’d think the weight would just fall off, but it didn’t. Any interruption in training saw the weight pile on quickly too (whether due to breaks from work demands, bad weather, holidays and holiday eating, illness or injuries). I think there is a tendency after long or hard exercising to believe that you are entitled to a few extra calories, and slightly larger serving sizes of food become habitual, and there goes any weight loss! It’s trivial to eat a few hundred calories extra, and not so trivial to burn them off. (See, for example, the chart at Also, there is evidence that exercise increases insulin secretion and growth hormone production (but that’s another post…). Certainly I have found myself famished after longer training events, and not satisfied with just one meal.

So why did it seem to work when I was in my twenties? I’m not sure—there are many factors that could have contributed. More importantly, why is it working now that I’m restricting carbs in my diet? I’m not even sure the exercise has anything to do with it. We’ve seen some of the fastest decreases in weight when exercising the least. For example, after an injury (I dislocated my elbow in early May) and cutting back on distances and speed, I spent a whole month with my weight consistently below the trend line on my weight loss graph. Afterwards, we started doing more distance again and pushing the pace, and the weight actually increased for a few weeks. (See our earlier posting about weight loss curves.)

From the male perspective:
My experience is, I think, quite typical of active males. I've always been a fast eater with a relatively high metabolism. Fast eaters tend to overeat a bit, and I tended to be 5–10 lbs above my ideal weight as I was growing up. I then went through periods when I was extremely active and periods when I was relatively sedentary. As a competitive varsity athlete in college, the weight naturally came off. It also came off when I spent summers mostly outdoors backpacking a lot and when I worked more physically active jobs organizing trail runs and doing construction work. I also ate more at those times. As I've gotten older, my weight has gradually gone up roughly along the typical pound-per-year curve. There have tended to be spikes when my level of activity decreased suddenly (when I stopped doing the physically demanding sport or job, and I had to relearn how to eat less and feel satisfied). The weight always seemed to have a local (in time) set point that was not very dependent on short-term variations in the amount I ate.

A few years ago, I switched once more from a period of high-activity work to more sedentary work, and my weight started drifting upward alarmingly even with significant efforts at reducing calorie intake. Essentially, I was starting to exhibit the typical set of symptoms known as metabolic syndrome: weight gain, increased waist circumference, moderate elevation of blood glucose levels, moderately elevated triglycerides, high blood pressure. If I paid close attention to my weight and caloric consumption, I could temporarily get some of the weight off, but it tended to come right back as soon as I stopped paying attention or went through a holiday period. Adding back in a strenuous exercise program also helped temporarily, but always tended to drive up my calorie consumption. Basically, I was confirming the usual experience that diet and exercise only sort of/maybe work to keep weight under control, and only if you pay close attention and keep paying attention. I've now lost all of the recent excess weight gain (about 25 pounds), though I'm still 30 pounds over what I weighed as a college athlete. The secret has really been exactly what my pediatrician told me 40 years ago before we all went low-fat, low-cholesterol: cut back on the simple starches and sugars! I don't make any attempt to calorie-restrict, or count calories—I'm rarely unusually hungry. I've just cut way back on the amount of sugars, potatoes, and grains I eat. End of story.

If anything, our relatively generous exercise schedule seems to lead to an ability to tolerate a few more carbs in our diet, but I’m sure if we restricted even more, we’d lose faster. Having adapted to a low carb diet, our metabolisms are acting in a carb-sparing capacity, using fat preferentially so that the glycogen is there for emergencies. It’s unclear whether we are even glycogen-depleted anymore. Certainly our blood glucose levels are stable—the body readily manufactures glucose from protein, and we eat sufficient protein so that we’ve never really been all that ketogenic (we’ve checked).

The body contains something like 400 g glycogen, stored in muscles and the liver. This glycogen binds 2.6 g water per gram of glycogen, and hence the total weighs 1440 g in total in the body—3.2 lb). Low carb critics often incorrectly state that the weight lost on a low carb diet is only “water weight,” due to the loss of glycogen stores. Those 400 g of glycogen theoretically contain about 4 kcal/g of metabolic energy (assuming efficient conversion), so that’s 1600 kcal worth of carbs that could theoretically be eaten each day if you burned all the stored glycogen for energy during exercise.

