Remembering Kevin Moore

Two-Year Study Shows Low-Carb Diet Effective For Weight Loss, Improves HDL, Cardiovascular Health Risks

Some of the harshest criticism about low-carbohydrate diets such as Atkins has been the supposed negative heart health implications due to elevations that take place in the cholesterol levels of dieters who restrict their carbs in favor of more fat and protein. Additionally, it is presumed that any weight loss that occurs on a high-fat, low-carb diet is quickly gained back making it a uniquely ineffective means for managing weight. Finally, bone health is supposed to suffer for people following a carbohydrate-restricted diet because the higher protein content allegedly promotes bone loss. However, all of these theories about low-carb diets have been summarily shot down by a brand new study funded by the National Institutes of Health (NIH) and published in the August 3, 2010 edition of the medical journal Annals of Internal Medicine.

Lead researcher Gary D. Foster, PhD, director of the Center for Obesity Research and Education and professor of Medicine and Public Health at Temple University, and his team of researchers noted that previous studies comparing low-fat diets with low-carb carb diets have failed to take into account the need for behavioral treatment as part of their dietary instruction which has resulted in poor weight loss outcomes. So they embarked on a 2-year randomized trial study of 307 participants placed on either a low-carbohydrate or low-fat diet combined with a “comprehensive lifestyle modification program” that took place in three academic medical centers, including the University of Colorado in Denver, Colorado, Washington University in St. Louis, Missouri, and the University of Pennsylvania in Philadelphia, Pennsylvania. The average age of the study participants was 45 years old with a mean body mass index (BMI) of 36.1. The study participants were split up into one of two diet categories:

Carbohydrate intake was limited to 20 grams daily with unrestricted consumption of fat and protein during the first 12 weeks of the study. Permissible carbohydrates were limited to mainly low-glycemic index vegetables. Participants were encouraged to eat 4-5 small meals every few hours and to use butter, mayonnaise and vegetable oils instead of margarine and they were discouraged from trying to “do a low-fat version of the program as it will disrupt weight loss.” At the end of the first 12 weeks, study participants were allowed to increase their carbohydrate consumption by 5 grams daily each week in the form of more vegetables, some fruit, and even whole grains and dairy products until their weight became stabilized. The principles outlined in the all-time #1 bestselling low-carb diet book Dr. Atkins’ New Diet Revolution were encouraged, but the study participants were never provided with a copy of the book. They were told to watch their carbohydrate intake primarily while urged to consume foods that are “rich in fat and protein” to satiety. The behavioral modification implemented with this group was “to limit carbohydrate intake.”

LOW-FAT DIET GROUP (154 participants)
Calories were limited to 1200-1500 daily for women and 1500-1800 daily for men with a fat/protein/carbohydrate ratio of 30/15/55. Study participants were strongly encouraged to keep their calorie intake reduced with a specific focus on cutting down on their fat consumption. The behavioral modification implemented with this group was “limiting overall energy intake.”

Participants in both groups received a 75-90 minute comprehensive, in-person group behavioral treatment led by a registered dietitian or psychologist with experience in weight control on a weekly basis for 20 weeks, then every other week for the next 20 weeks, and then ever other month for the rest of the study period where they were taught the SAFE method that includes Self-care, Adherence, Food records, and Exercise. Physical activity was identical for both groups in the form of walking beginning in week 4 with four sessions of 20 minutes each and then bumped up to four sessions of 50 minutes each by week 19. The study participants tracked their eating and exercise in a record book that was checked during their group sessions. A daily multivitamin was provided by the researchers to the study participants and encouraged for both groups.

What did Dr. Foster and his research team measure? Body weight was checked on each treatment visit where the participants wore light clothing and no shoes since this was the primary outcome they would be observing over the two-year period. However, additional health markers were checked out at 3, 6, 12, and 24 months, including:

Fasting blood samples were analyzed to look at HDL, triglycerides, VLDL, and LDL cholesterol. Precise measuring tools were used to obtain accurate lipid data.

After a 5-minute rest period, two blood pressure readings were taken separated by a 1-minute rest period. The average of the two readings was used to determine the blood pressure.

Ketone testing strips measured fasting ketone bodies in the urine and measured on a scale from negative to varying degrees of positive (trace, small, moderate, or large).

A checklist of 26 symptoms were looked at to determine if there was none, mild, moderate, or severe in the study participants. Documentation of symptoms was labeled “absent” or “present” with a further breakdown in the latter group. The researchers note that most symptoms were listed as none or mild.

Bone mineral density and body fat composition was used at baseline and then again at 6, 12, and 24 months. Sophisticated full-body scanning instruments were used to determine these numbers.

