Obese, but Metabolically Healthy: Is Weight Loss Beneficial? (Series Pt 5/5)

So what have we learned thus far?

1. About a third of obese individuals fail to exhibit the metabolic complications commonly attributed to excess weight.

2. These same individuals also seem to be at the same relative risk of diabetes and cardiovascular disease as equally healthy, but lean individuals.

3. Nevertheless, despite being metabolically healthy, some evidence suggests that excess weight may put such obese individuals at risk for early mortality due to other, non-metabolic, factors.

4. This latter point would imply that all obese individuals should be encouraged to lose weight, despite their metabolic health. This, in fact, is in line with guidelines developed by leading health authorities which currently recommend weight reduction as the primary treatment strategy for all obese patients, regardless of metabolic health.  However, as we learned yesterday, weight loss via caloric restriction among metabolically healthy obese may actually result in a deterioration in insulin sensitivity, thereby increasing risk of developing type-2 diabetes.

Now, as most of you know, when a completely counter-intuitive finding like this comes along, where even the study authors fail to come up with a plausible mechanism, it is up to other researchers to follow up with additional research to either corroborate or refute this original finding.

Because I am personally drawn to paradoxical and counterintuitive findings in science, I was very intrigued by the findings of Karelis and colleagues and decided to follow up their study, but include a few variations:

a) Since the study of Karelis et al. only used female subjects, we wanted to ensure this wasn’t due to a gender effect and thus included both men and women.

b) Additionally, to test the possibility that their finding was driven only by modality of weight loss (caloric restriction, in their case) we employed a number of weight loss interventions (diet alone, exercise alone, and the combination of diet and exercise).

c) Finally, while the original study only looked at insulin sensitivity, we decided to assess changes in other variables of interest (body composition, blood lipids, glucose and insulin levels, etc.).

In our study, which has just been published in the prestigious journal, Diabetes Care, a total of 63 metabolically-healthy obese men and women and 43 metabolically-unhealthy obese men and women participated in 3-6 months of exercise and/or diet weight-loss intervention.

And what did we find?

First, body weight, waist circumference, and total and abdominal fat mass were significantly reduced in all subjects – regardless of gender, modality of weight loss, and metabolic status.

Second, in contrast to the findings of Karelis et al., insulin sensitivity IMPROVED after weight loss in both the metabolically-healthy (by about 20%) and metabolically-unhealthy obese individuals. However, the improvement was greater in the metabolically-unhealthy subjects. See figure below.

Importantly, this improvement was similar across all weight loss modalities. In other words, dietary caloric restriction did not have a unique negative effect on insulin sensitivity.

Finally, while the metabolically-unhealthy obese individuals also showed improvement in numerous other outcomes (triglycerides, fasting glucose and insulin, HDL-cholesterol, and total cholesterol), a reduction in fasting insulin was the only other metabolic improvement among the metabolically-healthy obese. This latter finding is not surprising given the normal baseline levels of most metabolic risk factors among metabolically-healthy obese individuals. That is, since they were healthy to begin with – they can only get so much healthier after weight loss (ceiling effect).

Thus, we found no evidence of deterioration in metabolic profile among metabolically-obese individuals who lost weight via a lifestyle intervention.

While limited health care resources dictate the need to prioritize high-risk obese individuals for aggressive treatment, to imply that obese individuals who are metabolically healthy should not lose weight may not be the most appropriate public health message.  Such a public health message may be particularly misguided at a time when the prevalence of obesity continues to increase, despite a greater public awareness of the benefits of weight loss. In this context, our findings reinforce current recommendations which suggest that all obese individuals should be encouraged to lose 5-10% body weight.

Bottom line?

Although a fair number of obese individuals may have a perfect metabolic profile, it appears they may still experience negative consequences of their excess weight. Furthermore, weight loss achieved via lifestyle intervention appears to still bring about some metabolic benefit among previously healthy obese individuals (it certainly doesn’t seem to harm health). Given the numerous non-metabolic benefits of weight loss (mobility, joint problems, psychological status, sexual function, etc.), all obese individuals have something to gain from a modest 5-10% weight loss.

