Dietary fat vs. carb restriction in obese people (new research!)

I haven’t finished reading this yet but it looks interesting enough to post: Calorie for Calorie, Dietary Fat Restriction Results in More Body Fat Loss than Carbohydrate Restriction in People with Obesity.

Dietary carbohydrate restriction has been purported to cause endocrine adaptations that promote body fat loss more than dietary fat restriction. We selectively restricted dietary carbohydrate versus fat for 6 days following a 5-day baseline diet in 19 adults with obesity confined to a metabolic ward where they exercised daily. Subjects received both isocaloric diets in random order during each of two inpatient stays. Body fat loss was calculated as the difference between daily fat intake and net fat oxidation measured while residing in a metabolic chamber. Whereas carbohydrate restriction led to sustained increases in fat oxidation and loss of 53 ± 6 g/day of body fat, fat oxidation was unchanged by fat restriction, leading to 89 ± 6 g/day of fat loss, and was significantly greater than carbohydrate restriction (p = 0.002). Mathematical model simulations agreed with these data, but predicted that the body acts to minimize body fat differences with prolonged isocaloric diets varying in carbohydrate and fat.

I would try to make a witty observation at this point but I’ve only read the summary so far.


19 adults is not a good representativ basis for a medical study to draw any conclusions


The sample was small because they locked them in a room and controlled every part of their life for a week at a time. The study was limited by small sample size, but it was very strictly controlled and they had each group do both the low-carb and the low-fat diets and got the same reaction to the diets across the population.

I’m not saying we should reorganize the world around a study of 19 participants, but this study has merit even compared to a self reported study of a thousand people.


It’s suggestive. But, given the small number of participants, I would be interested in the individual results of each of the 19 participants. Did the results march lock step for each of the participants or did some participants’ results have the opposite result. By having two trials (the original group, and then the group with the opposite treatment) you essentially have two results. I.e., we’re in coin toss territory from a probability perspective.


Thanks or posting. From the summary the methodology looks good; actually one of the relative few forced isocaloric studies. The summarized results seem to be opposite of the headline, but after re-reading it 10 times, I think it’s just written very poorly. (Maybe there was a word-count limit.)

That’s terribly written! But it does support low-fat being superior for fat loss in the the very short term. And the effect measured is very specific. It should say:

Carbohydrate restriction led to sustained increases in fat oxidation and loss of 53 ± 6 g/day of body fat. Whereas fat restriction led to unchanged fat oxidation yet overall fat loss of 89 ± 6 g/day, a significantly greater amount.

The thing to keep in mind, regardless of the findings, is that these are diseased people and the study designed to measure effects on their disease state. The Soylent being food, not medicine, the question is–for me, at least–what the healthiest macronutrient ratio is for healthy people. In other words: in a generally healthy person with healthy weight, what macronutrient amounts result in superior longevity and healthspan?

If they could just find healthy people who would agree to be locked in that same lab for 20 years…

EDIT: somehow I’ve replied to myself.

1 Like

It is if the study is well designed and measuring specific enough results. The summary suggests so. But I haven’t read the full text.

Agreed. And I would go further to say this has at least as much merit for the narrowly measured fat loss effects, based on the tightly controlled food intake and measurements of specific fat loss parameters.

I suspect that self reporting studies are actually measuring hunger effects of fat vs carbs to a significant degree, no matter what the researchers are actually focusing on. I wish they had had a hunger instrument in this study.

That’s not true. (At least not what I think you’re saying.) Each group crossed over and did the opposite diet following their first, randomly assigned diet.

Here is the full text of the study for those interested in a closer read.

Although we don’t see the results of each individual my interpretation of Figure 3 chart A is that the person who lost the most fat on the low fat diet lost less fat than the person who lost the least on the low carb diet.

1 Like

How do you infer that? I’m only seeing (in those charts) what I assume to be averages.

Significant findings I see in the charts (and paper):

  • higher fat loss from low-fat (“cumulative fat change” and “Δ fat balance”)
  • results of the study largely fit the models (I’m assuming these were pre-study models)
  • body weight is less in low-carb group, explained by higher water loss in the paper

Figure G is confusing. It shows a higher change in fat mass with low-carb. But the accompanying text indicates that it’s their model projection, writing “the model predicted that the RF diet would lead to approximately 3 kg more body fat loss after 6 months.” Unless I’m misunderstanding how they’re using “Δ Fat Loss,” this appears to show the opposite.