Of course, that’s not how it actually works in aerobic or mixed aerobic/anaerobic exercise. Unless you are exercising beyond your anaerobic threshold for long periods of time, or your aerobic fitness is very poor, fat is the preferred energy source for exercise. Just keeping to a pace where my heart rate is elevated (130–160 beats per minute) means I must cross the anaerobic threshold at least some of the time, but even so, there’s no way that I deplete all my body glycogen during normal exercise. One exception might have been a time about two weeks into our low carb diet when we went on a particularly difficult route, climbing and descending about 2000 ft and covering about 11 miles (in a cold rainstorm!). By the end, I was cramping and weak (might have been due to the cold too), and my blood glucose was down to 76 mg/dl. Every other time, even if exhausted to the point of feeling ill, my blood glucose was never below 90 mg/dl. I was never in any danger of completely running out of glucose for fuel (or fat!). (See this nice discussion of fat vs. glycogen burning in distance training.)

So if you are eating 2000 kcal at 60% carbs as recommended by many medical professionals, that's 1200 kcal from carbs consumed every day. But it's very unlikely that you will use up and need to replenish that much glycogen every day from normal exercise and living activities. Unfortunately, the body naturally stores at least some of the excess as fat.

So, our conclusion to date is that exercising is a great thing to do, but is not necessary or necessarily even helpful for weight loss at all (though of course it is very helpful for general health, increasing insulin sensitivity, raising HDL levels, etc.) On the other hand, if you eat large amounts of carbs, you'd better burn it off somehow or the likely destination for a lot of the excess is your fat stores!

Saturday, July 26, 2008

Fruit is Good for you, Right?

We've all been taught that we should eat plenty of "fruits and vegetables." Fruits and vegetables are rarely separated out in nutritional guidelines. They are separated but paired in the USDA food pyramid which recommends eating about equal parts "fruits" and "vegetables." (A more detailed version of current USDA recommendations can be found at and Parents are typically happy if they can get their children to eat lots of fruit juice and fruit products (typically with added sugar) when they won't eat much in the way of vegetables.

Even among those who are supposedly being more discriminating about their food choices, fruits are almost always treated as something to eat in large, if not unlimited, quantity. For example, the Paleo Diet, which attempts to recommend a diet alleged to be similar to that of our hunter-gatherer ancestors, recommends consuming "all the fruit and non-starchy vegetables you can eat." (You are warned against eating too much dried fruit—most fresh fruits are about 80–90% water, so removing most of the water can dramatically increase the total quantity you can eat. Similarly, some people warn against drinking too much fruit juice which is another way you can consume a lot more fruit very quickly.)

Fruits are alleged to be good for you, because they contain a lot of essential vitamins and minerals as well as fiber and such newer necessities as "phytochemicals" and "antioxidants." So what's really in fruit? Even those of us who habitually read nutrition labels tend to be unfamiliar with the nutritional content of most fruits, since, as typically sold fresh, there aren't any nutritional labels. The following table gives some data for a lot of common (and a few less common) fruits.

Fruit Protein Fat Carb Fiber Sugar Fructose Glucose Sucrose Other
All based on 100 gm serving
Apple 0 g 0 g 13 g 1 g 10 g 6 g 3.2 g 0.8 g 7% C

Apricot 1 0 11 2 9 0.9 2.4 5.9 39% A, 17% C, 7% Potassium
Banana 1 0 23 3 12 4.9 5.0 2.4 15% C, 18% B6, 10% Potassium, 13% Mn
Blackberry 1 0 10 5 5 2.4 2.3 0.1 35% C, 32% Mn
Blueberry 1 0 14 2 10 5.0 4.9 0.1 16% C, 24% K, 17% Mn
Breadfruit 1 0 27 5 11

