Resistance training enough to lose fat and build muscle?

October 22, 2007 5:54 AM

Can you lose fat and build muscle with resistance training alone? I'm with a trainer who claims that building muslce through resistance training and reasonable eating is enough to destroy fat. He claims there's little need for specific cardio sessions -- that there's enough cardio in his resistance sessions and in daily life. Wrong?

My trainer says if I want to do a specific cardio session, 1x/week is plenty.

I'm trying to lose the extra 20-25 pounds I have packed on and get down to a lean, muscular figure. Will a good resistance training program be enough?

posted by Malad to health & fitness (31 comments total)

You lose fat by eating fewer calories than you burn--whether you're burning them through cardio, sleeping, basket-weaving or, yes, resistance training. So yeah, he's not wrong.

That's not to say that you CAN'T lose fat doing cardio, however.
posted by chelseagirl at 6:00 AM on October 22, 2007

I'm not sure if that's the most efficient way to lose weight. It might be the most effective long-term; I've read that your body burns more calories in a resting state if you have more muscle. It's certainly not ineffective.

If I were you I would try to combine cardio and strength training. As follows:

--do not rest between sets. Instead, alternate two exercises so that while you're resting the one muscle group, you're working the other group. This way you maintain a high heart-rate the whole time. In other words: bench, do pullups, and then go right back to the bench.

--do not do piddly little isolation exercises that work only one muscle, like the butterfly machine, bicep curls, etc. Do exercises that work broad parts of your whole body. Examples: pullups, squats, deadlifts, bench, push press, renegade row, sumo high pull...you can find videos of all this at the crossfit website.

This is how I train nowadays. I ran cross-country as a teenager and I was never as out of breath and overwhelmed as I am nowadays from alternating resistance exercises. If you do 5 sets of alternating sumo high pull and bench, or alternating deadlift and pullups, you can puke. In other words, you can definitely get a cardio work-out.

That said, I think the most effective way to burn calories is to work for a long time at a moderate intensity, i.e. running steadily on the treadmill for an hour. But this is not the best way to get fit, fast, and strong. On the other hand, with strength training you end up eating a whole lot of protein just to regenerate muscle tissue. And as you will gain muscle, you will also gain weight.
posted by creasy boy at 6:18 AM on October 22, 2007

Building muscle will increase your calorie burn. More muscle demands more calories. Also, weight training will do remarkeable things for your overall tone and shape. And, finally, though it's not steady cardio where your heart rate levels out for a long session, spending 30-60 minutes in the gym lifting and moving around has some cardio benefits, too. You'll be drawing on more oxygen and burning away. If the routine he's given you is fairly intensive, it may be doing some double duty for you.

Go with what your trainer suggests for a while, and see what your results are.

I can say that when I had about 60 pounds to lose, I dropped 30 doing cardio only, over the course of about eight months without extreme calorie restriction. When I added weight training in the pace of change accelerated unbelievably. My body seemed to snap right to after only a few weeks of focused weight training, and the last of the weight came off much faster. Weight training is powerful stuff, and I'll never consider an exercise program complete again without including it.

However, I don't want to give up cardio fitness either. I like knowing I can run a fair distance at a fair speed if there's a medical emergency, or spend a day out vigorously hiking without feeling too drained. THere are lots of good reasons other than weight loss to improve cardiovascular function, and weight training is not likely to expand your lung capacity. I wonder what the trainer would say about that?
posted by Miko at 6:25 AM on October 22, 2007

Well, he has a point, BUT you'd benefit by exercise in a low heart rate zone. Low and slow. There is a physiological purpose to it, training your body to burn fat as fuel.

Besides, cardiovascular fitness is important to overall health, which should be right up there as a main objective. Your trainer isn't one hundred percent wrong, but his priorities might not be exactly what YOU need.
posted by konolia at 6:32 AM on October 22, 2007

Yes, you can lose fat with resistance training alone. Resistance training burns calories. Activity burns calories, no matter what kind of activity it is. I have a circuit training video that rules my world. It's resistance and cardio together, which is ideal in my opinion. Resistance training can get your heart rate up, it burns calories, increases muscle and bone density. It's a good thing. You don't have to do cardio but...

