Let’s talk about this paper, part of the CALERIE studyEffect of Calorie Restriction with or without Exercise on Body Composition and Fat Distribution

I made a promise to myself, and may have mentioned it here, that I would no longer review single papers without also discussing the body of evidence. Today, I’m breaking that promise because I want to highlight this study as an example of what I feel is great control of important variables for the topic.

The deets:

  • 35 overweight participants (16 males, 19 females)
  • 25% calorie deficit reached one of two ways: calorie restriction (25%) or a combination of calorie restriction and exercise (12.5% + 12.5%).
  • CR + EX participants were allowed to choose their own exercise intensity and duration was adjusted to burn an appropriate number of calories.
  • Weekday breakfast and dinner were eaten at the research center and snacks, lunch and weekend meals were provided for take-home. During a 10-week period in the middle of the study, this was discontinued and subjects were given dietary guidelines to follow on their own. The diet was 55/15/30 PRO/CHO/FAT
  • Weekly meetings using cognitive behavioral therapy to improve motivation and teach people to eat and exercise. Pretty cool, huh?

Next to a metabolic ward study, where participants live in a hospital and all food intake and activity are monitored, this design is about the best you’re going to find in the exercise and nutrition literature. The fact that all meals and snacks are provided greatly increases the ability of the participants to adhere to the eating plan. We know that people often under-estimate and therefore under-report their calorie intake so this is a very important variable to control. (Hill 2001)

This goes the same for the exercise – 3/5 exercise sessions per week were supervised and all others were recorded using portable heart rate monitors. Excellent! This is far better than researchers telling subjects “we want you to do x number of minutes at x intensity on your own” and then praying it gets done. Cuz researchers pray – scientists are almost all religious…


Before I get to the results, let’s look at some other cool stuff about this study. I was literally cheering for these researchers as I kept reading.

“Yeah, buddy! Meals provided!” “DUDE! supervised exercise sessions! SWEET!” “Doubly-labelled water and a 14-day in-house feeding period! Fahkin’ Eh, Kehd!”

As my last exclamation alludes to, the team found the subjects energy requirements before the intervention by two separate 14-day doubly labelled water periods and a 14-day in-house feeding period – meaning metabolic ward conditions for two weeks. I should mention, this was all part of a larger study about calorie restriction and these groups weren’t the only ones in the larger study, but were used for this paper.

So, what did this study find? At 6 months, weight loss was equal between the CR and CR+EX groups:

-10.4 ± 0.9% for CR and -10.1 ± 0.9% for CR+EX at month 6

Fat mass loss – same story:

At month 6 the CR group lost an average 23.9 ± 3.0% (women: 21 ± 1%; men: 27 ± 6%) and CR+EX 24.8 ± 2.7% (women: 23 ± 3%; men: 27 ± 5%)

The exercise intervention did not make much of a difference in regard to fat free mass loss, which is the one drawback to this study; I would have loved to see some resistance training thrown in there, especially in a 4th group to see if that could make any differences in FM and FFM loss/retention.

So, what we have here is some pretty good evidence that you can create a calorie deficit through diet, or through diet plus exercise and as long as the deficit is the same you’ll experience the same result. The researchers rightfully mention that exercise carries other benefits like cardiovascular endurance and I would add that RT would provide strength and, again, possibly a retention of FFM.

In another paper Redman et al describe the metabolic adaptations (decrease in thermogenesis) found in each group and claim that the CR+EX group did not suffer the same metabolic slowdown the CR group did. As I review this paper, though, this doesn’t seem to be the case. And trust me, I got excited to think there might be a way to counter the adaptive thermogenesis seen after weight loss.

First, let me mention that a third experimental group was included in the analyses for this paper – a low-calorie diet group (LCD). Researchers found a reduced total daily energy expenditure (TDEE) in the CR and LCD groups at 3 months, but not significant at 6 months unless the two groups were combined. These changes persisted after adjustment for fat mass and fat free mass. The CR+EX group, however, did not show a drop in TDEE. This is where I begin to question the interpretation…

Remember that the CR+EX group is exercising 5 days/week burning a rough average of 500 calories per session. Average this over 7 days and you get 375 calories/day burned due to exercise sessions. The significant slowdown experienced by the CR group at month 3 adjusted for FM and FFM? 371 calories/day. So, it seems that what is really happening is the CR+EX group is experiencing the same metabolic and behavioral adaptations as the CR group, they’re just purposely making up for it by exercising their little tukases off. This is even more pronounced when the authors adjust TDEE for sleeping metabolic rate and the CR+EX group experiences non-significant increases in TDEE while the CR and LCD groups experience deficits again. To me, this means when resting energy expenditure is accounted for these people are burning more calories than the other groups. Of course, they’re on the treadmill 5 days/week. What is also important here is that the equations for estimated energy expenditure were created from measured baseline values, when all groups were sedentary. At follow-up, the CR+EX group is no longer sedentary so this equation for estimating TDEE will not apply!

I would have liked to see comparisons of resting metabolic rates and non-exercise activity between groups, since this is where the adaptation occurs. Take exercise out of the equation, and I bet you’d see similar adaptations between groups.


Take Home

First, bravo to the study’s designers for thinking ahead and producing a well-controlled study with some really cool abilities to decipher what’s going on with calorie deficits of differing mechanisms. It’s easy to rail on researchers for less-than-optimal study design, but we have to keep in mind they have a budget and participants, oddly enough, might not always be willing to live in a prison ward for 6 months at a time. I’ve emailed the authors to pat them on the back (although it may not mean much coming from some schmuck with a blog) and to ask a few questions about TDEE and their interpretation of the metabolic adaptation.

Now to the take-home points:

  1. If you want to lose weight, create a calorie deficit. This can be done through calorie restriction or with restriction and exercise combined.
  2. Choosing to add exercise, aside from cardiovascular or strength benefits, may also make maintenance easier by laying down the habit that will help you out-exercise the metabolic adaptation you’re most likely going to experience.
  3. Science is so much fun


Redman LM, Heilbronn LK, Martin CK, de Jonge L, Williamson DA, et al. (2009) Metabolic and Behavioral  compensations in Response to Caloric Restriction: Implications for the Maintenance of Weight Loss. PLoS ONE 4(2): e4377. doi:10.1371/journal.pone.0004377

Redman LM, Heilbronn LK, Martin CK, Alfonso A, Smith SR, Ravussin E. Effect of Calorie Restriction with or without Exercise on Body Composition and Fat Distribution. J Clin Endocrinol Metab. 2007 March ; 92(3): 865–872. doi:10.1210/jc.2006-2184.

Hill, R. J., & Davies, P. S. W. (2007). The validity of self-reported energy intake as determined using the doubly labelled water technique. British Journal of Nutrition, 85(04), 415. doi:10.1079/BJN2000281




Calories Don’t Count, but CALERIEs Do