48% C, 7% Thiamin, 6% Mg, 14% Potassium
Cherry (sour) 1 0 12 2 8 3.5 4.2 0.8 26% A, 17% C
Cherry (sweet) 1 0 16 2 13 5.4 6.6 0.2 12% C, 6% Potassium
Cranberry 0 0 12 5 4 0.6 3.3 0.1 22% C, 6% E, 6% K, 18% Mn
Date 2 0 75 7 66 32.0 33.7 0.5 8% Niacin, 12% B6, 8% Pantothenic Acid, 6% Ca, 14% Mg, 6% P, 20% Potassium, 18% Cu, 15% Mn
Fig 1 0 19 3 16

6% K, 6% B6, 7% Potassium, 6% Mn
Grapefruit (pink) 1 0 11 2 7 1.8 1.6 3.5 23% A, 52% C
Grape (red or green) 1 0 18 1 15 8.1 7.2 0.2 18% C, 18% K, 5% Potassium, 6% Cu
Guava 3 1 14 5 9

12% A, 381% C, 6% B6 12% Folate, 12% Potassium, 11% Cu, 8% Mn
Kiwi 1 1 15 3 9 4.4 4.1 0.2 155% C, 7% E, 50% K, 6% Folate, 9% Potassium, 6% Cu
Lemon 1 0 9 3 3

88% C

Lime 1 0 11 3 2

48% C

Mango 1 0 17 2 15

15% A, 46% C, 6% E, 7% B6
Melon (honeydew) 1 0 9 1 9 3.0 2.7 2.5 30% C, 7% Potassium
Nectarine 2 0 15 2 11 1.4 1.6 4.9 7% A, 9% C, 6% Niacin, 6% Potassium
Orange 1 0 12 2 9

75% C, 7 Thiamin
Papaya 1 0 10 2 6

22% A, 103% C, 10% Folate, 7% Potassium
Passion fruit 2 1 23 10 11

25% A, 50% C, 8% Riboflavin, 7% Niacin, 9% Fe, 7% Mg, 7% P, 10% Potassium
Peach 1 0 10 1 8 1.5 2.0 4.8 7% A, 11% C
Pear 0 0 15 3 10 6.2 2.8 0.8 7% C, 6% K
Pineapple (sweet) 1 0 13 1 10 2.2 1.7 6.5 94% C, 6% B6, 6% Cu, 41% Mn
Plum 1 0 11 1 10 3.1 5.1 1.6 7% A, 16%, C
Pomegranate 1 0 17 1 17

10% C, 6% K, 6% Pantothenic Acid, 7% Potassium
Raspberry 1 1 12 6 4 2.4 1.9 0.2 44% C, 10% K, 34% Mn
Rhubarb 1 0 5 2 1

13% C, 37% K, 9% Ca, 8% Potassium, 10% Mn
Strawberry 1 0 8 2 5 2.4 2.0 0.5 98% C, 6% Folate, 19% Mn
Tamarind 3 1 63 5 57

6% C, 29% Thiamine, 10% Niacin, 7% Ca, 16% Fe, 23% Mg, 11% P, 18% Potassium
Watermelon 1 0 8 0 6 3.4 1.6 0.1 11% A, 13% C

(All data from Vitamin and mineral content are given as % daily value. 'K' is Vitamin K; potassium is spelled out.)

As you can see, fruits are mostly sugar and water. There's typically about 1% protein, no fat, a modest amount of fiber, and a very small amount of more complex nutritive carbohydrate. Some fruits are admittedly good sources of particular vitamins and minerals, but you get a pretty generous "spoonful of sugar" with your daily dose of vitamins and minerals if you get them mostly from fruit. If you're after the trace nutrients, you'd be better off with vegetable sources anyway. Compare the following (the first three of which are arguably "fruits" in the botanical sense, although we usually call them "vegetables," because of the ways we tend to serve them).

"Vegetable" Protein Fat Carb Fiber Sugar Fructose Glucose Sucrose Other
Tomato 1 0 4 1 3 1.4 1.2 0.0 17% A, 21% C, 10% K, 7% Potassium, 6% Mn
Cucumber 1 0 4 0 2 0.9 0.8 0.0 21% K
Green Bean 2 0 7 3 1

14% A, 27% C, 18% K, 6% Thiamine, 6% Riboflavin, 9% Folate, 6% Fe, 6% Mg, 6% Potassium, 11% Mn
Broccoli 3 0 6 2 2 0.7 0.5 0.1 12% A, 149% C, 127% K, 7% Riboflavin, 9% B6, 16% Folate 6% Pantothenic Acid, 7% P, 9% Potassium, 10% Mn

There doesn't seem to be any good reason to eat fruits for their higher vitamin content; the vegetables have more vitamins anyway! Vegetables provide generous amounts of fiber as well.