Aerobic exercise has many health benefits and shouldn't be overlooked. Not only does it burn calories, it can increase mood, increase stamina, decrease anxiety, raise your HDL (good) cholesterol, and strengthen your heart and lungs.
posted by LoriFLA at 6:34 AM on October 22, 2007

Of course. You don't need any exercise to lose weight, so long as you burn more calories than you eat.
Of course exercise has benefits over dieting alone - you can increase your calorie deficit without depriving yourself of nutrition.
Building muscle will increase your daily calorie burn. 1lb of muscle burns approx an extra 50 calories per day fwiw.

Cardio is recommended for health reasons, for everyone, not just the overweight. But whether you're getting enough cardio in your daily life, depends on your daily life. eg. My daily life largely involves sitting on the couch working on my laptop. The biggest cardio workout I get is walking upstairs to the toilet - so if a trainer told me that there was enough cardio in my daily life, that would be rubbish. But your life is probably completely different.
posted by missmagenta at 6:39 AM on October 22, 2007

Malad, hope you don't mind if I cite facts from your previous question: you're starting from a point of reasonable fitness and you apparently live in New York.

Given that you'll likely be able to make it through a pretty intense resistance workout at a good pace and, by living where you do, get a lot more cardio than most Americans in your everyday life (and lots of opportunities to get more, e.g., by taking the stairs when possible), I think your trainer is on the right track. There's also something to be said for focusing on one form of training before you diversify. You may want to add in more interval cardio for health reasons eventually, but keep in mind that, contrary to popular belief, endurance cardio is not the end-all-and-be-all of fitness.
posted by backupjesus at 7:03 AM on October 22, 2007

If your diet is spot-on then yes, resistance training is enough. I spent a long time just lifting and I got in pretty decent shape.

However, I will say that i got in much better shape when I added in 20 minutes of cardio after every weightlifting session. It also gives you a little bit more wiggle room in your diet, which is helpful too.
posted by PFL at 7:10 AM on October 22, 2007

Yes, restricted calories and increased activity in the form of a resistance program will help you to lose weight.

It's also true that muscle burn more calories, and so added muscle will help, although the addition from that is probably negligible. (The estimates of extra calories burned by a pound of fat vary from 13-50 calories per day. A pound of fat is lost after you expend but don't replace 3500 calories. You can do the math and see that adding muscle is not a good strategy for losing fat, although it might help with keeping it off.)

The benefit of cardio is that you burn more calories more quickly. A ~half hour run that's four miles long will burn +400 calories. Since it uses different muscles from the resistance muscles it can be added on to the other workout.

There are some questions about whether or not aerobic exercise alone will maximize weight loss, but what seems clear is that resistance training alone will not.
posted by OmieWise at 7:12 AM on October 22, 2007

Oh, and konolia's suggestion that cardio "trains your body to burn fat" is poorly stated. The comment should read: "Exercising at low aerobic intensity trains your body to burn fat over glycogen when you engage in that exercise at that intensity in the future."

If you want to burn more fat, but don't care about training to preserve glycogen, then exercising at a higher intensity burns more calories from fat, even if the percentage of total fat burned is lower. This is because the number of total calories burned from all sources is increased when the intensity increases.
posted by OmieWise at 7:16 AM on October 22, 2007

For a contrasting opinion, this article from Runner's World (not an unbiased source on the question, obviously) does some calculations to show that 40 minutes of running burns significantly more calories than 40 minutes of strength training and that the "metabolic boost" of increased muscle mass does not compensate for that.
posted by hydropsyche at 7:30 AM on October 22, 2007

Hydropsyche, that article says that weight-lifting is not as good "largely because strength training involves too much sitting and resting between lifts". But there's no reason why you have to sit and rest at all.

If this guy's trainer is following the body-building model of using isolation machines and sitting around for a few minutes between each set, then I agree this is a terrible way to lose weight. But 40 minutes of strength training can also be 40 minutes of intensive cardio if you do circuit training or hop back and forth between exercises. So really it all depends on how the trainer approaches strength training.
posted by creasy boy at 8:34 AM on October 22, 2007

The counterpoint to the Runner's World article is probably this one from T-Nation.
posted by backupjesus at 8:49 AM on October 22, 2007

Go with what your trainer suggests for a while, and see what your results are.