All of the participants in the study were contacted via phone, mail, and e-mail for follow-up even if they dropped out of the study. And attrition rates were similar among both study groups. In the LOW-FAT DIET GROUP, the percentage of study participants who withdrew completely or had frequent absences from the scheduled meetings was 6% at 3 months, 12% at 6 months, 25% at 12 months, and 32% at 24 months. As for the LOW-CARBOHYDRATE DIET GROUP, attrition was slightly higher but statistically the same with 9% at 3 months, 16% at 6 months, 26% at 12 months, and 42% at 24 months. Although the study was funded by NIH (something low-carb science leaders like Dr. Mary C. Vernon has been pushing for years), they had no role in the “design, conduct or reporting of the study.”

What were the final results that the researchers found? Since body weight was the primary marker they wanted to observe, Dr. Foster and his team found that both groups lost a total of 11% of their starting weight at 6 and 12 months. However, by the end of the two-year study, the collective weight loss among both groups was 7%. One interesting bit of information they provided was the fact that there were “no statically significant differences in weight loss at any time point between the low-carbohydrate and low-fat diet groups.” But they did observe a “strong trend for greater weight loss” in the LOW-CARBOHYDRATE DIET GROUP before carbohydrate intake was elevated by 5 grams per day each week. In other words, as carbs were reintroduced into the diet, the differences in weight loss among the two groups became less pronounced until they were statistically the same.

Not surprisingly, urinary ketones were higher in the LOW-CARBOHYDRATE DIET GROUP at 3 months and 6 months as compared with the LOW-FAT DIET GROUP. But then something interesting happened–after 6 months, ketones bodies equalized among both groups as carbohydrate intake was increased gradually among the LOW-CARBOHYDRATE DIET GROUP which was the intent of the study. Yet it makes you wonder what would have happened had they kept carbohydrate intake at a level that would have been ketogenic over a longer period of time than 3-6 months. This was not observed in Dr. Foster’s study.

Regarding blood pressure, they saw similar decreases in systolic blood pressure among both study groups. Yet there were “significantly greater” reductions in diastolic pressure in the LOW-CARBOHYDRATE DIET GROUP at 3 and 6 months with a “strong trend” at 24 months. This is an observation that was previously recorded by researchers in this Duke University study published in the January 25, 2010 issue of Archives of Internal Medicine. On blood lipids, this was where the differences between the two groups became more pronounced, especially early on in the study.

The widest gap among the changes in this health marker were seen at 6 months when the LOW-FAT DIET GROUP saw a “significantly greater decrease” in LDL cholesterol levels, although this statistical change was negated by the time it was measured again at 12 and 24 months. Triglycerides decreased more among the LOW-CARBOHYDRATE DIET GROUP at 12 and 24 months and then became even at 12 and 24 months (again, as carbohydrate intake rose above ketogenic levels). Decreases in the VLDL cholesterol levels were “significantly greater” in the LOW-CARBOHYDRATE DIET GROUP at 3, 6, and 12 months, but not at 24 months (again, look at the ramping up of carbs among the study participants in that group). HDL cholesterol, commonly known as the “good” cholesterol, was predictably “significantly greater” in the LOW-CARBOHYDRATE DIET GROUP at every level measured ostensibly from the high-fat content of the foods they were consuming. This positive change in HDL cholesterol was the most notable difference between the two groups the researchers saw over the two-year period.

However, another key finding was with the bone mineral density. There were “no differences” between the two groups which is tantamount considering all the rumors that low-carb diets lead to osteoporosis and bone loss (this is ironic considering there is published data in 2006 showing a low-fat, low-calorie diet leads to bone loss). But the bone density was the same in both groups and the importance of this cannot be overemphasized since this is the first study of its kind to show that the bone mineral density change with low-carb diets is comparable to that of a low-fat diet. Another myth has been busted wide open with the published results of Dr. Foster’s study! Additionally, there was “no differences” in lean mass and fat mass between the two groups.

Symptoms from the LOW-CARBOHYDRATE DIET GROUP seemed to be “significantly greater” with reports of bad breath, constipation, and dry mouth during the first six months of the study which all could have been avoided with some simple measures included in most books and support groups for people following a low-carb diet. Curiously enough, although there have been headline-grabbing lawsuits purporting that a high-fat, low-carb diet caused heart disease in the past, there were “no serious cardiovascular events” that happened during this study. Dr. Foster noted that consuming a low-carb diet is not the heart health risk most medical professionals once feared.

A low carbohydrate diet is associated with favorable changes in cardiovascular disease risk factors at two years.

