A new study called “Effect of meal frequency on glucose and insulin excursions over the course of a day” questions the all too popular recommendation of eating every so often to keep blood sugar in check. It actually shows that such advice can be counterproductive, as high meal frequency leads to higher blood sugar levels compared to low meal frequency.

Yes, that’s right – eating every 2-3rd hour to manage blood sugar is nonsense and a myth that’s just about to die.

The World is Upside Down

It’s funny how mainstream health advice in regards to diet and meal frequency gets turned upside down by new and more accurate research. In 2010 we learned that three meals is better for appetite control.

We also saw more hard evidence for the fact that a higher meal frequency does not “stoke your metabolism” or lead to better fat loss. Fortunately, some mainstream media outlets are starting to wake up and has enough sense to inform people – as we saw when the New York Times debunked the myth about meal frequency and fat loss. Sad thing is that they’re more than a decade late. That’s OK though. Hell, even I was late to finding out but I’ve been talking about this nonsense since 2005.

We also saw the New York Times cover the benefits of fasted training. Again, the NYT is late to the party – I summarized that study back in September.

(Furthermore, I’ve been training fasted, and have been using fasted training as part of diet regimens for myself and my clients since around 2006. Check out the “Fasted Training”-tag.)

We should be thankful that a big and influential newspaper like the NYT is relaying this information to its large audience. Indeed, as far as mainstream media outlets go you can say that they’re “on the cutting edge” as far as new diet research is concerned…

…But if you want to stay on the true cutting edge, you better have Leangains bookmarked 😀 Or better yet, subscribe in a reader or by email.

I remember way back in the days when I first started talking and writing about meal frequency, fasting and fasted training on blogs and forums. Everyone thought I was crazy. Guess who’s laughing now. But I digress; back to the topic.

The Meal Frequency Myth

As you have probably heard and read many times over – in the form of weight loss advices given by health enthusiasts and dietitians alike – eating small meals every second to third hour is supposed to keep your blood sugar in check. That’s on top of other benefits that include boosting metabolism, improving appetite control, preventing muscle catabolism, and so forth.

If you are a regular reader of this site, you know that this is utter nonsense – and even in complete opposition to what actually happens. I thoroughly debunked these myths in “Top Ten Fasting Myths Debunked” (a must read if you haven’t read it yet).

Blood Glucose and Health

Maintaining blood sugar within a healthy range is important, as higher blood sugar levels may predispose people to insulin resistance and glucose intolerance. That’s on top of other diseases such as atherosclerosis and Alzheimer’s. Furthermore, it as long been hypothesized that high blood sugar (blood glucose) may accelerate the aging process.

It should be noted that the aging-hypothesis, and the link between blood glucose and Alzheimer’s, is based on animal and cell culture studies. As you can understand, it would be hard to establish a definitive link between blood glucose, aging and other diseases in free-living humans.

That said, it is not far fetched to assume that our bodies haven’t adapted to the modern diet with its high calorie intakes and highly refined carb sources. The abundance of junk food and highly concentrated carb sources provides endless opportunities for spiking blood glucose to heights that we are ill equipped to deal with. That there will be a backlash for those who continuously maintain higher blood glucose may not be a wild theory.

In summary, maintaining blood sugar within a healthy range is very important for individuals with poor glucose tolerance and insulin sensitivity (“prediabetes”). However, it might also be of interest for just about anyone who wants to live a longer and disease-free life.

Blood Glucose and Meal Frequency

While debunking myths about meal frequency and metabolism is easy, the scientific literature on meal frequency and blood glucose (BG) is harder to explore and reach a conclusion on. For one thing, there is the issue of taking blood samples at the right time point in order to make a fair assessment on the result.

Let’s say we want to track average BG levels during a day of either 3 or 6 meals. We split meals equally in terms of calories (i.e. 3 x 800 kcal vs 6 x 400 kcal) and measure BG at various time points. The first time point is one hour after the first meal. Needless to say, the first reading is going to show a higher value after the 800-kcal meal (3-meal group).

On the other hand, the 3-meal group will have a lower BG reading in the third and fourth hour of the experiment – just as the 6-meal group is eating or finishing the second meal.

The above is an example but illustrates the problem. Proper timing of sampling is a huge confounder – and this has been handled poorly in earlier studies on meal frequency, BG, and insulin. In order to make a fair assessment of the results, researchers need perfect timing in relation to meals or draw several blood samples throughout the day. Otherwise, the results will be highly misleading.

Another monumental confounder has been using different nutrient compositions of meals. Not standardizing calorie and macronutrient composition in the above mentioned context makes any conclusion drawn from results worthless. Each nutrient has an independent effect on BG and insulin. Carbs raise BG and insulin the most, protein much less so, and fat the least.