What miko said. You're paying this person a lot of money to be an expert, so give the program a chance unmodified (or adding just 1x cardio/wk). If, after say a month, you don't feel you are on the way to results you want, then think about switching up your workouts (or your trainer).
posted by and hosted from Uranus at 9:05 AM on October 22, 2007

Thanks for the responses so far!

But 40 minutes of strength training can also be 40 minutes of intensive cardio if you do circuit training or hop back and forth between exercises.

Yeah we basically do two different exercises at a time, alternating sets between them with no rest (usually 3 or 4 sets each). Then between these pairs of exercises, there might be a minute or two of rest -- just enough to set up the next pair. Typically there are four such pairs in the hour. So I guess that is cardio-ish.
posted by Malad at 9:19 AM on October 22, 2007

He's right to be saying that. Not to say there isn't a place for cardio, but it tends to be overdone and lead to a constriction and reduction in metabolism just when you want to build it.

Seconding T-Nation.
posted by Mr. Gunn at 9:26 AM on October 22, 2007

You lose fat by eating fewer calories than you burn

There are others who commented along the same lines. Those statement does not appear to be supported by current science. Caloric intake depends on your cells' demand for energy, and you stay hungry until that demand is met (Unless you simply stop eating, in which case hunger goes away in a couple of days). If you're exercising your caloric demand will be higher, therefore you will be hungry more often and need to eat more. Your cells will burn more glucose or fatty acids if no glucose is available. If there's to much glucose, then your body never burns the fatty acids and fat begins to accumulate in the body. This happens even if you exercise all day. Although exercise does reduce insulin levels to some degree, it does so only for about two hours. This time shrinks if your diet is high carbohydrate.

I posted an answer on another question describing what science has shown us about how fat gets accumulated in the human body. If you're looking to lose pounds, cut out carbohydrates, but eat until your satisfied (proteins and dietary fat). This reduces glucose and therefore reduces insulin and promotes your body burning fat as fuel. Exercising will reduce insulin levels even further. When you get down to the weight you want, simply start adding carbs back into your diet until you find the limit your body is able to metabolize while still burning enough fatty acids to keep you from gaining weight.

If you need more verification that exercise doesn't help with weight loss while having a high carb intake, simply run the numbers that others have laid out:

A pound of fat is lost after you expend but don't replace 3500 calories.

~ half hour run that's four miles long will burn +400 calories

That means a 4 mile run at 6 minutes per mile will take 24 minutes. At 450 calories per 24 minutes, to lose 1 pound you'd need to run 2.8 hours (Math check: 3500/450= 7.7 24 minute segments, so 7*24= 168 minutes, 168/60 = 2.8 hours). To lose 20~25 pounds you'd need to run somewhere between 56 and 70 hours hours. If your body was a static system and you stopped eating altogether, this could be done by running 2.8 hours a day for about 20~25 days. Unfortunately your body is dynamic, and over the course of the next 24 hours you will replace the 3500 calories burned by eating. If those meals are high in carbs, you'll probably eat more calories with a percentage of them going to fat.

One caveat to all this is different people have different tolerances for insulin. If you are overly insulin sensitive, only a small amount of insulin is needed which keeps your insulin levels naturally low. This would counteract the effects of a high carb diet, and combined with exercise could cause you to not accumulate fat even on a high carb diet. On the other end of the spectrum, if your suffering from hyperinsulinemia, then following the same routine just described, you could gain weight while doing that much exercise.
posted by herda05 at 8:03 PM on October 22, 2007

That means a 4 mile run at 6 minutes per mile will take 24 minutes. At 450 calories per 24 minutes, to lose 1 pound you'd need to run 2.8 hours (Math check: 3500/450= 7.7 24 minute segments, so 7*24= 168 minutes, 168/60 = 2.8 hours). To lose 20~25 pounds you'd need to run somewhere between 56 and 70 hours hours. If your body was a static system and you stopped eating altogether, this could be done by running 2.8 hours a day for about 20~25 days. Unfortunately your body is dynamic, and over the course of the next 24 hours you will replace the 3500 calories burned by eating. If those meals are high in carbs, you'll probably eat more calories with a percentage of them going to fat.

