General fitness, health and nutrition · Public discussion

What good is BMI?

Started by Joe User · · Last activity · 4 posts · 523 views

Thread details

What we know about this thread

Original section
General fitness, health and nutrition
Published
1 February 2004
Last activity
2 February 2004
Original author
Joe User
Posts
4
Discussion status
Public discussion
Total views
523
Views / 30 days
0

The navigation and discussion metadata provide context. Posts remain in their original chronological order.

Showing posts 1–4 of 4
Posts remain in their original chronological order.

Text size
  1. Searching past threads, there seems to be general agreement (including my own) that BMI is not a
    measure of body fat composition.

    So what good is BMI? And why do some reputable books and articles still suggest using BMI as a
    measure of "fitness"?

    As we know, BMI is weight divided by square height in kg/m^2. The interpretation of BMI is the same
    for both male or female; and it does not take "build", age, or even some subjective notion of
    fitness into account. For example, see nhlbisupport.combmicalc.htm from the US NIH.
    "Normal" (whatever that means) is BMI 18.5 to 24.9. Period.

    How could that be the same for male and female? Isn't it common scientific knowledge that all things
    being equal (height, weight, age, fitness level, etc), females generally have higher percent body
    fat than males, on average?

    Of course, there are exceptions. That is why statistical samples yield some kind of bell-shaped
    curve. But the NIH's own percent body fat tables show that for age 20-24, for example, "optimal"
    %fat is 18.9-22.1% for women, whereas it is 10.8-14.9% for men.

    (The "margin of error" is reportedly +/-3 percentage points. Dontcha just love it when statistics
    are reported to a greater degree of accuracy than the standard error?)

    Of course, the correlation between BMI and labels like "underweight", "normal", "overweight", and
    "obese" is based on statistical studies and linear regression (of some other method of statistical
    correlation). I can only assume that the standard error is so great that it overrides any difference
    due to gender.

    Does anyone know the "margin of error" for the BMI? Can someone point me to a report of the
    statistical study? Just how old is BMI, anyway?

    My purpose in making those comments is not to write a provocative manifesto, but to highlight some
    context to why I question the usefulness of BMI as an individual.

    My conclusion is that BMI is virtually useless -- except perhaps for its original purpose, which I
    believe was to simplify motality tables for insurance purposes.

    (Right? But IMHO, even that is suspect because BMI fails to take "fitness" into account.)

    I welcome knowledgable opinions on the subject.

  2. 31 Jan 2004 16:54:54 -0800 in article
    <[email hidden]> [email hidden] (Joe

    User) said:

    Searching past threads, there seems to be general agreement (including my own) that BMI is not a
    measure of body fat composition.

    So what good is BMI? And why do some reputable books and articles still suggest using BMI as a
    measure of "fitness"?


    BMI gives reasonable results for average built average height individuals. For very tall and very
    short people it gives wrong results. Muscular individuals also get wrong results. For example
    practically all NHL ice hockey players have BMI > 25, which classifies them as overweight, although
    their body fat percentage is usually around 6%.

    --
    Matti Narkia

  3. Once upon a time, our fellow Matti Narkia rambled on about "Re: What good is BMI?." Our champion De-
    Medicalizing in sci.med.nutrition retorts, thusly ...

    Quoted message said:

    BMI gives reasonable results for average built average height individuals. For very tall and very
    short people it gives wrong results.

    Well Mr. Science Geek, why don't YOU geeks change the formula to account for height and body build?

  4. This link, off the url you poosted, clarifies the issues you raise.

    nhlbi.nih.govrisk.htm

    The ranges of "normal" etc. are so broad as to take into account gender and other such normal
    differences. The case of body builders etc. is different in that they are indeed on the tail of the
    bell curve of the populations, they are abnormal in that regard. Fitness and body fat don't equate.
    I n the old days of soviet olympic teams their power lifters had great bellies but most likely could
    not be faulted for fitness. Fitness and body weight are only indirectly related. For purposes of
    population based research the bmi does a very good job as an independent variable and/or must be
    normalized with other variables to isolate the item one is trying to isolate in the research
    question. It describes the great bulk of folk who in the real world reside in the hump of the bell
    curve, not the abnormal situations of fashion models, body builders, etc. whose numbrs are so small
    as to mean nothing.

Active in the last 60 minutes

Active in this thread

0 users · 0 guests ·0 bots ·0 total

No signed-in users are active right now.

No known search crawlers active right now.