The YAM CanFoG - Results and Discussion

I am quite pleased to be reporting that the Young Adult Male following Canada's Food Guide (YAM CanFoG) study has be terminated early based on my predefined criteria for a clinically-significant weight loss (+/- 3 lb / 2 weeks). While the study title essentially says it all, for those interested, I have provided the details of my little mis-adventure on a previous blog post: "The YAM CanFoG - Design".

The study objective was to determine if Canada's Food Guide provides the appropriate amount of calories for young men (19-30 years old). My research hypothesis was that consuming only the foods recommended would result in negative energy balance and weight loss. I meant this in the general sense, acknowledging that the diversity in food supply and variability in energy expenditure was bound to generate combinations that would produce a positive energy balance. Indeed, I previously demonstrated that my 7-day food record included only 50 different food items, but could provide anywhere between 1,205kcal and 3,308kcal per day while adhering to the recommendations for number of servings within each food group.

I decided to start at the beginning, looking at the study that was used to generate the food groups, daily servings and recommendations for food choices in the guide. I reported my findings in the post: "Canada's Guide to Weight Gain?". Briefly, Canada's Food Guide was developed with a focus on nutrition, looking at what model would result in adequate intake of nutrients while limiting the risk of toxicity based on the food choices of Canadians. There were 39 different nutrients variables that needed to be considered in this modeling process:

In addition to looking at these nutrients in each of the 16 age-and-gender groups for the 50 food models being considered, the researchers assessed energy balance. Erring on the side of caution, the researchers compared estimates of energy intake with energy requirements based on sedentary activity level. For most age-and-gender groups (8/12), the median estimated intake was within or below the median energy requirement range.

Overall, the process was transparent and well-designed, and is a useful model for developing and evaluating dietary guidelines for a population. However, there were a few limitations worth noting:

1. Dietary intake of Canadians was derived from the Federal Provincial food and nutrition surveys for BC, Manitoba and Ontario for adults, and Quebec for children.
2. The 500 simulation diets that were used to test each dietary model, while representative of the population intake, may not capture the food choices of the individual.
3. Validation of the dietary model, in particular its interpretation and application by the public, was not assessed by fielding testing in free-living subjects (to my knowledge).

For males 19-30 years old, the median estimated intake of 2,344 was below the median estimated energy requirement range of 2450-2550kcal per day. While these findings support my research hypothesis, given the limitations of the study, further investigation was warranted. To this end, I decided that I should put Canada's Food Guide to the test. Admittedly, I "borrowed" the idea from physical trainer, Eric Briere, who is currently (as far as I know) still following the food guide, and gaining weight.

I first approached this mathematically by comparing my 7-day food record (~2000 kcal/d) with median estimated energy requirements for males 19-30 years old (2450-2550kcal/d). This approach also supported my research hypothesis, but given the numerous sources of variability and error, the only way to be certain was to actually test it out.

It was an n of 1 study, and I was the subject. I'd hardly consider myself to be representative of the average man. At 5'5" and 138.8 lb, I would expect my energy requirements to be below the median range. However, as pointed out previously, my diet is arguably lower in added fats and sugars than the general population, and would therefore provide less calories.

I am pleased to note that compliance with the diet was 100% (no cheating). I also managed to adhere (somewhat) with the recommendations in Canada's Food Guide to obtain at least 30-60 minutes of moderate physical activity every day. My average was 70 minutes/d, although some days I was sedentary (15 minutes of activity), and others I was quite active (up to 150 minutes of activity). This weekend effect depicted in the table below is probably consistent with what most individuals experience who work the standard M-F shift.

I wore a BodyBugg during the experimental period (thank you Dr. Alan Titchenal). Unfortunately, I didn't obtain this device until the afternoon of Day 1. I have provided the values for readers, but excluded it from analysis. My estimated energy expenditure was 2,458 kcal/d (2,151 - 2,790 kcal/d), and my number of steps was 7,000 steps/d (4,607 - 10,359 steps/d).

These findings indicate that I was in negative energy balance of ~500kcal/d for the two weeks of study. Using the 3,500 calories = 1 lb of body weight, I would be expected to lose ~1 lb/wk while following Canada's Food Guide. My actual weight change can be seen below:

As you can see, my weight loss of 3.1 lbs exceeds what would have been expected by the math by roughly 150%. There are several possible explanations for this discrepancy. First and perhaps as likely as any other, we are dealing with relatively small changes in weight, so slight errors in measurements and estimates can have a larger impact. Another possibility is glycogen storage depletion. Our bodies are able to store excess carbohydrate in our diet (~2,000kcal worth) as glycogen. This reserve of carbohydrate provides glucose for cells that need it (ie. red blood cells, brain cells) during periods of fasting such as between meals. Glycogen is a branched molecule that traps water, which partly explains why individuals following low-carbohydrate diets tend to experience a more rapid weight loss early on in the diet. As their glycogen stores become diminished, they lose weight as water (~1:3 glucose:water). My diet wasn't low-carbohydrate (53% of calories), but if I was in negative energy balance of ~500 calories per day, it is reasonable to assume that my glycogen stores may have been slightly depleted.

There are a limitation of my study that I need to acknowledge. I consumed less calories than the estimated median intake derived from the population. It is difficult to account for this difference, particularly given that my range in daily intake (1813-2090kcal/d) was below even the 10th percentile of 2100kcal/d. My diet is compliant with the food guide, leaving me to wonder if this inconsistency is not at least partly related to the use of vague, lenient verbiage such as "select lower fat milk alternatives" in the food modeling.

I am admittedly happy that the experiment was terminated early. Prolonged negative energy balance, even for two weeks, is unpleasant to say the least. It was a great learning experience for me, and has certainly got me thinking, just not about food guides.

Katamay SW, Esslinger KA, Vigneault M, et al. Eating Well with Canada's Food Guide (2007): Development of the food intake pattern. Nutrition Reviews 2007; 65(4): 155-66.