What? I'm not sure what you're talking about, except that you conveniently just assert that exercising is likely to result in increased calorie consumption. (The especially if you eat carbs part is just your own hobby-horse.) But I don't think anyone is arguing that serious weight loss is possible without restricting calories, so you're arguing against a straw man. Sure, when people begin to exercise they tend to lose fat, regardless of diet, but you may be confusing the focus of the question (exercise) with a prescription for weight loss.
posted by OmieWise at 4:24 AM on October 23, 2007

Herda05? Not to derail, but while hormones play a role in energy consumption, thermodynamics do as well--and there is just no possible way that eating fewer calories than you burn would not result in your body using fat, as well as glycogen and muscle to varying degrees, as fuel to make up an energy deficit.
posted by chelseagirl at 6:19 AM on October 23, 2007

I believe herda05's argument is based on Good Calories, Bad Calories by Gary Taubes, which, whereever one stands on low-carb diets, is a fascinating read. For all of criticism of Taubes, he ends up describing a diet that's pretty much Walter Willett's pyramid with the option of more red meat and maybe fewer whole grains.

Taubes' argument about exercise, which is supported by animal studies but, like pretty much all nutrition theories, needs more/better human studies, is that high insulin levels from refined carbs are fat-sparing, so creating a caloric deficit through exercise in a hyperinsulinemic person will not burn (much) body fat. Since the energy has to come from somewhere eventually (i.e., it may come from glycogen in the short term, but that glycogen will need to be replaced), it will come from lean tissue, short-term metabolic slowdown (lower heart rate, lower body temp, less semiconscious/unconscious movement, etc.), or food consumption. Caloric restriction, in this case, will result in some combination of muscle loss and a slower metabolism, so telling someone in this position to exercise without changing their diet will end up with a result somewhere between unproductive and really bad.

In comparison, a person with low insulin levels (either genetically or through a low-carb diet) will have adipose tissue that's releasing free fatty acids, so the energy can come straight from fat. It's still calories in/calories out, but the low-insulin approach maintains the metabolic rate both in the short and long terms (since lean mass is spared) while resulting in less hunger since the stored fat enegry is readily available to the body.

Taubes is certainly not the originator of these ideas -- he traces the history back quite a ways, in fact -- but he offers a lot of supporting evidence to make a convincing case that low-carb approaches should be the subject of broader investigations. Once again, though, he ends up relatively close to the current dogma, so his "they just won't listen!" attitude seems like posturing at times.
posted by backupjesus at 6:54 AM on October 23, 2007

Here's the concluding part of Gina Kolata's review of Taubes' book (It should be noted that neither are scientists):

But the problem with a book like this one, which goes on and on in great detail about experiments new and old in areas ranging from heart disease to cancer to diabetes, is that it can be hard to know what has been left out. For example, Taubes argues at length that people get fat because carbohydrates in their diet drive up the insulin level in the blood, which in turn encourages the storage of fat. His conclusion: all calories are not alike. A calorie of fat is much less fattening than a calorie of sugar.

It’s known, though, that the body is not so easily fooled. Taubes ignores what diabetes researchers say is a body of published papers documenting a complex system of metabolic controls that, in the end, assure that a calorie is a calorie is a calorie. He also ignores definitive studies done in the 1950s and ’60s by Jules Hirsch of Rockefeller University and Rudolph Leibel of Columbia, which tested whether calories from different sources have different effects. The investigators hospitalized their subjects and gave them controlled diets in which the carbohydrate content varied from zero to 85 percent, and the fat content varied inversely from 85 percent to zero. Protein was held steady at 15 percent. They asked how many calories of what kind were needed to maintain the subjects’ weight. As it turned out, the composition of the diet made no difference.

As I read Taubes’s book, I kept wondering how he would deal with an obvious question. If low-carbohydrate diets are so wonderful, why is anyone fat? Most people who struggle with their weight have tried these diets and nearly all have regained everything they lost, as they do with other diets. What is the problem?

On Page 446, he finally tells us. Carbohydrates, he says, are addictive, and we’ve all gotten hooked. Those who try to break the habit start to crave them, just as an alcoholic craves a drink or a smoker craves a cigarette. But, he adds, if they are addictive, that “implies that the addiction can be overcome with sufficient time, effort and motivation.”

I’m sorry, but I’m not convinced.