So then maybe fruit sugars are somehow better for you? Certainly many "natural foods" advocates seem to believe that "natural" and "unrefined" sweeteners are somehow OK when table sugar is not. But the sugar in most fruits is nutritionally almost the same as that of table sugar. Table sugar is sucrose which is a disaccharide made up of equal parts fructose and glucose. Most fruits also have approximately equal amounts of fructose and glucose either separately or combined as sucrose. The body breaks apart the sucrose right away. Fructose and glucose are metabolized somewhat differently. While fructose has an apparently lower glycemic response, it appears to be more harmful overall resulting in elevated triglyceride levels and fat storage much more than does glucose. So focusing on apples and pears with their higher fructose content is probably even worse than going for the fruits with more balanced sugar content. A more detailed discussion of the metabolism of fructose and glucose can be found in Basciano et al., "Fructose, Insulin Resistance, and Metabolic Dyslipidemia," Nutr. & Metab., 2005, 2:5.

So my conclusion from all this is that fruits are not health food! They're by no means terrible, especially when consumed in moderation as whole fruit, but as macronutrients, they are primarily bags of flavored sugar water. Particularly if you are trying to cut back on total carbohydrate consumption and on sugar consumption in particular, you should seriously consider limiting your fruit intake and increasing your vegetable intake instead, or at least focusing on lower-sugar fruits such as raspberries, blackberries, and cranberries. For us, fruit is relegated to the role of dessert, to be consumed once or twice a day in ~½-cup portions, to alleviate carb cravings. This way, the sugar loading from fruit is kept to a minimum of 20–25 g.

Sunday, July 20, 2008

More on the Thermodynamics of Body Weight

The previous post was intentionally general with no specific finger-pointing to misuse or good use of thermodynamic arguments. The misuse is so widespread, that I still prefer not to provide specific references. However, it is worth pointing to a few recent papers that have taken the issue more seriously.

First, I should note that to a high percentage of even trained scientists and doctors, "thermodynamics" means "equilibrium thermodynamics." This is all that is taught in a typical introductory course. Unfortunately, the human body can rarely be modeled accurately as being in any sort of equilibrium. At best, it may be reasonable (at least for some time scales) to model it as being in something approaching a "steady-state." As such, a complete analysis must include a study of "dynamic," "kinetic," or "nonequilibrium" effects, i.e., you must look at all inputs and outputs, and potentially the rate of change of inputs, outputs, and the state of what's inside.

One recent paper which tackles the issue head-on is by Feinman and Fine: "Nonequilibrium thermodynamics and energy efficiency in weight loss diets," Theoretical Biology and Medical Modelling 2007, 4:27. The authors specifically focus on the greater weight loss observed in low-carbohydrate diets with an emphasis on the specific "kinetics" of fat storage and dissipation arguing that simple equilibrium models fail, because real bodies, and especially real bodies that are in transition (gaining or losing weight) can be far from equilibrium requiring the consideration of dynamic effects. While they find that no experiment exists that measures all relevant variables, they are able to find evidence that dietary carbohydrate controls fatty acid storage and release via its effect on hormone levels (particularly insulin) and that this "nonequilibrium" effect can explain the greater weight loss of low-carb diets.

Another paper by Schulz and Schoeller, "A compilation of total daily energy expenditures and body weights in healthy adults," Am J Clin Nutr 1994, 60:676, reviews 22 studies which use an important (if somewhat expensive) technique for measurement of what they call "total expenditure energy" (how much energy is used over, say, a day or a week for all physical activity. The technique uses "doubly labeled water" (both the hydrogen and oxygen atoms are non-standard isotopes so that the differential kinetics of hydrogen and oxygen can be measured). The authors present data for various populations from elite athletes to normal and overweight individuals. Most notably, the total daily energy expenditure is found to vary by a factor of more than 3. To a limited extent, they were able to separate out the contributions of basal metabolic rate and physical activity, fat-free body mass and excess fat. While not specifically focused on thermodynamic equilibrium, it does add further evidence against any simple "calorie is a calorie" model of diet and weight.