This coincides with Dr. Ronald Krauss’ January 2010 study vindicating saturated fat as a contributor in heart disease. And once again, another myth about high-fat, low-carb living has fallen by the wayside.

The conclusion of the study zeroed in on two findings that Dr. Foster and his team of researchers considered most important. First, they contend that neither dietary fat nor carbohydrate intake were a factor in weight loss when they are combined with a comprehensive lifestyle intervention as was the case in this study. Second, since weight loss success over the period of the study was the same among both study groups, the only real lasting change that took place was in the HDL cholesterol of the low-carb dieters. They explained that the LOW-CARBOHYDRATE DIET GROUP “has greater beneficial long-term effects on HDL” than the LOW-FAT DIET GROUP did. This is a positive take-home message that validates previous research on low-carb diets.

Note that this study lasted two years which is considered a long-term research project. The complaint that low-carb nutrition has not been examined over an extended stretch can no longer be used by those who oppose carbohydrate-restriction (and yet again another myth bites the dust). Another important lesson from this study was the need for behavioral support in order to be successful. This is why follow-up with bariatric physicians or even a low-carb doctor for severely obese people is vital to attaining the desired results. The researchers readily admit that their purpose with this study was not to see the results of a long-term “Induction”-styled Atkins low-carb diet beyond 12 weeks. But the fact that even this low-potency low-carb diet did as well or better than the low-fat, low-calorie diet over a two-year period cannot be ignored. Just think if they had instituted a plan based on all the new science such as the runaway 2010 bestseller The New Atkins For A New You by high-fat, low-carb diet researchers Dr. Eric Westman, Dr. Jeff Volek, and Dr. Stephen Phinney what this improved, higher-potency low-carb diet could have done–no doubt there would have been upwards of 50-75 pounds of sustained weight loss annually with massive improvements in all the key health markers that were seen early on in the LOW-CARBOHYDRATE DIET GROUP. Now these findings will have to wait for the next low-carb study to come around for the next researcher to discover–although the trivial weight loss that has been shown in these studies can be discouraging to would-be low-carb dieters.

Dr. Foster gave low-carb diets the ultimate compliment after he analyzed the results of his study and offered up this final conclusion.

These long-term data suggest that a low-carbohydrate approach is a viable option for obesity treatment for obese adults.

That’s high praise indeed and interestingly enough there is already a physician-training program in place with CME accreditation teaching medical professionals from around the world the value of using carbohydrate-restricted diets with their metabolically-challenged patients. It’s from a group called Innovative Metabolic Solutions (IMS) led by Dr. Vernon, Dr. Westman, and New York Times science writer and bestselling author of Good Calories Bad Calories Gary Taubes. No doubt interest in such educational programs like this one offered by IMS will continue to grow as research from Dr. Foster and others continues to be published in major medical journals in the coming years.

If you appreciate the time and effort it took to put this quality low-carb diet research together over the past few years, then send Dr. Gary D. Foster an e-mail of thanks to gfoster@temple.edu. Let him hear your gratitude for letting the results of the science he found guide him in his research. He’s a rare breed and deserves to be recognized for his stellar ethics and procedures with the publication of this fabulous low-carb study!

  • Great study breakdown, Jimmy!

    • THANKS Mike! Took me six hours, but totally worth it. GREAT STUDY!!!

  • Mario

    Great review, Jimmy!

    One question. Did they report the carb percentage the low carb group where eating at the end of the study?

    • Nope! But we can safely assume it was above ketogenic levels (over 60g carbs daily) and much higher than they needed to produce the kind of weight loss experienced in the first three months when they were at 20g daily.

  • “… it is presumed that any weight loss that occurs on a high-fat, low-carb diet is quickly gained back making it a uniquely ineffective means for managing weight.”

    I always get a kick out of that, as you, I and your readers are well aware; You don’t go off of a low carb diet, we change our eating habits for life.

    • As I like to say, a PERMANENT and HEALTHY lifestyle change.

  • Tula

    Nice to see this being scientifically validated. I’ve always known this was the case, since my own cholesterol has not been negatively affected by low-carb eating. In fact, the last time it was tested, my HDL was so good that it was 19 points above the high end of the listed “normal” range. My LDL, though, was toward the low side of the normal range, so that proved enough to me 🙂

    • It really works to improve lipids, but that doesn’t follow the template that low-carb is unhealthy. This study is going to open doors for low-carb in the mainstream of health thought again.

  • Tron


    I read this study as well. I couldn’t find anything regarding the reporting of nutrient intake for the subjects, which I obviously think is critical. Gardner did a nice job of this in the JAMA article in 2007. Do you know if it was tucked away in a appendix somewhere (all I found was the “summary for patients” and the lifestyle modification program appendix)? Thanks

    • I got a copy of the entire paper, Tron, and there is no specific mention of the actual nutrient intake of the study participants.