Note, particularly, the second paragraph. That's why I called it a hobby-horse. There are many people who've decided that all carbs are bad, and I'm sure many of them have lost weight following their convictions. That doesn't make the assertions any more definitive or scientific, however.
posted by OmieWise at 7:20 AM on October 23, 2007

Because Taubes has me in a question-everything kind of mood, I tracked down what I'm moderately sure is the article Kolata is talking about here. It's based on studies from the '50s and '60s, but was published in 1992 -- there may be earlier works based on the same research.

IANA nutritionist or much of a statistician, but I found three items of note:

1) Of the 13 adult subjects, only 4 had BMIs over 25, and only 1 over 30. Given that all subjects were over 40 years old, this does not seem like a population predisposed to whatever causes obesity.

2) None of the subjects was fed 0% carbs -- 15% appears to be the bare minimum (which makes me wish Kolata had cited a particular article). 15% fits most definitions of "low carb," but it's not Atkins-induction-style minimal carbs.

3) All carbs were in the form of celerose, which is a form of glucose. A liquid cocktail of 15% glucose calories, 15% milk protein, and 70% fat would seem custom-made to trigger a insulin response and provide plenty of blood glucose and fatty acids to get taken up by the adipose tissue.

On the calorie-is-a-calorie front, Taubes is not suggesting that calories are metabolised at different rates, but rather that insulin controls where those calories end up. Some of the animal-study examples he cites are anecdotally interesting, but he readily admits more work needs to be done.

(And, lest we forget, Kolata believes that sustained fat loss is pretty much impossible; in the long run, she doesn't think exercise will help, either.)
posted by backupjesus at 8:55 AM on October 23, 2007

Thanks backupjesus. I just finished reading the study you linked to. Kolata's article actually mischaracterizes the study completely. From the abstract:

Diets rich in fat may promote obesity by leading to a greater deposition of adipose-tissue triglycerides than do isoenergetic diets with less fat. This possibility was examined by a retrospective analysis of the energy needs of 16 human subjects (1 3 adults, 3 children) fed liquid diets of precisely known composition with widely varied fat content, for 15-56 d (33 ± 2 d, iı ± SE).

The study was designed to test if diets rich in dietary fat lead to obesity over diets with more carbohydrates. It was not intended to test whether reducing insulin levels humans would reduce fat accumulation and weight. The study does not counteract the carbohydrate idea (that refined carbs and high glucose cause weight gain) of weight gain at all. It is also evidence against the idea that high-fat diets cause fat accumulation and weight gain.

Chelseagirl, the alternative idea of weight gain does not break the laws of thermodynamics. Rather scientists advocating the alternative ideas argue that the 1st law of thermodynamics has been misinterpreted. I don't want to get to far away from the OP's question, so I'll quickly summarize. The 1st law says:

Change in energy stores (fat, glycogen, etc.) = Energy intake (calories eaten) - Energy expenditure (calories burned)

The standard model assumes the right side of the equation causes the left side. Overeating and no activity cause fat storage. But that may not be the case. It's just as possible that any change in the left side causes the right side. A change in energy stores could be the driving force behind positive caloric balance and result in overeating and reduced activity.

Here's a link that describes how the second law works with regards to chemical reactons.
posted by herda05 at 10:46 AM on October 23, 2007

OmieWise, the metabolism of carbohydrates is not my own personal hobby-horse that I'm throwing out there. There are plenty of clinical animal and observational human studies showing that reducing refined carbohydrate and sugar consumption reduces fat accumulation. There's also plenty of evidence regarding using high carb diets to increase weight (sumo wrestler diets for one).

As far as the OP's question, I thought it was this:

I'm trying to lose the extra 20-25 pounds I have packed on and get down to a lean, muscular figure. Will a good resistance training program be enough?

To which my response was a definitive No. Dietary changes are necessary to lose weight.

What science (obesity research, endocrinology, diabetes research, physiology) is showing is that simply exercising (it doesn't matter what kind) will not cause you to lose 20-25 pounds. Evidence indicates that our bodies work to keep homeostasis, a state of internal stability against the variability of the outside world. Your energy storage is controlled involuntarily by the body, much like your height is not something that you can consciously control. Increasing activity levels does not cause weight loss because your body uses caloric intake and activity level to maintain energy storage. If you voluntarily increase activity level, your body compensates by increasing caloric demand (and hopefully intake). If the caloric intake does not match demand, then hunger ensues. Hunger continues until caloric demand is met, at the same time the body reduces energy levels to compensate for the lack of caloric intake.