And not being completely unable to resist highlighting one paper that is unwilling to draw the right conclusions from its own data, consider Brehm et al, "The role of Energy Expenditure in the Differential Weight Loss in Obese Women on Low-Fat and Low-Carbohydrate Diets," J Clin Endocr & Metab, 2005, 90:1475. Their paper reports the results of weight loss comparisons for obese women following the two diets for six months. The Low-carb group lost 50% more weight despite not being specifically calorie-restricted. (Food diaries indicated that the caloric intake was nevertheless similar between the two groups.) Not being able to measure any significant difference in resting energy expenditure or physical activity, the authors fell back on the old standby: the low-fat dieters must have cheated and underreported actual food consumption! The possibility that the results could be real apparently wasn't acceptable.

Saturday, July 19, 2008

"Thermodynamics" of Weight Gain and Loss

A common belief among the proponents of calorie restriction and exercise as a means of weight loss is that "thermodynamics" (or more particularly, conservation of energy) says it must be so. The basic premise is that:

Energy-in - Energy-out = Energy-stored-as-fat

Therefore, if you want to lose weight (make Energy-stored-as-fat negative), then Energy-out must exceed Energy-in. This is then interpreted to mean that you must either eat less or exercise more to change this balance. People then go into elaborate calculations as to how many Calories you burn with various sorts of exercise and how many Calories are in different diets.

A few authors have recognized that this same equation can be interpreted a little differently. They argue that no one can actually regulate their energy intake precisely enough to have a constant weight over any length of time. Therefore Energy-stored-as-fat must be regulated by the endocrine system and be more or less fixed. Then, the thing that has to change when you eat more or less is the Energy-out; i.e., the body adjusts its energy expenditure (without a specific conscious exercise program) to burn available excess energy, and conversely, if you get an unusually large amount of exercise (run a marathon, for example), then you will naturally eat more to replace the energy used.

I believe the real picture is a good deal more complicated than any of these simple descriptions. As a physicist and thermodynamicist, I look at a "control volume" (a human body in this case), and consider all of the flows of mass and energy in and out. Any net differences must be stored internally. Inputs include the food and drink consumed and the air breathed in (most of the time, direct energy inputs [mechanical or thermal] can probably be ignored as small). Outputs include waste excreted, air breathed out (with a different composition!) and any mechanical work performed. It is important also to recognize the energy associated with differences in the internal potential energy associated with the any chemical changes that may occur between matter taken in and matter excreted. These represent a form of possible difference between energy-in and energy-out.

So, consider someone who is eating, say, an "extra" 1000 Calories a day. He could store it as a quarter pound of fat and gain a couple of pounds per week. He could get restless and move around a lot more, expending at least some of the excess as mechanical energy. His peripheral capillaries could dilate so that he loses more energy thermally through his skin. He could simply excrete some of it as undigested. I have always suspected, for example, that there is an upper limit to the rate at which food can be processed in the body, and if you exceed that limit by binge-eating or feasting on a particular day, then most of the excess is simply excreted rather than stored.

People often report that even large increases in the amount of exercise they get does not result in weight loss. Some portion of this may be a consequence of muscle development ("good" weight gain). But to a large extent, it is very hard not to just eat more in response to the body's repair and recovery processes after exercise.

Similarly, if you try to lose weight by starving yourself, your body tends to try to conserve energy to compensate (and you may be chronically suffering as your body tells you to eat more!). People who literally go on starvation diets (either intentionally or as a result of disaster) often report a permanent reduction in their "metabolism" to the point where they eat much less after the period of starvation (or quickly gain back any weight lost).

