Time and again so called studies of “calorie restriction” (CR) have failed to implement well-designed restriction, randomizing overweight subjects instead, making them no different from the thousands of weight-loss studies. But if we ever want to confirm or refute the efficacy of CR we need high quality data. Alas, rarely do we get such data. So, always remember that CR must be implemented in lean and healthy animals & humans.
Unsurprisingly, the CALERIE study did it again (1, 2). That is, failing to randomize lean people. They already once messed this up with their phase 1 pilot study, which was forgivable. At least we are getting close at a BMI of 25.1 (23.8, 26.4) and n=220, but it's not good enough. There are two reasons for this failure:
1. Americans are getting heavier and heavier and it is hard to find lean people,
1a. Therefore the maximum BMI they allowed was 28 , clearly overweight by most criteria,
3. They excluded volunteers below a BMI of 22.
Their excuse for the latter is flimsy (2):
“The lower BMI limit was selected primarily for safety reasons. That is, the commonly used standard of underweight is BMI less than or equal to 18.5 kg/m2, and the lower limit provides an adequate buffer from this threshold as participants lose weight over the 2-year intervention.”
Oh my, oh my. And they do not even give a reference. I wonder why? Sarcasm aside, the reason is obvious, since the 18.5 BMI-cutoff is arbitrary. If you trust the epidemiology anything below some 22-25 is unhealthy (3). If you include other strands of evidence, i.e. controlled trials in animals including primates, short term trials in humans and (arguably, weaker, “cross-sectional”) epidemiology (see ref. 4. for a brief review) you must admit that we simply do not know and that the optimum might as well be at very low BMIs or rather very low calorie intakes.
Why then does a study trying to test whether low calorie diets are safe and healthy assume the conclusion that they are not?
If they were really concerned that people could lose an unhealthy amount of weight, they could simply adjust the CR% for people falling below, say, BMI ~19 or even drop people below a certain BMI from the study. Or implement lower CR% for those between BMI 20-22. Or anything else, really.
Are they incompetent? No, I hope I am wrong and the researchers had a good reason for their flawed design e.g. ignorant ethics committees or difficulty in recruitment. The latter is plausible since their relaxed criteria still yielded 98% recruitment losses (see below).
Maybe I am underestimating the severity of weight-loss from 25% CR, but I doubt it, c.f. Biosphere 2. The latter combined pretty severe initial 30% CR, hard work and considerably leaner participants. Nonetheless only the women reached a BMI of 18.5 down from 21.2, although variability was high. Overall biomarkers improved. So where is the evidence for harm? (p. 455, ref. 4).
Are the CALERIE investigators inexperienced? This is very possible. I do not see co-authorship by known CR-experts: Weindruch, Spindler, Fontana, Holloszy, M. Rae or the teams carrying out the primate studies.
I am curious if subset analysis will salvage this study.
Screening and Recruitment
(adapted from reference 1 and 2)
1. Contemp Clin Trials. 2013 Jan;34(1):10-20. doi: 10.1016/j.cct.2012.08.011. Epub 2012 Sep 5. Comprehensive Assessment of Long-term Effects of Reducing Intake of Energy Phase 2 (CALERIE Phase 2) screening and recruitment: methods and results. Stewart TM, Bhapkar M, Das S, Galan K, Martin CK, McAdams L, Pieper C, Redman L, Roberts S, Stein RI, Rochon J, Williamson DA; CALERIE Study Group.
2. J Gerontol A Biol Sci Med Sci. 2011 Jan;66(1):97-108. doi: 10.1093/gerona/glq168. Epub 2010 Oct 5. Design and conduct of the CALERIE study: comprehensive assessment of the long-term effects of reducing intake of energy. Rochon J, Bales CW, Ravussin E, Redman LM, Holloszy JO, Racette SB, Roberts SB, Das SK, Romashkan S, Galan KM, Hadley EC, Kraus WE; CALERIE Study Group.
3. Harrington M, Gibson S, Cottrell RC. Whitlock G, Lewington S, Sherliker P, et al. Body-mass index and cause-specific mortality in 900 000 adults: collaborative analysis of 57 prospective studies. Lancet. 2009;373:1083–1096.
4. Fontana, L., Coleman, R. J., Holloszy, J., & Weindruch, R. (2010). Calorie restriction in non-human and human primates. HANDBOOK OF THE BIOLOGY OF AGING,.