This was the point of my paragraph that you excerpted. I wasn't asserting that exercise increases caloric demand, I was stating an accepted fact of physiology. If you find a reference stating that exercise does not increase caloric demand, I would love to read it. I have not come across a single citation in anything I've read thus far.

Yes, restricted calories and increased activity in the form of a resistance program will help you to lose weight.
...
posted by OmieWise at 7:12 AM on October 22

But I don't think anyone is arguing that serious weight loss is possible without restricting calories, so you're arguing against a straw man.
...
posted by OmieWise at 4:24 AM on October 23

There are at least two other statements regarding restricting calories in the comments above. The one I've excerpted is from your first comment. Unless I'm reading it wrong, you are arguing that weight loss is possible by restricting calories. 20~25 pounds is serious weight loss and restricting caloric intake to lose that much weight is pretty much doomed to failure. Eventually hunger will get the best of you, while also causing a reduced energy level.
posted by herda05 at 11:45 AM on October 23, 2007

Kolata's article actually mischaracterizes the study completely.

How does Kolata's article mischaracterize the study? She says that it asked if a calorie was a calorie was a calorie, and the study did indeed find that to be true (within the limits of it's design, which she layed out precisely). Her larger point, that prior conviction seems to drive much of the writing on this topic appears to be borne out by your own reaction. While I agree that the study supports the fact that high-fat diets do not cause weight gain. It does however support the idea that they do not promote weight loss over diets with higher levels of carbohydrates. Indeed, they quote studies that suggest that 1) gaining weight is not influenced by the type of diet fed (Forbes), and 2) that switching from a high fat to a lower fat diet caused a need for increased calories and still resulted in weight loss (Prewitt). While the two studies are contradictory, neither supports the notion that diet composition in terms of macro-nutrients is the determining factor for weight gain and loss. In all, there's nothing in the article that suggests that diet composition is the determining factor, and the study suggests that "a calorie is a calorie is a calorie."

The study does not counteract the carbohydrate idea (that refined carbs and high glucose cause weight gain) of weight gain at all.

This is completely incomprehensible to me. Certainly a variance of between 15-70% of calories from simple sugar would produce a variance in insulin response were such a response so very determinative of weight gain? If "carbohydrate hypothesis" sees no difference between those percentages then it's hard to understand how it might be applicable to the real world, or indeed, how it might account for the huge populations who eat largely carbohydrate bound diets without dying of obesity.

The bottom line is: Kolata is correct to suggest that this is a complex field in which there is enough evidence to pick and choose your studies to support your chosen conclusions.
posted by OmieWise at 11:57 AM on October 23, 2007

Sorry, didn't preview.

I'm not sure what you mean by doomed to failure. People lose weight by restricting calories and adding exercise all the time. Almost all success stories I've ever seen include both components, and I've seen very few success stories that come from changing the percentage of macronutrients alone. Certainly the National Weight Loss Registry reports that 94% of people who lose weight and keep it off exercise.

This was the point of my paragraph that you excerpted. I wasn't asserting that exercise increases caloric demand, I was stating an accepted fact of physiology.

Go back and read your excerpted paragraph again. You aren't accounting for any metabolic use of energy other than those from the exercise you talk about. Someone running ~3 hours per day is burning well over 3500 calories over 24 hours. It's a telling admission. You've decided to do the math without actually adding in all the variables. Sure, you can eat enough calories to replenish any deficit, but the more calories you burn the harder it is to do that. If you put some effort into restricting your intake so as not to replace all that are burned, you'll lose weight.
posted by OmieWise at 12:10 PM on October 23, 2007

The study that backupjesus linked (and we're assuming that's the study in question) to did not test whether a calorie is a calorie. It was testing the hypothesis that diets rich in dietary fat promote obesity. Those are two completely different testable hypothesis.

Also, keep in mind that this study wasn't actually performed on the subjects, but gleamed from records from other studies:

This technique was made possible by the careful collection of data on the effects of formula feeding, in studies done over many years by EH Ahrens and his colleagues at the Hospital of the Rockefeller university

These doctors did not hospitalize patients and perform any tests. They took an initial cohort of records from other trials where patients were fed liquid formula diets:

The records of all patients studied from 1955 to 1965 were searched for consecutive periods in which patients were fed formula diets of differing CHO-to-fat ratios.