I don’t think it’s terribly written, within the technical bounds of a medical journal. If it were this way in a lay press article, I’d agree with you.

That is a key observation, here. The intervention period was only six days. We do know that the body adjusts very rapidly to changes in carb intake… and that fat metabolism doesn’t adjust for days, it sorta kicks along, unless you constantly feed on carbs to actively suppress it all day long. But, since these folks were in a metabolic ward, they were unable to consume extra between meals, and if the higher carb diet (and higher metabolism after a carb spike) led them to be a little hungrier after, they didn’t to appease that hunger with snacks.

It does make sense that in an obese person - who has a lot of fat available even when they don’t eat it - that the fat metabolism can just keep chugging along for a long time.

It’s VERY expensive to do a controlled study where the subjects are given all calorie-counted meals, and it’s VERY expensive to do a controlled study in a metabolic ward where calorie expenditures and type of calorie expenditure is measure. Each of these people had 22 days of exactly-measured meals provided, as well as metabolic ward time.

Having people self-report how much they ate, and just weighing them after some time, is profoundly cheaper - but also so sloppy that you need enormous sample sizes to get a statistically significant result - and then, because you don’t REALLY know how much they consumed, your statistically significant results may be accurately measuring how people bias their food estimates on low-fat versus low-carb, rather than measuring their diet results on low-fat versus low-carb.

That being said, this is a VERY exacting way to do the research - and because the numbers are very accurate, you can get statistically significant results with smaller samples, as long as the response in each person is fairly consisten. Which it was. You can tell by the details in the numbers and the tight error bars - and the level of significance they’re getting (p<.001 and p<.005) are just excellent.

Actually, no - every person did all of it. They came in, the were fed a standard diet for 5 days while monitored, and then they were fed a reduced-carb diet for 6 days while monitored. Then they left for a couple weeks, came back, and did 5 days of starndard diet followed by 6 days of reduced-fat. The only difference between the individuals was that half of them did the reduced-fat test first, left for a few weeks, and then came back to do the reduced-carb test. They study shows it this way:

They don’t offer the full dataset in the paper - even in the extended PDF publication. If you want to contact the study authors, they may share the data.

However, you don’t need it. The statistical analysis answers your question:

If you look at the dots for day 4 and day 6 (blue diamonds for Reduced Carb and red squares for Reduced Fat), you’ll see little lines going up and down indicating a range. The group had results averaging at the blue dot, plus or minus… the bars show you the range of plus or minus, expressed as standard deviations. The tight scope tells you the standard deviation withing the group was very small… 68% of the people in the group fall within those bars (that’s plus or minus one Standard Deviation.) That means that for the most people even the worst result on the Reduced Fat diet is better than the best result on the Reduced Carb diet.

It’s pretty close to “lock-step.” This is a very rigorous study with a very strong result.

There’s still individual variation, to be sure - some people really are metabolically different. But what I take away from this is that most of us are very, very metabolically similar. The big differences in our dieting results come more from our tremendous differences in tolerance for particular foods, particular diet types, tolerance for hunger, and especially how much we misrepresent to ourselves how much we can or should eat, how much we did eat, etc… and the tiny biological differences in our metabolism are almost certainly swamped by all those psychological choices we make.


THANK YOU for pointing this out. I hadn’t seen this yet, and it is a really excellent study which has long been on my “I wish someone would do this study” list!

1 Like

Several days of the RF diet led to a steady fat imbalance of 840 ± 60 kcal/day, or equivalently 89 ± 6 g/day of body fat loss (Figure 3A), which was significantly greater than the steady rate of body fat loss of 500 ± 60 kcal/day, or 53 ± 6 g/day, achieved during the RC diet (p = 0.0002) (Figure 3A).

I’m interpreting the ± as a range of results, not a statistical average.

Taking another look at that one, I agree with you. I wonder if it’s mislabeled.

The practice is to report the statistical average and the +/- variance in terms of S.D. (standard deviation.) So if they say:

body fat loss of 53 ± 6 g/day

They’re saying the statistical average lost was 53 grams per day, and that the statistical S.D. was 6 grams. So 53 ± 6 is the range 47 to 59 grams. Statistically, 68% of their study population fell into the range 47 to 59 grams of fat lost. (68.2%, based on the definition of one Standard Deviation.)

