June Challenge - What's a Healthy Weight?

[Mitra]

I've decided to move over here with some of the issues raised in this week's maintenance thread. If your aim in being slim is to look good on the beach, then read no further. Is slimmer always better? How low does your body fat need to be? Is abdominal fat more important than overall fatness?

I don't know if I can answer all these questions, but I'm going to start by looking at A prospective study of adiposity and all-cause mortality: the Malmo Diet and Cancer Study, by Lahmann PH, Lissner L, Gullberg B, Berglund G (http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=12006635&query_hl=14&itool=pubmed_docsum).

I have to admit that one factor that influenced me in choosing this paper is that the full version was available on-line, free. So much for the purity of scientific research :rolleyes: .

[banshee]

Mitra, I'm with you on the "freely available" thing. After all, I'm already spending a bunch of time reading, and I'm trying to get my exercise in every day, so I'm all for being lazy about what I read. :p

I'll be interested in hearing what the research says about this. I know that the lady at the fitness center who did my body analysis was pleased when I told her I didn't care what I weighed as long as my body fat percentage was healthy. :D

[LisaS]

I'm especially interested in this study because my family is all Swedish and so I finally "match" the gene pool (more or less) of the group being studied. I'll be reading it over lunch today :)

[LisaS]

well, that was interesting. WHR seems to be more important than BMI - at least in older women - is that what you understood as well? To me, this just confirms that VAT is evil ... or at least a sign of other bad things going on -

what do you think about the weight/fat thing though and mortality - is less fat/lower wt associated with mortality because so many diseases end up being wasting diseases? a little chicken or egg --- did you get that impression? I can certainly say that in the "little old ladies" that I grew up with in church (again, these were 1st gen Swedes or other Scandanavians) - the sturdy ones were a lot more lively than the skinny ones of the same approximate age - so less observed morbidity at about the same age. I don't remember any really "in between" sizes in the seniors from when I was growing up.

[Mitra]

The study (http://www.obesityresearch.org/cgi/content/full/10/5/361) involved over 27,000 Swedes, aged between 45 and 73, and reports on their death rates over the next 5.7 years. As a measure of robust good health, death in the next 5 years is not perfect, so all the results need to be taken in that context - it's not showing quality of life, or whether a 50 year old can expect to live to 90 rather than 70. It is indicating how various measures of excess fat predict death rates over the relatively short term.

The measures used were
BMI (Body Mass Index = wt/ht^2)
Body Fat percentage
Lean Body Mass
WHR (Waist to Hip Ratio)


Generally the participants were considered in four separate groups:
Men 46-59
Men 60 - 73
Women 45 - 59
Women 60 - 73


For each group, the members were divided into five according to their fatness as judged by the various measures. This is a very common way of doing an analysis, but one of the disadvantages is that you can get uneven intervals, and that the end groups can contain extremes that you don't know about - for instance, the very thinnest group could include anorexics, but we don't know if it does, or if most of the members were just slightly below the cutoff point. Similarly at the top end, for a group of women above 35% body fat, we don't know if there were a sizeable number who were well above that, or if most were just slightly above.

That's not to say this is a "bad" study, just that it's fairly broad brush, and you can't expect too much detail.

Body Mass Index (if you want to see the detailed numbers, the link to the study is at the top of this post).
For men, the lowest risk was found for those with a BMI of about 25 - 27. For middle-aged women tended to have lower risks for lower BMI, though the results weren't statistically significant (ie you can't be sure the differences weren't just due to chance). For older women (above 60) there was a lower risk for BMI higher than 22. There was no sign of any increase of risk for either sex in the 25-29.9 range often described as "overweight," or "pre-obese."

Body Fat
For the middle-aged women, increased body fat increased the risk of death, but for the over 60s the risk reduced for higher fat.

For men of any age the lowest risk was 17-20% BF, and the risk associated with higher levels was more for older men.

Fat Distribution
The effects of fat distribution (judged by waist/hip ratio) didn't depend on age for either men or women. Overall, more fat round the middle increased the risk, especially for women. For men, there was less difference between the effects of fat around the middle, and overall fat levels, whereas for women fat around the middle was much more significant: the risk for those with a ratio greater than 0.835 was twice as high as for those with a ratio less than 0.75.

All of these risks are presented as relative risks. The women with the lowest waist/hip ratio had 55 deaths out of 3,363 (0.1% per year) and those with the highest had 120 (0.25% per year), so the doubling that's quoted (their relative risk numbers take into account age, exercise &c) actually represents an increased risk of about 0.1% per year, or one in a thousand. Of course, it's possible that those factors leading to an increased risk during this short period would continue to exert an influence, and maybe have a life-shortening effect later on.

My Conclusions
I didn't see any reason to disagree with the authors' conclusions. Body fat is a more useful measure than BMI. Waist to hip ratio, independent of body fat, is important for women. Men seemed to do best with BF 17-20%, whereas for women, while low bodyfat seemed more healthy in the younger group (45-60), higher levels reduced the risk in older women.

I have another pape (http://www.annals.org/cgi/reprint/138/1/24.pdf)r, based on the Framingham Heart Study, that looked at people aged 30 - 49 for 40 years, so that should give more idea of long term effects. And, of course, there are other issues such as osteoporosis that we mentioned in the maintenance thread, where higher body fat levels were protective (except if you exercise, when LBM becomes significant).

