Low-Carb And Low-Fat Diets Battle To A Draw

A new study comparing a low-carbohydrate diet with a low-fat diet found no important differences in weight loss or other important outcomes between the two diets. Some experts believe the result shows that the  debate over the relative worth of these different diets has been overblown and confirms the view that calories count. Others say that the issue is not yet settled and that the study did not test the effectiveness of a true low-carb diet.

The Diet Intervention Examining The Factors Interacting with Treatment Success (DIETFITS) trial, published today in JAMA, randomized 609 overweight adults to follow a low-fat diet or a low-carb diet. There were no significant differences between the groups at the start of the study in diet or other important factors and both groups reported a calorie reduction of 500-600 calories per day.

After one year there was no significant difference between the groups in weight loss (5.3 kg in the low-fat group and 6 kg in the low-carb group). Dietary intake, as self reported, differed significantly between the two groups during the study. Carbohydrate consumption dropped from 241 g to 212 g in the low-fat group and from 246 g to 132 g in the low-carb group; fat consumption dropped from 87 g to 57 in the low-fat group and from 92 to 86 g in the low-carb group; and protein consumption dropped from 92 g to 84 g in the low-fat group and stayed the same at 93 g in the low-carb group. As expected there were significant differences in lipids between the two diets. LDL cholesterol was significantly lowered only in the low-fat diet group while HDL increased only in the low-carb diet group. Triglycerides were lower in both groups, but the reduction in the low-carb group was much larger (-28 mg versus -10 mg).

It is worth noting that the study retained a relatively high percentage of participants, with 4 out of 5 participants in each group remaining in the study for the full year.

It should also be noted that the study also tested whether a genetic test or a metabolic test could predict weight loss success with either diet. This portion of the study was entirely neutral, suggesting yet again that “precision medicine” is not yet ready for prime time.

A Closer Look at the Two Diets

Christopher Gardner

Results of the study seem likely to add fuel to the debate about what constitutes a healthy diet and what exactly defines a true low-carb diet. Low-carb proponent Gary Taubes pointed out in an email that the study did not put people on a true low-carb ketogenic diet, which usually aims for dramatically lower carb levels ( around 30 g day or less) than those achieved in this study. But the lead  author of the study, Christopher Gardner (Stanford University), said that that was intentional. “We did not have a precise set goal for the participants,” he said in an email. Participants were encouraged to follow their assigned diets and to go as low (in fat or in carbs) as they could.”

“We asked the low-carb group to try to get to 20 grams/day over the course of the first 8 weeks (and most did!!). We explained that this number had been selected somewhat arbitrarily as simply something very low to start with as an ANCHOR. When they got there, we asked them if they thought that it could be feasible to do that level for the rest of their lives. If not, then we explained it wouldn’t work, because if they went “off” that diet, the weight would come back on. We wanted them to find a level that was: 1) As low as they could go, and 2) A level that could conceivably become an “eating pattern” (not a DIET) that they could follow forever more.”

“The answer we got was that, in our study, with our population, cutting their carbs from ~44% of calories (baseline) to 30% (at 12 months), and from 246 g/day (baseline) to 132 g/day (at 12 months), was the AVERAGE that our real people could do in the real world,” Gardner summarized.

Gardner agrees with Taubes that it is certainly possible to study diets that are much lower in carbs. “Another type of study is one where food is purchased and made for participants, and those participants can be observed or even confined to have more control over the diet. In those very rigorous settings, lower carb is certainly possible,” he said. But it is unclear whether it would be possible for a large population to follow such a diet, it should be noted.

In at least one respect the study could be considered a victory for the low carb movement. Taubes notes that “the advice for both low-carb and low-fat interventions was to avoid sugar, flour and processed foods, so in that sense both diets were carbohydrate restricted… Gardner refers to this in the paper as ‘healthy low fat’ vs ‘healthy low carb’ and then defines healthy as absent sugar and highly refined grains and including green vegetables.”

Taubes speculates “that the weight loss may have been similar not because any diet works if you stick with it and cut calories (one possible interpretation) but because of what these diets had in common — avoid sugar, refined grains, processed foods. Whether the low-carb arm would have done even better had Gardner kept their carbohydrates low is something this study can’t say. (And Ornish [low-fat diet proponent] would probably say the same thing about fat consumption.)”