Portion of cheesecake on a plate
Feast your eyes upon this beauty. I ate a lot of cheesecake this Christmas. The events that transpired will not go unnoticed and I shall tell you more about it soon. Keep in mind that I don’t recommend this practice for optimal blood glucose control. (You might wonder what a picture of a cheesecake slice is doing in the middle of a diet study review. First of all, your brain needed a break from the science-talk. Second of all, cheesecake is a hell of a lot better than lame graphs, a bodybuilder holding a dumbbell, or some chick doing yoga.)

The New Study: Greater Accuracy

The new study seeks to improve on the lacking methodology used in past studies. From the paper:

In contrast to previous research, this study used frequent blood sampling to track glucose and insulin concentrations to three and six subsequent nutrient ingestions.

OK, so let’s look at how the study was conducted, what the results showed, and what we can take away from it all.


The participants arrived to the laboratory fasted, after which baseline blood samples were taken. On three separate occasion, each participant was then fed the following 1500-kcal diets:

  • 6 CHO: 65% carbs, 15% protein, 20% fat, split 250 kcal x 6.
  • 3 CHO: 65% carbs, 15% protein, 20% fat, split 500 kcal x 3.
  • 6 PRO: 35% carbs, 45% protein, 20% fat, split 250 kcal x 6.

Meals were taken in the form of liquids; carbs in the form of sucrose and corn syrup, protein in the form of soy protein. Fat came with the protein supplement. Certainly not “ideal” but liquid meals are standard in these types of experiments.

Meals were eaten every second hour starting at 7 A.M. for the 6-meal groups (6 CHO and 6 PRO) and every fourth hour starting at 7 A.M. for the 3-meal experiment (3 CHO). Blood samples were drawn every 15th minute during the study period (7 A.M. – 7 P.M). The results were added together and values for BG and insulin were then calculated to establish averages for each diet.


Baseline (fasted) glucose and insulin values were similar across the three study days. Let’s look at the average BG values for each diet-experiment.

  • 6 CHO: 710.0 +-251.0 mmol/L*min
  • 3 CHO: 522.7 +-99.3 mmol/L*min
  • 6 PRO: 442 +- 121.0 mmol/L*min

The 6 CHO-experiment exhibited significantly higher BG values than the other groups. Despite identical carb and calorie-intakes, those who ate 6 meals had 30% higher blood sugar values than those who ate 3 meals. That’s a rather striking difference considering the energy- and nutrient-matched condition.

The difference between 6 CHO compared to the high-protein experiment (6 PRO) was even more pronounced (60% higher), but this is not so strange considering the effect of protein on BG.

Insulin values were not significantly different between the CHO-groups and the PRO-group had the lowest values; again, this is not unexpected given that carbs are more insulinogenic than protein.


The authors of the paper sums up the results

The present study is one of the first to investigate glucose and insulin excursions in response to altered meal frequency and macronutrient composition in healthy young adults over a 12 h period. Our primary finding is that consumption of 6 frequent meals in 12 h resulted in higher blood glucose levels over the course of the day than the consumption of 3 meals, although there was no difference in the insulin response between these two conditions.

The implication here is that it seems insulin was able to do its job better, that of lowering blood sugar, with less meals.

There has been considerable promotion both by the medical community and the lay press to consume 6 meals per day for weight loss or for glycemic control but our data indicate that the glucose AUC is 30% higher over the course of the day with a frequent high carbohydrate feeding than when consuming 3 meals per day.

This is just a nice way to say that mainstream diet advice is a bunch of bullshit.

This could potentially have profound implications for individuals with glucose intolerance or those with type 2 diabetes, and should be studied further in this population.

Not only is it bullshit, but it might even be counterproductive for some individuals.

My Thoughts

A few remarks:

1. The subjects were lean, healthy and young; 18-35 year old males and females with an average of 12% and 30% body fat respectively. Thus they were metabolically healthy and representative of the health-conscious crowd. If they saw a negative effect of high-frequent feedings, one can speculate about what someone in worse metabolic condition, i.e. poor glucose tolerance or insulin resistance, might experience.

2. Ironically, eating six meals a day is pretty much the norm among the fitness crowd – partly due to the belief that it will help be beneficial for blood sugar control…when it does the exact opposite.

3. While the study design was excellent, it can be argued that the sample size was fairly small (8 subjects). This makes the results less definitive than if the sample size were larger. However, this is still a larger sample size than some studies that have been cited when questionable claims of benefits of high meal frequency has been brought up (e.g, this widely cited study by Speechly, et al that used 7 subjects). Furthermore, a sample size of 7-10 subjects is quite typical in these kinds of studies.

