New BMI Calculation

Options
themorethebetter
themorethebetter Posts: 8
in Health and Weight Loss
So I guess most people on here will probably know this being health gurus and all, but in January these researchers made the BMI scale more accurate.
http://people.maths.ox.ac.uk/trefethen/bmi_calc.html

now why is that more accurate? The old formula is weight(kg)/height(m)^2 or 703*weight(lb)/height(in)^2. However Dr Nick Hale, Professor of Numerical Analysis (whatever that is!) at Oxford, points out that we live in not a two-dimensional (as the ^2 suggests) but three-dimensional world. He notes that "It has been known for a long time that people don't scale in a perfectly linear fashion as they grow". His new formula is

1.3*weight(kg)/height(m)^2.5 = 5734*weight(lb)/height(in)^2.5

He himself still admits that he's no doctor and that the BMI system is very general at best. Anyways, who is going with this new updated version, and who is sticking with the old one? (I have a guess your choice might have something to do with your height!)

Replies

  • mmcdonald700
    mmcdonald700 Posts: 116 Member
    Options
    Interesting, hadn't heard of a new calculation. I'm in a healthy weight range and this "new" bmi is 0.6 lower than my other one. The healthy weight range is 2-3 lbs higher at the lowest and highest weights. I consider myself to have kind of an average body comp (not super muscular, not super high body fat %) so I guess that's why mine doesn't change much and I would expect at least that much error in the bmi anyway. I don't think it makes a huge difference in general and I still think body fat %, measurements and the scale in combination are better numbers to look at - I guess BMI is a good general tool for seeing if you're over or under weight.
  • concordancia
    concordancia Posts: 5,320 Member
    Options
    If BMI has a formula, it is because it was derived from data points. The I in BMI is "index": it was derived from measuring a bunch of people of all heights and weights and determining their body fat % to come up with averages.
  • Francl27
    Francl27 Posts: 26,371 Member
    Options
    I don't understand what the difference is, concretely?

    I mean, my gripe with BMI is that people with a lot of muscle can show as overweight, when they're not. Now it's even worse.
  • snappingosprey
    snappingosprey Posts: 28 Member
    Options
    My "new" BMI is only 0.10 less than the "old" BMI

    Entering a weight in the "healthy" range (not quite in the healthy range yet lol) only showed a difference of 0.09 between the old and new BMI calculations.

    Somehow I don't see this new calculation catching on
  • bennettinfinity
    bennettinfinity Posts: 865 Member
    Options
    If BMI has a formula, it is because it was derived from data points. The I in BMI is "index": it was derived from measuring a bunch of people of all heights and weights and determining their body fat % to come up with averages.

    So for average people of average height, BMI tends to be a fairly good approximation of healthy BF%, but if you set aside that data points used to calculate the average were collected in the 19th century (people are generally taller in the 21st), the current formula ignores that fact that people are 3-dimensional.

    Picture an average person (height/weight); that person will be right in the middle of the 'healthy' BMI range. Because the 3rd dimension is ignored, If you scale that person down (keeping all proportions equal), the BMI number for that person will drop - conversely, if scaled up the BMI will increase - even though BF% remains the same in all cases.

    It's generally known that shorter (than average) people can simultaneously maintain a healthy BMI and an unhealthy BF%. On the flip side a taller person can maintain a healthy BF% and an Overweight/Obese BMI.

    BMI is decent at indicating a generalized healthy metric for *populations*, but is terrible when applied to individuals. Any metric that is guaranteed to label a bodybuilder or NFL linebacker as obese probably shouldn't be considered the gold standard.

Categories