Doctors suggest that the ‘overweight’ need help, despite being at the lowest risk of death

The standard way of assessing body weight in Europe and the USA is the body mass index (BMI) which is calculated by dividing someone’s weight in kg by the square of their height in metres. It is generally accepted that BMIs of 18.5-24.9 are ‘healthy’, and that those of 25.0-29.9 mean someone is ‘overweight’, with the implicit suggested here being that they need to lose weight if they are to optimise their health.

Previously on this site, like here I have questioned the basis for these BMI categories. To begin with, for any individual the BMI is a quite useless gauge of their health status. Why? Because it takes absolutely no account of the composition of the body. It is therefore possible for a very healthy individual packing plenty of muscle and very little fat to end up being classified as ‘overweight’. It is also possible for someone with relatively little muscle mass and plenty of fat to be classified as ‘healthy’. Also, even if someone is carry excess fat, where it is found in the body appears to determine it’s likely impact on health (weight around the midriff appears to be particularly toxic). The BMI takes no account of this either.

Even when we forget individuals and focus on populations, there appear to be good reasons for querying the validity of the BMI as it is traditionally used. Proponents say that being ‘overweight’ increases the risk of certain conditions including heart disease and type 2 diabetes. These and other conditions that are more common in the overweight are important, but focusing on them can narrow our focus somewhat.

Some argue, and I would be one of them, that when judging the effect of any lifestyle factor as health, we need to take as broad a look as possible. A risk factor for one condition might be protective for another. For example, sunlight exposure may induce skin cancer, but the evidence appears to show it can reduce the risk of many, many other forms of cancer. It is also associated with a reduced risk of other conditions including cardiovascular disease and multiple sclerosis. Focusing on skin cancer can therefore give us a very skewed impression of the overall impact sunlight has on health.

An article published online in the journal Circulation [1] this week provides an assessment of the impact of being overweight on health and risk of death. It is an official statement from doctors and scientists acting for the American Heart Association. Some of the paper focuses on the apparent ability of being overweight has to increase the risk of conditions such as heart disease, type 2 diabetes and gallbladder disease. This is used to justify the BMI classification. Plus, the authors also take the line that those who are overweight run the risk of becoming obese, where there are even greater risks for health.

However, as I said, it makes sense to take as wide a view as possible in these matters. And one way to do this is to assess the impact of a lifestyle factor on overall risk of death. When we do this with the BMI, we find quite of lot of evidence which shows that, overall, those in the ‘overweight’ category have a lower risk of death than those in the ‘healthy’ category [2,3]. Notably, one of these studies [2] found that overweight individuals were at a statistically significant reduced risk of risk of deaths not related to cardiovascular disease or cancer compared to those of healthy weight.

The authors of this week’s paper in Circulation discuss the evidence pointing to the benefits of being overweight, and they also point to the potential deficiencies of these studies. Epidemiological studies such as these look at associations between things, and can’t be used to conclude beyond a shadow of a doubt that being ‘overweight’ is healthier than being of a ‘healthy’ weight. However, it looks like the best evidence we have does indeed suggest that this might be the case.

The authors conclude, however, that effective weight maintenance and obesity-prevention strategies be developed and implemented for all individuals above normal weight. So, the suggestion here seems to be that ALL overweight individuals (who appear to have the lowest risk of death of all, and some would argue, therefore, the best health overall too) need active intervention. This, to me, seems to be an example of treating a problem that doesn’t exist. We have no way of telling if these individuals are going to gain unhealthy weight or not. Why should ALL of them be subjected to some weight-related intervention? Also, what is the evidence for there being any benefit to such an approach, in terms of health, mortality and cost-effectiveness? The authors do not present any.

What seems to be going on here is an attempt by some members of the medical and scientific community to persuade us that people who are well need to be treated as though they are not. I don’t suppose that it will as too much of a surprise to learn that almost all of the authors of this paper have received research grants, speaking fees or consultancy fees by companies that supposedly have weight-related solutions at hand (including food companies, drug companies and Weight Watchers International).