Have a wonderful weekend,


References and Further Reading:

Janiszewski, P., & Ross, R. (2010). Effects of Weight Loss Among Metabolically Healthy Obese Men and Women Diabetes Care, 33 (9), 1957-1959 DOI: 10.2337/dc10-0547

Related Posts Plugin for WordPress, Blogger...
This entry was posted in News, Obesity Research, Peer Reviewed Research, Series. Bookmark the permalink.

43 Responses to Obese, but Metabolically Healthy: Is Weight Loss Beneficial? (Series Pt 5/5)

  1. Thanks for a great series, Peter. One question: You mention in the first point that 1 in 3 obese are metabolically healthy. Why did you choose to use the looser definition instead of the stricter one proposed in your colleagues’ study, which lowers the number of MHO to 6%? It seems to me that the presence of any risk factor which is independently associated with morbidity and mortality disqualifies one from being “metabolically healthy”, which is why the stricter definition makes more sense.

    • You bring up a good point Chris. To date there exists no established criteria for defining metabolically-healthy obese. The two most common schools of thought either use the upper tertile or quartile of insulin sensitivity (alone) or the lack of presence of metabolic syndrome. Beyond that have been a number of variations of that theme. Thus, if you consider all the studies looking into this issue the most appropriate number in terms of prevalence of these individuals in an obese population would be about 6-40% – widely-varied depending on the definition used. Even if you take Dr. Kuk’s approach and include only individuals with no metabolic risk factors – this will vary from study to study depending on what factors were even measured in that study. If one study includes inflammatory markers, or measures of coronary calcium, or endothelial function while others don’t – you can see how this creates a problem.

  2. Pingback: Obese, but Metabolically Healthy: Is Weight Loss Beneficial | Weight Loss Inside

  3. Andreas Johansson says:

    What if anything, then, explains the result of Karelis et al? Were they just confounded by noise in a smallish population?

    • Great question Andreas! It is one that neither the authors of the study , nor the editorial that accompanying that publication, nor I could answer. To this day, I have no idea. This is primarily why we felt it was necessary to see if we can replicate the findings.

  4. Great series! I really enjoyed reading it

  5. Suzanne Willis says:

    I assume the individual lines in each graph represent unique individuals, while the yellow dotted line is an overall fit to the total sample? If so, it looks as though, while there was improvement on the average, there were individuals whose insulin sensitivity decreased. The graphs are hard to read but I count one metabolically unhealthy man, four metabolically unhealthy women, two metabolically healthy men, and four (or five; one slope is close to flat) metabolically healthy women where this appears to be the case. If this is true, would it be worth investigating further to determine individual characteristics that might affect whether weight loss would decrease or increase insulin sensitivity in a particular individual? This would seem to be particularly interesting in light of the fact that long-term weight loss is difficult to achieve, and many who succeed in the short term regain the lost weight (and more, in some cases). You have mentioned that this cycle of weight gain and loss is more detrimental to health than simply maintaining at a higher weight, so being able to identify subgroups for whom the weight-loss intervention might not be beneficial would seem to be a good idea.

    • Suzanne, you’ve pointed out a very important issue in clinical research – that is, the heterogeneity of response among subjects to any given treatment. This is almost always the case looking at individual responses and is one of the reasons large sample sizes are often needed to cut through the random noise. It certainly would be interesting to investigate what factors influenced this outcome, but alas, the number of subjects in this study is too small and the number of possible confounders too great for such an analysis.

      • C.S. says:

        It is understandable that you wouldn’t be able to improve the conclusions of your study by examining the case histories of individual patients, but it seems to me that you and others are missing the point by throwing your hands up. Part of the purpose of the scientific method is to develop new hypotheses, so if you can investigate the histories of the study participants whose results were different, maybe you can find a significant variable to test in the future.