They then formulated a criteria for inclusion from this cohort:

For inclusion in this analysis, each period of formula feeding had to be [greater than] 2 wk and the subject had to have remained weight stable (to within 1 kg) within that period.

They then rejected various records from this cohort using the said criteria:

Reasons for ejection of records were lack of feeding of two different formulas or the absence of feeding periods of [greater than] 2 wk. A total of 47 patient records were reviewed to obtain 16 (13 adults and 3 children) fitting the above criteria.

So, let me summarize my understanding of this study:

1) The researchers took 47 patient records from various other studies which did not have anything to do with the hypothesis the researchers were currently testing.

2) They then created a criteria that eliminated anyone who's weight deviated more than 1kg, lack of feeding of two different formulas, absence of feeding periods longer than 2 weeks.

3) This left the researchers with a cohort of 16 out of 47, approximately a third of the original cohort.

#2 indicates to me that the finding that "weight did not change" was not so surprising a conclusion. If you only include patient records where the patients remained weight stable, then you will find that your cohort will not have lost or gained weight. This finding derives from the very fact that you just eliminated any record where the patient's weight changed.

#3 poses even more of a problem for me. What happened to the 31 other patient records? How many lost weight and were eliminated by the criteria chosen? How many gained weight and were eliminated from the criteria?

So yes, I think reader's of Kolata's article will believe she is pointing to "definitive studies done in the 1950s and ’60s by Jules Hirsch of Rockefeller University and Rudolph Leibel of Columbia, which tested whether calories from different sources have different effects". When it appears (I say appears because we still don't know exactly which "studies" she's referring to) she's referring to:
a) a single study
b) that doesn't even test whether a calorie is a calorie
c) uses selective criteria to eliminate contrary evidence of it's conclusions

My conclusion is that Kolata's article mischaracterizes the study, and what's more the study itself was poorly designed to prove it's own findings.
posted by herda05 at 1:07 PM on October 23, 2007

You're right. I had missed the exclusion criteria of <> of feeding of two different formulas or the absence of feeding
periods of feeding periods >2 wks." That's the way I read that paragraph. The weight criteria was used before they reached the group of 47.

It does not invalidate the work that they used an existing data set, though. Multiple uses of data sets is quite common, and many studies are designed to collect more data then they strictly need in order to make the data set more useful. This may also be why the lack of weight fluctuation criteria was put in place, as the data set might not have supported an analysis that could do away with confounding variables.

More to the point, what the study measured was the energy needs of patients on two different diets. What they showed was that those patients did not have changes in energy requirements even in the face of significant changes in diet composition. If they had, they could not have stayed within the exclusion criteria.

(I also meant to point out before that of course what the study doesn't measure is anything to do with protein.)
posted by OmieWise at 1:32 PM on October 23, 2007

Whoa, that got screwed up in active preview. That should say that I missed the +- 1kg exclusion criteria, but while that might weaken it, that doesn't invalidate the study. They were measuring something else.

They do list the exclusion criteria for the 47-->13, lack of time on a diet and lack of two diets. It's written poorly, but it seems like that's what they mean.
posted by OmieWise at 1:53 PM on October 23, 2007

OmieWise, we seem to have diverged from the OP's question rather severely. I answered Malad's question with the best knowledge I have of the topic. This includes research of medical literature as well as practical application. I'd further recommend to Malad to be skeptical. It's good that he's questioning his trainer. He should question how each of us has come up with our answers for his question. He then needs to integrate that information and adjust his behavior (either exercise or diet). To progress to his goal, he'll need to measure whatever his choice was against real world results. What works and what keeps him healthy should dictate how he proceeds, not what someone wrote on AskMeFi.

As to that study, I'd be very hesitant to use it in support of anything. The data set was hand picked from an originally small cohort, rather then randomized across the entire cohort. The end product was an even smaller cohort. We don't even know the criteria for their original inclusion in the initial studies. It leaves to many uncontrolled variables to make the study worth very much.
posted by herda05 at 4:17 PM on October 23, 2007

In other words you support Kolata's overarching point, which is that selected studies are embraced or ignored by partisan observers based less on science than on preconceived ideas. I don't except Kolata or myself from this.
posted by OmieWise at 6:58 PM on October 23, 2007