On the low fat diet, it was

89 ± 6 g/day of body fat loss

So the general range was 83 to 95 grams of fat lost, and 68% of their population fell into that range.

They also said that the difference between the two results was very statistically significant:

(p = 0.0002)

A p value of .05 or less is statistically significant, generally speaking. Usually, the results would be closer together than 53 and 89… and each would be spread out wider than +/- 6, so you would need a huge sample size to get a difference that is significant and gets the p value below .05. Here, because the difference between averages is much bigger than the variation ranges, you have a very, very significant result of p = .0002. That means that if you ran this test with random inputs, you’d only have a 0.02% chance of getting a result like this - and we usually call a finding significant if there’s anything less than a 5% chance of a random input leading to the same result!

This happens because the SD of +/- 6 is so tight. 68% of the population falls with one SD, and 95.4% fall with two SD.

So 95% of the people on reduced carb lost 53 +/- 12 grams of fat - the range is 41 to 65 grams of fat lost.
And 95% of the people on reduced fat lost 89 +/- 12 grams of fat - the range is 77 to 101 grams of fat lost.

This is what I meant when I said early that for most people (95%), even a big loser on low-carb (65 grams) loses less than the lowest loser on low-fat (77 grams.) The distance between the 53 and the 89 averages, combined with the very tight SD for each of those figures, makes for a significant finding even with a fairly small sample size.

(99.9% of the population falls within three SD of the average.)

Yeah, the text in the paper is contrary to the labeling on the figure:

Figure 3G illustrates the mathematical model simulations of 6 months of selective isocaloric restriction of dietary fat versus carbohydrate at the level implemented during the inpatient study. The model predicted that the RF diet would lead to approximately 3 kg more body fat loss after 6 months of perfect adherence to the isocaloric diets.

So it’s an error. This is, however, the online pre-press publication proof:

Publication stage:
In Press Corrected Proof

There will be correction between this and the published version, and perhaps more correction after people send in their criticism. This is also a good reminder that although this appears to be a very rigorous and well-done study with a very clear result, it’s still just one study. It has to be judged in balance with other rigorous studies in the same area, which may or may not agree with this finding.


The study being very expensive is irrelevant for a conclusion of the results, And I’m not comparing to self reported studies.

This is a study in one of the most complex and fluctuating scientific fields, where you have to count for thousands of factors, mainly the reason why small samples don’t say anything conclusive other then for those 19 persons and people having a very similar situation, ‘excluding’ the rest of the populous in a meaningful scientific way.

only Conclusion i can see this suport, is more and similar studies following it’s guidelines on a bigger scale, as ti is showing some interesting data.

While it is an interesting study, and being isocaloric was necessary here to get a real direct comparison… In general I don’t find isocaloric studies to be all that helpful for people in the real world trying to lose weight. Did one group feel any more or less satiated than the other? One of the benefits of the low carb diet (supposedly) is that it is more satiating and thus results in a naturally lower caloric intake.

I think what will be far more interesting to see is how effective Soylent 2.0 with 47% fat is at helping those who are trying to lose weight as well as the general health of the thousands of users after being on such a (relatively) high fat diet for a while. Forget n=19, let’s try n=10,000… many of those results we’ll see right here on discourse.

1 Like

Right this is what I was thinking. As far as nutrition science though it is an important result that the weight loss is significantly different, I wonder what the underlying mechanism is and if it would have other relevance on health aside from weight loss.

All of their diets had 3x the sodium of Soylent (!)
And their “low carb” diet was 29% carb, 50% fat and 21% protein… Pretty close to Soylent 2.0 with 33/47/20.

What I’m having a hard time understanding though… is they admit the low carb diet increased fat oxidation (RQ -= .05) versus the low fat (RQ -= .004) yet people still lost more fat? The implication that even though fat oxidation was increased with low carb, the mere presence of dietary fat reduced body fat loss relative to the low fat diet? The RQ data was 24hrs, I thought they mentioned a plateau although I don’t recall seeing the data… Most of what I’ve read implies more than a few days to really mobilize fat oxidation… Although one thing I’ve never found (and am still searching for) is the delta RQ as a function of carb % (or fat %) in the diet (after a few weeks at that diet). It would be interesting to see the RQ during the same fixed exercise for someone who had been on a two week Soylent 1.3 diet vs a Soylent 1.4 diet vs Soylent 2.0…

Caveat: I am not an expert and I’m not sure I completely understand this.