  • You nailed it, Jimmy. Great post.

    • THANKS Keith! This is a HUGE study that needs to be trumpeted far and wide.

  • Tron

    Jimmy Moore wrote: “Nope!”

    To quote John McEnroe: You cannot be serious!

    Did the authors not find it relevant to record and report the intakes of subjects in a 2-year diet study that’s theoretically testing the differences between nutrient compositions of the diet (but again, they essentially ruined this by making one diet restricted in calories (low-fat) and the other ad libitum))?

    Two years of study and funding from the NIH and no report of intake. Not even a questionnaire? I know the subjects were keeping “food record booklets.” Did they hand them in at some point? Did the authors have access to them?


    • Well, I’m looking at the study right now and there is no mention of the breakdown of the diet at the end of the study. That’s why this whole business of raising carb intake by 5g/day each week was ridiculous. Just keep ’em at ketogenic levels and all of the weight loss and blood markers would have been SIGNIFICANTLY GREATER improvements with the low-carb group. DUH!

  • Wow, Jimmy, quite a write up – I’m impressed that you got something that concise & insightful in only 6 hours! 🙂

    Nice job, thank you!

    • THANKS Sarah! I wanted to share everything I could about this important study so people could learn the truths that are contained within it.

  • Jimmy is right about the lack of participants’ nutrient and caloric intake. That data is notoriously inaccurate, anyway.

    Or perhaps this group will publish a series of articles based on this one study, including the nutrient intake estimates. You know what they say, “Publish or perish.”

    This study indicates that successful long-term weight loss requires counting something, in this case either carb grams or calories. I bet most people find it easier to count carbs. [I remember diets that counted fat grams – are they still out there?]

    The intensive behavioral program here would be difficult to implement on a large scale. I suspect it had a lot to do with the success of both programs. It would be great if on-line support groups had the same effect.

    You have to wonder if the high HDL cholesterol (thought by many to be heart-protective) in the low-carbers would yield lower rates of stroke and heart attack after one or two decades of compliance.


  • Great article! Very informative! Thank you.

  • Mike Schudel

    One thing I think that was missed in your fantastic write-up of this study and in the study’s conclusions is that the low-carb group presumably was not watching the number of calories of their food, just the amount of carbohydrate. The fact that the low carb was equally effective as the low-fat (and calorie restricted) diet would indicate one of two things are true:

    1. If the low carber’s on average ate more calories than the low-fat calorie restricted eaters, then the low-carb diet must the metabolic advantage that Atkins talked about or, as Gary Taubes believes, the calories don’t really count.

    2. If the low carbers ate as many calories or fewer than the low fat calorie restricted eaters, they must have done so voluntarily which would configm Atkins and Taubes believe that the carbs trigger the hunger and restricting them from the diet causes the behavior change without the need for ‘will-power’.

    Either way, I think this is an extremely important outcome that came out of this and I would argue that either way, the fact that the outcome of the low-carb diet was equivalent to the low fat calorie restricted diet but that the low carbers did not have to fight hunger or count calories (whether they ate more or at as much or less voluntarily) proves that the low-carb diet has a definitive advantage over the low-fat calorie restricted diet.

    Also, I strongly agree with Jimmy that if the low carber’s stayed at the Induction Level (or at least at an on-going weight loss level)until the desired weight was lost, I suspect the result would have been far more favorable for the low carb group.

    Thanks again for the great write-up!!

    • Mike, I agree. And Taubes made this point in his recent Q&A Session I posted on my blog following his lecture to the IMS participants.

  • Mark Hoelter

    I am not a regular reader of your blog, except through my wife’s abridgments, and she is both a regular reader and fan, not only of your personal story but also of your objectivity as well as respect toward others, even others of different opinions.
    We’ve both shifted to a South Beach, and then to a modified Atkins, low-carb diet, and are about to begin a slow-burn resistance training. In telling our prior personal trainer of our moves, in a gracious email in response, he attached the following study (which I’ve reduced to a tinyurl): http://tinyurl.com/2bu7w3m.
    Wonder if you’ve seen this study, which appears to be more recent (and if you’ve already commented on it, just tell me where to look), and how it strikes you. I don’t think of NIH as automatically objective and unbiased, but my knowledge base is pretty shallow here. Nor have I studied it closely.

    • THANKS Mark! The link you provided was broken and did not point to the study you are referring to. But just because it’s NIH-funded doesn’t necessarily make it valid. E-mail me at livinlowcarbman@charter.net the entire URL and I’ll be happy to take a look.