  6. Jennifer says:

    But in the end, what is the point in recommending weight loss to anyone, regardless of their size, if the vast majority can’t keep off the weight permanently?! It seems you are just encouraging weight cycling, since you know how poorly dieting works as a weight loss measure.

    • Joanne says:

      Yes – I would be curious to see how these same subjects fared with respect to their ‘metabolic health’ after the nearly inevitable regain of their lost weight.

    • Nathaniel says:

      As someone who has lost the weight and kept it off, I would request you take a more positive attitude towards those in my former and – according to you – current position. The point of recommending weight loss is to aid the individual, not to up the statistics.

      • Joanne says:

        Sorry, no offense was intended. The point I was trying to make was that since it’s been shown in other studies that the majority of people who go on weight loss diet regain their weight (and then some, in many cases) we need to consider the metabolic consequences of this. Could the weight cycling make a formerly metabolically healthy obese person, metabolically unhealthy? Might they be better off if they hadn’t dieted in the first place? It’s all fine and well to talk about the benefits of weight loss, but is it ethical to encourage someone to do so if the chances of them succeeding are not great and we don’t know the net effects on their health in the long run?

        • Katje Sabin says:

          I too am interested in the answer to this question.

          Also, how much weight did the subjects lose? Did it drop below their “set point”?

          What do you think of the set-point theories (at least, as far as they are explained in the HAES book by Linda Bacon)… ie, when you lose a lot of weight, the body tends to see the loss as famine and resets the set point even higher… and more firmly… than before, and that (so far) there is no healthy or reliable way to reset the set point lower?

          Thank you.

        • Nathaniel says:

          Thanks, Joanne. My comment was directed more at Jennifer questioning the need to recommend weight loss to anyone rather than your probing deeper into the conclusions of the results of Dr Janiszewski’s research (although the word “inevitable” did sting a little, I must confess). Nevertheless, I am happy to tell you what I think.

          It seems to me that there are many health solutions that can have serious unintended consequences while not being 100% effective, or effective at all. Look at HIV medication, chemotherapy, bone marrow transplants, to name but a few. Medical practitioners recommend them regularly and – if they do their jobs right – counsel their patients through the pros and cons of these treatments.

          I believe a similar ethical situation exists in recommending weight loss, although in this case the individual has much more control over the outcome and success of the “treatment”. While failed weight loss or yo-yo dieting may do more harm than good, I believe medical practitioners should indeed steer patients towards weight loss while indicating the consequences of both successful and unsuccessful outcomes. Perhaps knowing that failure may cause additional negative consequences can be an extra motivator to a successful outcome.

          If there is a high relapse rate, that does indeed require further effort to eradicate. But that is a separate issue. I believe counseling people towards better health is still the optimal solution.

          • MollyMurr says:

            I don’t think that dieters need any additional negative consequences of failure, they already are being told that they’ll die if they stay fat!

            Only 5% of people who diet will keep the weight off. To compare being fat to having leukemia or HIV is problematic. An average fat person aged 35 would probably live another 5 years even staying at their weight, but without treatment the HIV or leukemia patient probably would not.

  7. Troy says:

    Thank you for such a good series. As a future primary care physician I really enjoy you going over useful primary research. I have a few questions as I am not at my university this rotation and can get through the pay wall form my home computer.

    Did you look at any patient based endpoints like MI, development or progression of diabetes and stroke? Were all measures taken on proxies for patient health (triglycerides, HDL, insulin sensitivity)?

    Second what is the confidence interval for the reduction in both the metabolically healthy and unhealthy groups?

    Third is there increasing benefit for weight loss beyond 5-10%?

  8. Nathaniel says:

    As a former obese metabolically unhealthy individual with the resulting insulin resistance, who is now a very fit personal trainer (still with an insulin resistance, of course though), I found an excellent series that has given me some great tidbits to convey to frustrated clients. Thank you very much.