In their conclusions, the authors state that ‘debating the relationship between BMI in this [overweight] range and total mortality misses broader implications’. I’m left wondering what broader implications can be more important than the evidence linking being ‘overweight’ with lowest risk of death.


1. Lewis CE, et a. Mortality outcomes, and body mass index in the overweight range. A science advisory from the American Heart Association. Circulation, 8 June 2009 [epub ahead of print publication]

2. Flegal KM, et al. Cause-Specific Excess Deaths Associated With Underweight, Overweight, and Obesity. JAMA. 2007;298(17):2028-2037.

3. Adams KF, et al. Overweight, obesity, and mortality in a large prospective cohort of persons aged 50 to 71 years old. N Engl J Med. 2006;355:763-778

6 Responses to Doctors suggest that the ‘overweight’ need help, despite being at the lowest risk of death

  1. Tom 13 June 2009 at 5:41 pm #

    Despite our best efforts, human mortality remains at 100%. Lowering the mortality from one cause absolutely has to raise it somewhere else.

  2. Dr John Briffa 15 June 2009 at 4:46 pm #


    ‘Risk of mortality’ refers to risk of death over a defined period of time. No, of course you can’t put off death inevitably, but it is possible to delay it.

  3. Derek 16 June 2009 at 1:45 am #

    I am of medium height but much broader in the shoulder than normal. I have large hands and wrists and a ‘normal’ watch will not fit me. I have calf muscles over 45cm in circumference. I admit that I have a ‘middle age spread’ too, but my niece has about the same height as me and is built like Olive Oyl. I have often wondered how the same (one and only) BMI table can apply to both of us. Surely this table should at least come in ‘Skinny’, ‘Normal’ and ‘Large’ variations to be of practical and non-depressive use.

  4. Hilda Glickman 16 June 2009 at 2:28 am #

    Tom. What a silly comment. Should we try not try lower premature death, then. When some people try to help other become healthier others just make facetious comments!

  5. M. Cawdery 17 June 2009 at 12:03 am #

    There was a prior report by Flegal
    JAMA. 2005 Apr 20;293(15):1861-7.
    Excess deaths associated with underweight, overweight, and obesity.
    Flegal KM, Graubard BI, Williamson DF, Gail MH.
    National Center for Health Statistics, Centers for Disease Control and Prevention, Hyattsville, Md 20782, USA.
    which was preceded by another report which claimed the reverse. The report cited above had to be written because the original had made arithmetical errors

    They have been trying to talk their way out this FACT ever since.

    Also There are reports that the obese actually survive hospitilalisation and surgery better than normal weight individuals.
    These reports were based on hospital records!

    In short if it is Medical Correctness (MC) you want, listen to Big Pharma, the “x-spurts” and the KOLs (Key Opinion Leader). If you want the truth search the internet!

  6. Trinkwasser 19 June 2009 at 12:31 am #

    “Some of the paper focuses on the apparent ability of being overweight has to increase the risk of conditions such as heart disease, type 2 diabetes and gallbladder disease.”

    IMO they got that the wrong way round, people prone to “metabolic syndrome” tend also to put on weight. It’s the insulin resistance, and probablly leptin/resistance.

    Having said which our family includes what Michael Eades calls “metabolically obese” people like me, who are skinny but have all the markers of obesity (probably omental/visceral fat rather than visually obvious subcutaneous fat) and here it definitely works backwards: my plump aunt was told at eighty she had the blood pressure of a thirty year old, whereas skinny old me and others were the exact opposite, I had the BP of an eighty year old when I was thirty and my tiny mother has had chronic high BP for about 50 years.

    So trying to fit everyone into a canonical weight range is doomed by genetic and other factors. Probably changes from *your* normal might be indicative of something, I lost nearly ten kilos when I had (undiagnosed) gallstones and the only time I gained weight was from eating too many dietician-approved carbs. But again people with different gene sets appear to gain and lose weight almost at random, including in some cases monthly, oh and Rugby players boxers and athletes in and out of season.

Leave a Reply