All of this means, that if you really want to change body weight (either up or down), you really need to be tinkering with the endocrine system to stimulate release or storage of fat. That's rather complicated, and I don't understand it very well myself. It's a bigger topic than I want to tackle today, anyway. Moderate amounts of exercise are probably beneficial. But probably the simplest thing that most people can do is to control carbohydrate consumption, particularly simple sugars and starches (including those found in "whole" grains and most fruits). These tend to set off a chain of hormonal events including a large spike in insulin production, triglyceride production and ultimately fat storage. If you keep your insulin and triglyceride levels low, you tend to take fat out of storage instead.

Anyway, the purpose of this post is not to promote any particular approach to weight management, but merely to argue that if you're going to invoke "thermodynamics" in discussing it, make sure you are properly accounting for all the energy inputs, outputs, and transformations that are occurring, and be very careful that you are drawing the right conclusions.

Thursday, July 17, 2008

Among Diabetics, only the Mediterranean diet decreased plasma glucose?

The recent New England Journal of Medicine article comparing long term results with low fat, Mediterranean and low carb diets published yesterday, and made quite a splash in the media. The article is free to the public (Thank you NEJM!) As usual, either the authors or the media, in their fervor to get a "message" out to the public, screwed it up again. The message everyone seems to be taking from this study is that Mediterranean diets are better for diabetics than low fat or low carb diets, the other two diets tested, when that is not what the article says at all!

It doesn't help that the authors state in the abstract that

"Among the 36 subjects with diabetes, changes in fasting plasma glucose and insulin levels were more favorable among those assigned to the Mediterranean diet than among those assigned to the low-fat diet..."

Note that they did NOT say "more favorable than among those assigned to the low-carb diet." Was this a concession to mainstream medicine? Did they feel that they would take too much flak or not get funding for their next study if they went too far in supporting low-carb "fad" diets?

It is beyond me why they did not state in the abstract that those on low carb diets had the lowest glycated hemoglobin levels, since we know that glycated hemoglobin levels are a much better measure of average blood glucose levels and blood glucose control than a few isolated fasting glucose values. This seems to me to be one of the strongest points from a public health perspective, and it needs to be said louder.

Think I am hallucinating? Reading into the data what I want to see? Right there on page 235 (and continuing on page 238) the article says:

"Among the participants with diabetes, the proportion of glycated hemoglobin at 24 months decreased by 0.4±1.3% in the low-fat group, 0.5±1.1% in the Mediterranean-diet group, and 0.9±0.8% in the low-carbohydrate group. The changes were significant (P<0.05) only in the low-carbohydrate group (P = 0.45 for the comparison among groups). I suspect that Mary Vernon and Eric Westman will have similar observations (see their their earlier thoughts on the ACCORD trial and this one (with a hoard of other authors too). (Thank you for pushing the issues with mainstream medicine!)

The authors of the NEJM article calculated the "homeostasis model assessment of insulin resistance" (HOMA-IR) according to the following equation (see Diabetes Care 1997;20: 1087-92) as follows : insulin (U/ml) × fasting glucose (mmol/liter) ÷ 22.5. Their observations about HOMA-IR are interesting, and serve as the basis for their conclusion that the Mediterranean diet worked better for diabetics than the low fat diet. What they again fail to state (though it is also shown in Fig. 4F) is that the HOMA-IR improved most on the low carb diet in non-diabetics. This point is certainly relevant from a public health perspective for diabetes prevention, is it not?

Another curiosity is that the low carb dieters were counseled to choose vegetable sources of fat and protein (ostensibly to reduce trans-fats—Since when is animal fat a significant source of trans-fats?)? I'd like to know what if any effect this urging had on the dieters, and what effect it had on their lipid levels and other parameters measured (if those measures are even relevant…).

Ranting aside, this was a nicely done study, with better control than in many, and for a longer period of time. It should end the speculation and criticism that low carb diets show short term benefit only. It's clear that the weight losses were much better at 5–6 months for both low carb and low fat diets, and that relaxation of dietary stringency most likely resulted in weight re-gains. Certainly, the increase in triglycerides in the low carb group demonstrates an increase in carb consumption (despite the higher fat intake during the more stringent phase) and is likely responsible for the weight re-gains. The Mediterranean diet seemed easier to maintain, judging by the better weight maintenance, but it's not clear why this should be.