Body fat loss was calculated as the difference between daily fat intake and net fat oxidation measured while residing in a metabolic chamber.

The reduced fat diet (obviously) had less dietary fat intake. There’s a handy graphic at the top visualizing this:

Fat oxidation was increased with the low carb diet, but carb oxidation decreased even more:

Only the RC diet led to significant sustained adaptations of carbohydrate and fat metabolism. At the end of the RC diet period, net fat oxidation increased by 463 ± 63 kcal/day (p < 0.0001) (Figure 2G) and net carbohydrate oxidation decreased by 595 ± 57 kcal/day (p < 0.0001) (Figure 2H). In contrast, only the first day of the RF diet led to a significant reduction in net fat oxidation by 96 ± 64 kcal/day (p = 0.01) (Figure 2G) and an increase in net carbohydrate oxidation of 147 ± 49 kcal/day (p = 0.01) (Figure 2H) compared to baseline.


[Figure 2] (G) Net fat oxidation increased substantially during the RC diet and reached a plateau after several days, whereas the RF diet appeared to have little effect.

Our relatively short-term experimental study has obvious limitations in its ability to translate to fat mass changes over prolonged durations. It could be argued that perhaps the fat balance and body fat changes would converge with continuation of the diets over the subsequent weeks. However, this would require that the net fat oxidation rate somehow increase above the observed plateau with the RC diet, and/or the RF diet would have to result in a swifter decrease in fat oxidation. Neither of these possibilities was apparent in the data and did not occur in the mathematical model simulations of prolonged diet periods. If such a convergence in body fat loss were to occur with prolonged RC and RF diets, the physiological mechanism is unclear.

Also interesting:

The model also indicated that the RC diet would lead to increased net protein oxidation compared to the RF diet (Figure 2I), a trend that was apparent in the 24-hr urinary nitrogen data (Table 3).

While the simulated weight loss with the RC diet was close to the observed value, the RF diet led to substantially more weight loss than was predicted by the model. This was likely due to body water losses that took place via mechanisms outside the scope of the current model (see the Supplemental Information for a full description of the model).

Table 4 presents the baseline overnight-fasted plasma measurements along with the changes in response to the RC and RF diets. Both RC and RF diets appeared to significantly decrease plasma C-peptide, insulin, insulin resistance, leptin, adiponectin, total cholesterol, and HDL. Plasma HDL and total cholesterol decreased to a greater extent with the RF diet, and LDL decreased only with the RF diet. Plasma TG decreased only with the RC diet. Plasma b-hydroxybutyrate and ghrelin increased only with the RC diet. Plasma GIP increased with the RC diet and decreased with the RF diet.

Since it might be possible that different ratios of carbohydrate and fat would lead to different results, we simulated body weight and fat mass changes after 6 months of eating a variety of 30% reduced-energy isocaloric diets varying in carbohydrate and fat, with protein fixed at baseline levels as illustrated in Figure 3H. The model predicted that weight loss increased with decreasing carbohydrate. However, body fat loss was relatively insensitive to isocaloric substitutions of dietary fat and carbohydrate, suggesting that the body acts to minimize differences in fat loss when the diet calories and protein are held constant. In fact, the experimental RC and RF diets resulted in close to the maximum predicted differences in body fat loss. In other words, the modest differences in body fat loss achieved by the diets used in our experiment are probably greater than would be observed with other ratios of carbohydrate and fat.

Translation of our results to real-world weight-loss diets for treatment of obesity is limited since the experimental design and model simulations relied on strict control of food intake, which is unrealistic in free-living individuals. While our results suggest that the experimental reduced-fat diet was more effective at inducing body fat loss than the reduced-carbohydrate diet, diet adherence was strictly enforced. We did not address whether it would be easier to adhere to a reduced-fat or a reduced-carbohydrate diet under free-living conditions. Since diet adherence is likely the most important determinant of body fat loss, we suspect that previously observed differences in weight loss and body fat change during outpatient diet interventions (Foster et al., 2010; Gardner et al., 2007; Shai et al., 2008) were primarily due to differences in overall calorie intake rather than any metabolic advantage of a low-carbohydrate diet.