4. There was considerable variance in the average BG values of subjects in response to the 6 CHO-diet; a whopping +-251 mmol/l*min compared to the +-99 mmol/l*min for the 3 CHO-diet. The implication of this is that it seems some subjects handled 6 meals very poorly while everyone handled 3 meals well, relatively speaking.

5. It’s unfortunate that the researchers did not compare six high-protein meals to three high-protein meals by including a 3 PRO-experiment in the study. However, I’d be surprised if the results of such an experiment would differ much from those of the 6 CHO vs 3 CHO-experiments.

6. Another conclusion by the authors, or shall we say affirmation of fact, is that of the benefit of increasing protein intake relative to carb intake as an effective preventative measure against the metabolic syndrome and Diabetes Type 2. I’ve talked about the benefits of high-protein diets numerous times in the past, but usually as an effective diet strategy for maintaining low body fat and minimizing fat gain during overfeeding.

However, despite the fact that there is overwhelming scientific support for the positive effects of high-protein diets on fat loss, weight management and health markers, many medical professionals and dietitians are still hesitant to recommend high-protein diets. In the bizarre world of public health professionals, a high-carb (45-65%), low-protein diet (15-20%) is still recommended. Most recently, such dietary advice was given in this shameful publication: “The Dietary Guidelines for Americans, 2010.”

It’s strange to see such ignorance given the growing waistlines of the American population (and elsewhere). When scientific evidence for the effectiveness of high-protein diets are discussed, it is lamely stated that “further research is warranted” – but this has been said for the last 30-40 years, ever since the beneficial effects of high-protein diets started to surface.

This is all strongly related to findings in dietary epidemiology, which is the cornerstone of public diet and health guidelines. But it’s completely worthless. You easily can “prove” that higher protein intakes and meat causes weight gain. But like I’ve said many times in the past, correlation does not imply causation.

Public health recommendations must be given based on controlled studies – where diets are standardized and compared against each other in a controlled and methodological manner (such as the one discussed in this article). Only then can we draw conclusions based on the collected evidence.

For a thorough explanation of why the results of dietary epidemiology can be highly misleading, read “Top Ten Fasting Myths Debunked” where I discuss this in relation to meal frequency and breakfast. As you will note, results from interventional studies dispute the results found in dietary epidemiology.

7. Lastly, a disclaimer: As with every new finding in nutritional research, more studies are needed to to confirm the results found here. While this study was an improvement over older studies, due to better methodology and greater accuracy, nothing can be said with full certainty before the results are replicated (i.e. with a larger sample size).

There is always a chance, however slim, that some of the participants were significantly different from the Average Joe or Jane with identical characteristics. If that was the case, their results, be it that they were genetically wired to handle three meals better, not well suited for frequent feedings, or any combination thereof, would skew the average values for the group and perhaps falsely show statistical significance when there is none to be had.

That said, it would be unlikely that the results found in a future study would be in complete opposition to the ones found here. So for now, I think we can safely pronounce the blood sugar myth dead and state that the latest and best scientific research suggests that a lower meal frequency is superior for blood glucose.

Fake Diet Gurus

Take heed of diet gurus and “health experts” who claims that a high meal frequency is ideal for health, metabolism or fat loss. The advice has always been questionable from a physiological point of view, but the truths have remained hidden and obscured to the public. Buried in academic papers that are unavailable to the lay person or hard to decipher for anyone without solid understanding of the topic. (…And not the “solid understanding” displayed by the so-called experts.)

Remember, there was no incentive to talk about the meal frequency myth, but there has been, and still is, lots of monetary incentive to have people believe that smaller meals and frequent feeding is important. Supplement companies makes billions from people that gulp down protein shakes, bars and meal replacements in the false belief that it might be beneficial.

However, we’ve all been there, believing the myths. Eating breakfast, drinking whey shakes, watching the clock for the next meal, worrying about catabolism, etc. Given that the noise of those who have something to gain by keeping these myths alive is much higher than those who spread the right information, it’s perfectly reasonable that we buy into the things we hear everyone saying. After all, it is not until fairly recently, in the last few years, that the myths has been exposed – by myself and others.

The evidence against any supposed benefits of high meal frequency is mounting, while the evidence that speaks for a lower meal frequency is emerging. The information has been presented – it is available. Now you just can’t get a free pass if you keep spinning the same bullshit about how meal frequency stokes your metabolism and all that other nonsense.

In this day and age, anyone that claims to be a health or diet expert and keeps regurgitating these myths displays the mark of incompetence. Be wary of them. They are still around and their number is legion. Confront them when you can and watch them squirm when you ask them to back their claims up with scientific evidence. Ignorance is inexcusable today if you claim to be an expert.

That’s all for tonight, folks.