[Mitra]

I'm not going to discuss the Framingham paper in detail. It used the familiar WHO classifications of BMI, but didn't (for reasons they didn't explain) include any underweight participants, so there's no info on the effect of low BMI. The quick and dirty version is that being in the overweight category at 40 reduces your life expectancy by about 3 years compared with "normal" weight, and being in the obese category reduces it by about 6 (the effects were very similar for men and women). Losing or gaining weight over the next 20 years didn't seem to affect the risk much.

One of the statistics that stuck in my mind from Anthony Colpo's new book was that men who are diagnosed with diabetes at 40 die on average 12 years earlier than otherwise (14 years for women). (As an aside, in talking about factors affecting heart disease, I think he barely mentions weight - blood sugar, antioxidants, dietary fat and many others are covered, but I don't remember seeing much about body fat.)

So, pulling some of these random facts together:


Being obese shortens your life about half as much as being diabetic (presumably diagnosed by high blood sugar);
Excess fat around the middle is more dangerous than total fat, especially for women;
Older women with low body fat have lower bone density;
Losing weight later on didn't change life expectancy in the Framingham study;
Exercise is good.


Of course, having a lot of excess weight has many effects on your life that won't show up in mortality tables - like making everything harder work; not fitting in standard seats, clothes; not being able to keep up with your children/grandchildren and all the rest. But you don't need to be a superwaif to improve those things.

So I'm wondering ... maybe we should just keep our blood sugar under control (by eating low carb, for instance ;) ) and not worry too much about weight? Billie was right all along - throw your scales away :D.

Unfortunately, once you start to look at issues other than death (which has the advantage of being very easy to measure!), it widens the scope hugely - so I don't think I'll be able to cover every health issue that could be related to weight! But I haven't seen anything yet that suggests being at the thin end of normal is better than being at the heavy end, or even that being a little bit overweight is very hazardous, particularly as you get older.

[Gabriel Guzman]

I remember some time ago (actually long time ago...) that we had a discussion about the real value of BMI as indicator or predictor of anything. I've always been against using BMI as a predictor of risk and the reasons have been discussed by many not only in the published scientific work but also in the media. To me, the inadequacy of BMI starts with its failure to indicate where there is excess fat deposition (truncal, abdominal, etc.), plus its failure when body composition and complexion are considered.

Body fat percentage, on the other hand, may be a more powerful parameter to track any progress of nutritional interventions aimed for weight loss and body recomposition. Waist-hip ratio, on the other hand, may be a better indicator of risk as this paper suggests. However, since this paper analyses all cause mortality, coronary heart disease as cause of death is also included, though not specifically discussed in relation to WHR or the other parameters. To me this is important because while they show a convincing relationship between all cause mortality and WHR, and if we don't separate CHDs, we may conclude that WHR may also correlate with CHDs as cause of mortality. Until recently, it had been accepted that excessive visceral fat is associated with insulin resistance and other metabolic risk factors for coronary heart disease because of its contribution with free fatty acids to the portal system. However, an elegant study published in 2004 demonstrated that free fatty acids derived from visceral fat was much less than the relative amount derived from lipolysis of subcutaneous fat. In fact, only 5% and 20% of portal vein FFAs originated from visceral fat in lean and obese subjects, respectively and the data demonstrated that visceral fat is not as important as subcutaneous fat in supplying FFAs to the liver in lean or in most obese persons.

That, of course, doesn't take any importance to the fact that WHR may be a better parameter to focus on when evaluating the efficacy of any nutritional intervention aimed to improve health and weight loss.

What was very interesting to me was the 'bi-modal' (some people also call it 'bi-phasic') results in women. Normally, when one encouters results like those, it's hard not to think of two or more factors being involved in the behavior of the sample. There is no much discussion about it but it's interesting that younger women had a higher risk of all cause mortality with increase WHR but not so older women. It would be interesting to explore what other factors, maybe not so obvious, are still underlying the data to produce that bi-modal behavior. For example, and of course this is just pure speculation (which I don't like but it helps to illustrate), older women may have a 'good' type of adiposity and by 'good' I mean that say trans-fats were not part of their diet.

Younger women, while having the same degree of adiposity, may have a 'bad' type of adiposity in which trans-fats were part of their diets. Trans-fats may not have been available or not as abundant (the same for processed food), thus older women may not have been in too much contact with them, even though they accumulted a certain amount of adiposity. Younger women may have grown up immerse in an environment were transfats are not only more abundant but present in almost everything, thus affecting them in a way different from just adding adiposity. In this very speculative example, while adiposity may be the same for both groups, a yet unknown underlying factor influenced by the amount of transfats in the diet may be part of their increased mortality. In any case, this is why we read and try to interpret data, right? :) And of course, how come men didn't follow the same trend? :confused: And there goes the theory...:p Remember that my example was just illustrative!

Thanks for the article Lisa!

By the way... did you celebrate Mittsummer? That was my favorite celebration, not just because of the beauty of the Summer in Sweden but also because with exception of the potatoes, the midsummer dinner had nothing but Protein Power kind of food!!!

We'll have our own little celebration next Sunday with some neighbors... hopefully my Gravad Lax will be good enough!