If BMI is such a flawed system, then why are obesity statistics based on it?.?

They say 60% of Americans are obese, yet this is according to BMI. A BMI of over 25 is classified as overweight and a BMI of over 30 is obese. Yet I know plenty of people with BMI's over 25 and 30 who have healthy levels of body fat. I do realise on the other hand, there may be people with a healthy BMI, yet still be carrying to much body fat and this may balance figures out somewhat, but I get the feeling that the 60% figure may well be somewhat exaggerated. I'd like to see doctors start checking patients body fat percentages and see how this affects the obesity statistics. It would be interesting to know whether the number would stay the same or go up or down. Yes, I've heard waist measurement and waist/hip ratio is a much better predictor of obesity than BMI. Very true Cara, especially Asian people.

Public Comments

1. I think because it's some type of base to go by. It might be flawed, but at least it gives a range to help as a guideline
2. you have a very good point.
3. there have been studies that have shown a stronger link between waist size and obesity related illnesses than BMI. A larger waist size is almost always due to an increase in visceral fat, whereas an increase in BMI could be gained by putting on muscle. Now we all know that someone who adds twenty pounds of muscle is generally heathier than someone who puts on twenty pounds of fat.
4. This is because it is convenient enough to be used to estimate statistics for large groups. Percentage body fat is infinitely more accurate but it would be extremely difficult to use to measure the statistics of a large population. Here is an article about measuring body composition: http://www.spartafit.com/articles/measuring%20body%20composition.php
5. I agree. Also, although it doesn't affect obesity statistics, certain groups of people are incorrectly considered underweight. People that are naturally thin or have a small frame are "underweight" even though many are healthy. Age and gender also play into BMI's flaws.
6. because it is easy to measure and non invasive that is why it is used. i think a bmi of 30 or under is more realistic than 25. the bmi does not take into account muscle mass. its flawed in many ways but there is no doubt america has a real problem with obesity and we are bringing up the rear.
7. it's easier for a Dr. to say that your BMI is too high vs. saying that your body fat is too high. Dr's don't like to use the word fat because people are so dam sensitive these days. when I was growing up fat people where fat and skinny people where skinny. when it concernts the health sometimes it's not beneficial to "sugar-coat" things.
8. It's widely used because it is easy to calculate and the required data for it is easily obtained when performing studies. Even with its flaws it is still useful for giving a general picture of things.
9. BMI replaced height/weight tables, when it was eventually recognised that they were meaningless. These tables were introduced by insurance firms (originally Metropolitan Life in New York) in order to make different categories for life insurance. Weight-for-height was a simple way to divide the population into categories; the assumption was that 'fatter' people would die sooner, so they would be given poorer 'odds' (bigger premium). Healthier thin people were given better odds (lower premium), so that they would be eager to 'bet' (buy a policy) Because the insurance industry is a form of bookmaking, the categories don't need to be 'real' (eg fantasy football). You only need categories so that one can be 'backed' against another. Bookmakers (and insurers) make their profit from the difference in the amounts gambled on different results; who wins only matters to the gamblers, because a bookie (or insurer) won't lose either way. (The common idea that a bookie loses money when an 'outsider' wins is a MYTH. It is sometimes also thought that the bookie loses out when a 'favourite' wins; also a myth. The odds are generated BY the betting (NOT by form or circumstances); if they are accurately laid, then a small profit arises from EVERY penny that has been bet.) The medical system has reorganised itself massively during the 20th Century. When population obesity was being classified as a medical matter, the Metropolitan Life tables were seen by medical authorities as a convenient 'standard' to use. It was assumed that they were valid (but not accurate enough); however, the weight of medical evidence eventually made it obvious that they were complete bollocks. BMI has had a similar life cycle. It is clearly just as inaccurate as height/weight tables (being calculated from height and weight). The only difference is in the boundaries between the categories. The more available bodyfat analysis becomes (electronic body composition analysis, or the more accurate skinfold measurement) the more obvious the system's flaws appear. The adverse evidence has not yet reached the point where a different system is neededd to avoid embarassment to the medical authorities, but it can't be far off. Some sports teams have recently abandoned BMI as a training & selection tool, because it classifies their most muscular players as hopelessly unfit to play.
10. Actually, there is new research that questions whether the BMI can predict the health of people. Now, the issue is physical fitness (as in cardio fitness) and not BMI as an indication of health. Also, statistics can be used in unethical ways by people wishing to influence others. You have to be educated and read between the lines when it comes to the use of statistics. Generally, ask yourself (1) Who commissioned this study? (2) What was the purpose of this study? (3) How large was the sample population? (4) What was the methodology used to obtain the data? (5) How did researchers use this data to make a statistical inference about the population under study?. If you have know idea as to how these numbers were obtained and for what purpose, then you cannot be certain of its reliability as a source. There is so much conflicting research on obesity out there today that it is hard to take any of it seriously. In my opinion, the best indicator of health really is cardio fitness tests - how long a person can stay on a treadmill and probably not pass out. And thinness is not a realiable "proxy" for health. I know of obese people (fitting the definition of BMI>30) who are aerobics instructors in gyms (both men and women), there are also football players, and regular people who get plenty of physical exercise and would be considered obese by the stats. And, there are plenty of skinny people considered normal weight by BMI standards, like my brother whose health habits include pizza, fried n'fatty foods, smoking cigars, binge drinking, and sleeping all day rather than exercise. My brother's health habits should be considered high risk factors for hypertension, high cholesterol, stroke, or lung disease regardless of his skinny frame. You really have to watch yourself in making sweeping generalizations about people's health from questionable stats in the media. That is just another way of the media duping people who refuse to critically think about what is being stated.