  9. Lyn says:

    I really enjoyed this series. Apart from the information itself, it’s a really good example of how research questions are formulated and how good research builds on such questions.

  10. cass_m says:

    Interesting study. Thanks for the series. I follow Dr. Sharma as well and am intrigued by his premise that obese people should target weight stabilization if they are metabolically healthy. Definitely more work needs to be done for people who do not improve metabolic health with weight loss; perhaps the problem is effective help for weight loss so the cycling doesn’t start. Shouldn’t all adults target weight stabilization?

    Like Nathan I have lost and kept off weight for several years by changing my lifestyle. I have (self imposed) diet restrictions but am not on a diet.

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

  12. Sandy Brooks says:

    Hi Peter, I came across this via a metafilter post and then recognized your name. Anyway, I just wanted to say hello and that I enjoyed reading the series. Looks like you’ve been busy since project ALLIFE.

  13. Nancy says:

    As a dietitian I find if the dietary changes are sustainable and work with the person’s taste preferences, budget, cooking level etc (unlike many diet programs) that people increase their odds of keeping the weight off. Some of my clients who come in who can’t understand why they aren’t losing weight even though they focus on healthy food have made significant progress once they learned the appropriate calorie level to aim for, portion size information, how to emphasize lower calorie density foods such as fruits and vegetables, the right types of snacks to eat, avoiding skipping meals, and reducing portions of high calorie density foods and reducing liquid calories.

  14. Jennifer says:

    I wasn’t meaning to be negative when I said that most dieters regain weight. It’s a very well-known problem. There are some individuals who can keep off a lot of weight permanently, but it’s unusual. So, I wonder how you can address this problem in light of the study results, which seem to indicate that all obese people should lose weight.

    I also think it would be useful to do some research into metabolically unhealthy overweight people and see what percentage of these people have a history of weight cycling. Has weight cycling contributed to their health problems? What percentage of these people can attribute at least some of their obesity to the weight gain that often results from a big weight loss?

  15. Pingback: Wednesday Round Up #121 | Neuroanthropology

  16. Pingback: Recent Reading – 9/25/10 | Everyday Biology

  17. Pingback: Why researchers should blog | higher ed marketing

  18. Pingback: Science of Blogging | How and why all scientists should be blogging | Blogs | 24 Media Labs

  19. Pingback: Social Media For Scientists Part 3: Win-Win | | NEW RESUME PORTALNEW RESUME PORTAL

  20. Pingback: Social Media For Scientists Part 3: Win-Win - DOCUMENTS – DOCUMENTS

  21. Pingback: Social Media For Scientists Part 3: Win-Win | | RESUME PORTALRESUME PORTAL

  22. Pingback: UTILITY DOCUMENT » Social Media For Scientists Part 3: Win-Win » UTILITY DOCUMENT

  23. Pingback: Social Media For Scientists Part 3: Win-Win | | RESUME DOCUMENTRESUME DOCUMENT

  24. Pingback: Social Media For Scientists Part 3: Win-Win - NEW RESUME PORTAL – NEW RESUME PORTAL

  25. Pingback: Social Media For Scientists Part 3: Win-Win | Letter Format

  26. Pingback: Communicating Science 2.0 | Max Bingham | Science Writer and Editor | Blog

  27. Pingback: Social Media For Scientists Part 3: Win-Win | Next Job Application

  28. Pingback: Social Media For Scientists Part 3: Win-Win | Job Application Letter

  29. thanks for this; it is really interesting but more — inspiring. For a person like me with a “sluggish” metabolism and no health risks other than obesity, I am constantly asking myself “Why do this? Wouldn’t the kids rather have a happy fat mom than a thinner grumpy one?” So it is very helpful to be reminded that I am still at risk of early mortality.

    Thought you would like to know that your work is doing some good amongst the non-scholarly!

  30. Pingback: Beneficial Body Weight Control