Weight-loss variability in response to the same diet

Whoever said weight loss was a straightforward process has probably never tried to go from a BMI of 40 down to 24 kg/m2. I must admit that only a few years back, in the early stages of my graduate training, I was equally naive. I thought: it’s a case of simple math – once your energy expenditure exceeds your energy intake, you lose weight.

It certainly sounds simple, and yet it rarely is that simple in practice.

Take for example a recent study published in the November issue of Obesity by Gasteyger and colleagues which investigated the effect of the size of the weight loss study, in terms of numbers of participants recruited, on the weight loss achieved among the study participants.

In the study, the authors looked retrospectively at data collected during a prior multicenter trial to see if the size of the study center (how many people were recruited to lose weight at site A versus site B, for example) had any influence on the success of the weight loss treatment.

In total, 22 different centers recruited anywhere from 4 up to 85 participants to undergo a 8 week low calorie diet. The subjects BMIs were in the range of high 30’s to low 40’s (class 2 and 3 obese).

At the end of the intervention the average weight loss among all the subjects was approximately 10 kg (or 10% of initial body weight) – not bad for 8 weeks!

However, the average weight loss for a given weight-loss center ranged from 5.8 to 11.8% – despite the exact same intervention.

Interestingly, the authors found a relationship between the number of subjects recruited at each center and the average degree of weight loss observed among those subjects.

Specifically, for every increase in 10 subjects to the center’s study population, the expected average weight loss for participants in that center increased by approximately 0.5%!

So if two obese people joined the same weight-loss study, the one who joined a bigger center, with more subjects, would be expected to lose more weight on the exact same plan.

Not so straightforward, after all.

How do we explain these findings?

Here’s the authors’ best guess:

“The main reason for the correlation between weight loss and the number of recruited subjects per center may be that principal investigators, study coordinators, and dieticians working at centers with high numbers of recruited subjects have more experience than those working at smaller centers; therefore, they are probably more efficient in treating and counseling subjects during a low calorie diet.”

Have a great weekend,


Gasteyger, C., Christensen, R., Larsen, T., Vercruysse, F., Toubro, S., & Astrup, A. (2010). Center-Size as a Predictor of Weight-Loss Outcome in Multicenter Trials Including a Low-Calorie Diet Obesity, 18 (11), 2160-2164 DOI: 10.1038/oby.2010.118

Related Posts Plugin for WordPress, Blogger...
This entry was posted in News. Bookmark the permalink.

13 Responses to Weight-loss variability in response to the same diet

  1. Yes, differences in treatment centers are important, as the authors of the study imply. Genetic differences can also be important. And genetic differences that are sensitive to life choices – how much and what you eat, exercise, sleep, alcohol and tobacco use, for example – can also be important. I will briefly summarize three studies here:

    1. In 2006 (PubMed ID = 16971225), a report of a Spanish population of males and females showed that change in BMI for people on a hypocaloric Mediterranean diet were sensitive to the genotype of a variant (rs8179183) in the LEPR (leptin receptor) gene. Those with the C allele at this position showed less of a decrease in BMI.

    2. In 2007 (PubMed ID = 17544366), a report of Korean females showed less of a drop in BMI when the women carried the A allele of variant rs659366, located in the UCP2 gene, and were on a low-calorie diet.

    3. A report from 2009 (PubMed ID 19543202) showed that in a group of White Americans, both sexes, that there was also risk of less change in BMI when carrying the A allele of variant rs8050136 in “obesity” gene FTO and were engaged an exercise regimen.

  2. M says:

    I don’t have access to this article, but was the weight loss at the centers normally distributed?

    If the weight loss was lognormally distributed, and you are dealing with small sample sizes, you might get a higher mean with more sampling.

  3. WRG says:

    And what about gender and age? Were they comparing men in their 20s to post-menopausal women? Factors such as these are incredibly important.

    • thedocsquawk says:

      Was there any documentation of how much adherence to the diet/exercise or extra exercise there was amongst the participants? I thought that would be the most important factor.

  4. Jamie says:

    They don’t mention adjusting for lean body mass. Surely differences in this, and hence basal metabolic rates, could account for different weight losses in response to the same diet?

  5. Sean Wharton says:

    Great article Peter. I agree with the study. I am a bit bias as our centre has over 20,000 pts and we feel that the volume of patients has certainly helped us to become experts in the field. I have not read the study has yet, but clearly the ability to account for known and unknown variables is primarily based on sample size, and randomization (as this was 1 diet, there was no randomization, therefore sample size will account for variability). Inherent flaw is the small number of patients at the s”small number pt centres”. Although presents as epidemiological error I still agree with the premise of the conclusion, a conclusion that did not need a study to verify. Look at surgeons, barbers, and basketball players and many other professions. The more you do, the better you are. You need the right tools and proper outlook, but with practice and more chances to do what you do, everyone gets better.

  6. Pingback: Weight-loss variability in response to the same diet – PLoS Blogs (blog) | Dr. Simon Ourian

  7. Pingback: Quick Links | A Blog Around The Clock

  8. Leslie Nolen says:

    All good points, particularly regarding the risk of small sample sizes.

    Perhaps there’s also a “social interaction” effect. Could be as simple as numerically more exposures to peers; or a much more subtle effect. For example, perhaps simply witnessing greater numbers of peers, all pursuing the same goal, affects the outcome for some yet-to-be determined reason.

    I’d have to be convinced that there’s a measurable difference in practitioner expertise between the smaller and larger centers, and further that it actually affected client outcomes. Anecdotal evidence suggests that you get to a point of diminishing returns pretty fast – does seeing 20000 pts vs, say, 500 or 5000 really confer greater know-how?

  9. hello there and thank you for your information – I’ve certainly picked up anything new from right here. I did however expertise some technical points using this website, since I experienced to reload the website a lot of times previous to I could get it to load correctly. I had been wondering if your web host is OK? Not that I am complaining, but slow loading instances times will very frequently affect your placement in google and can damage your high quality score if ads and marketing with Adwords. Anyway I am adding this RSS to my email and could look out for much more of your respective fascinating content. Make sure you update this again soon..

  10. free world says:

    Awesome issues here. I’m very satisfied to look your article. Thanks so much and I am looking forward to touch you. Will you please drop me a mail?

  11. Hello there! I know this is somewhat off topic but I was wondering which blog platform are you using for this website?

    I’m getting fed up of WordPress because I’ve had problems with hackers and I’m looking at options for another platform. I would be fantastic if you could point me in the direction of a good platform.

    Here is my web blog weight loss help