The authors were not cagey at least about the more favorable effects on lipids with the low-carb diet, and even with less compliance and some weight re-gain in later phases of the diet, these results were maintained.

I am left with my usual confusion over what constitutes a "Mediterranean diet" though—as described, it seems that the only significant difference between the "low-fat" diet and the "Mediterranean diet" was the substitution of olive oil for some of the meat fat (and a slight increase in the total amount of fat). Does substitution of 5–7 nuts for a few pretzels turn a low fat diet into a Mediterranean diet? Certainly most "low fat" dieters believe they are eating "healthy" diets high in grains and vegetables, but low in "evil" fat. This study represents another nail in the coffin of the low fat dogma, and in that respect as well, I am pleased.

Saturday, July 12, 2008

What's the Expected Shape of a Weight Loss Curve?

We "discovered" low-carb diets about 6-7 months ago by reading Gary Taubes' recent book, Good Calories, Bad Calories (highly recommended, if admittedly somewhat controversial). Deciding that some of our own struggles with weight control even (or especially) in face of a 30-40-mile per week exercise program might be due to excessive carb consumption, we embarked on our own dietary revisions and experiments.

But regardless of the weight-loss method adopted, no one seems to show any accurate data on weight loss vs. time. You can find plenty of anecdotal data: "I lost 50 lbs. (10, 20, 100, whatever) in x weeks (months, days)." You can find before and after photos. You can find data correlated with lipid profiles, blood pressure, cardiac incidents, diabetes, even death. You can even find guidelines that suggest that you should plan to lose, say 1–2 lbs. per week, but is that right? Does it depend on how much you need to lose? Should the weight loss be linear over some period of time? Obviously, it can't be linear indefinitely, since your weight must necessarily level off at some point. The next simplest model might be an exponential approach to some target weight value. That model might arguably be correct if you make an incremental change to your diet/lifestyle that causes you to have a new natural weight setpoint. If that new setpoint is a bunch of pounds different from where you are now, what is the expected time constant for approaching the new set point? Or is that model too simple too? And for that matter, how much noise should you expect to see in the data? What "normal" day-to-day fluctuations should one expect to see? None of these questions seems to be answered in the literature, even though they are clearly easy to measure and clearly of great psychological importance to anyone trying to lose weight.

As scientists, we, of course, have been keeping records! The data are still somewhat limited and anecdotal. Observations so far:

(1) Low carb diets really work for us as means of reversing persistent long-term weight gain. We'll talk about our particular diets in another posting, but so far, we've seen losses of 10% body weight without much effort or hunger. It's a permanent diet change, but once you get used to it, it's painless, healthy and easy to continue.

(2) Over the time we have recorded so far (about 7 months), we still can't clearly distinguish between a linear fit and an exponential fit to the data. But all that really proves so far is that, if the exponential model is valid, we are seeing a time constant of somewhere in the range of 10-14 months. (We have each lost about 20 lbs so far and expect to be able to lose another 20 or so.)

(3) There are normal daily fluctuations of ± 1.5–2 lbs (remember that if you drink a pint, it's a pound added, and there are all sorts of things that can easily cause variations in fluid retention of a pound or two).

(4) There may be ~monthly cycles that may or may not match menstrual cycles that account for a couple of pounds of cyclic variations.

(5) There appear to be effects (usually temporary) associated with heavy exercise that go beyond simple fluid replacement issues. Again, these can account for temporary weight increases of a pound or two.

(6) There are also clearly seasonal and environmental effects due, at least in part, to the body's response to average temperature. This is another fluid retention effect, I think.

(7) There is often a perceived "plateau" effect where one seems to be stuck at certain levels for up to a few weeks at a time and then suddenly lose a few pounds. These are often reported anecdotally, especially when measurements are made infrequently (weekly or less often) but, in our case (daily measurements), we can't really prove that they're not just artifacts of the scatter in the data. We certainly see clusters of data points that are above and below the long-term trend lines, but we rarely see points that are appreciably outside a 1–2 lb. tolerance band about the longterm trend lines.