Darrel Francis (Imperial College, UK) thinks the study helps solve “3 things I always wanted to know about dietary intervention in cardiovascular prevention.”

“First, does eating the same number of calories of (say) low-carb high-fat diet, change your metabolism so that you burn up more calories? Second, outside the laboratory environment where calories are kept fixed, do such diets help you eat less calories? Third, do they reduce cardiovascular event rates?

“Dozens of carefully controlled experiments (meta-analysed by Kevin Hall) have demolished the first question: if there is a difference in energy burning rate, it is the equivalent to half an Oreo per day.

“So if there is no magic metabolic energy-burning effect, maybe LCHF helps you limit your calorie intake – the second hypothesis. This new trial destroys that thought too. People lose just as much weight on low-carb as low-fat diets, over a period of a year. ~5kg is worth having, but you can do it with either type of diet.

“Finally, the third question. While the study was not designed to measure events, it presents fascinating information on changes in risk markers. We all want to have lower glucose and insulin, and lower LDL, if we can easily manage it, don’t we?

“The low FAT diet managed to lower glucose and insulin just as much as the low carb diet! All that shuffling in the shadows with ketone breath was for naught.

“And the low CARB diet raised LDL by a little more than the low FAT diet. In my mind, there is no longer a reason to favour recommending a low-carb high-fat diet, unless this is a specific preference of the patient. No better at getting weight off or lowering glucose, and it worsens blood lipids. Goodbye and good riddance.”


  1. ‘Deceptive’. Would be the genteel summation.
    Eliminating ‘sugars’ and ruthless reduction of ‘grains’ are foundation stones of LCHF, yet we find these inserted into BOTH arms…
    The unmentioned Confounder is that most people on a ‘diet’ or Trial will be more rigorous in their choices of snacks and other unrecorded intakes. – The extra paperwork and subsequent ‘guilt’ another disincentive and more distance from their normal eating pattern and hence a “free” health-improver. !

  2. James DownUnder says

    Seriously, ANY move away from the current modern eating patterns can only be beneficial. – British Advised Diet (BAD), and the notorious SAD – Standard American Diet.
    – And the level of carbs allowed is not really a “LOW” carb level.
    I see that fruit is not mentioned, and seeing that modern-day fruits are bred – by selection and GM’d – for very high sweetness, fructose levels are only controlled by the elimination of Processed ‘food’.
    Sorry, can’t take this one too seriously, except to say the slightly better weight (fat? lean muscle?) validates the LCHF approach, – despite what I believe to be RDR. / ‘rigging-for-desired-result’.

  3. ‘Taubes notes that “the advice for both low-carb and low-fat interventions was to avoid sugar, flour and processed foods, so in that sense both diets were carbohydrate restricted”‘: he’s caught them out. Which also makes a monkey of Francis’s pompous statement.

    Added to which let me pour derision on “While the study was not designed to measure events, it presents fascinating information on changes in risk markers”. By risk markers I assume he means any variable that is, or is assumed to be, a positive correlate with something deadly. But I give not a hoot for his risk markers: it’s the attendance of the Grim Reaper I want to know about. I don’t even much care about the weight of the patients: tell me about their death rates. The trial seems to have lasted a year. How long was the follow up?

  4. Frank Hummer says

    I agree with Janet’s and James DownUnder’s comments. I think the study does more to support LCHF than the headlines will likely reflect.

  5. Isn’t Chris Gardner a vegan? He’s at least a vegetarian. I admit to being impressed by his ATOZ study where he actually reported the results which went against his beliefs.

    IMO this time he decided not to make the same mistake twice.

    Anecdotally we routinely see large increases in HDL while trigs go through the floor on low carb diets, not to mention blood glucose effectively normalising and insulin (and insulin resistance) tumbling.

    When this fails to happen in a trial this is evidence that it has been fixed. As here. Again.

  6. Carol Vassar, MD says

    The real accomplishment of this weightloss study is the retention. 80% still enrolled at the end of 1 year is quite good. Was there a pretrial phase in which people could drop out? How was the weight loss at 2 years? 3 years? That is the important goal.

  7. dearieme says

    “How was the weight loss at 2 years? 3 years? That is the important goal.” Is it? Don’t you think that mortality and morbidity are more important?

Speak Your Mind