Poor man’s BMI

I admit it; I perseverate about body weight but then, I used to be morbidly obese. Accordingly, Eric sent me Measuring tape trumps a scale. The article is based on the study published in today’s issue of The Lancet (paid subscription required). Apparently, Dr. Salim Yusuf, director of cardiology at McMaster University has found that BMI or Body Mass Index is a poor indicator of obesity.

Waist-to-hip ratio, not body mass index (BMI), is the best obesity measure for assessing a person’s risk of heart attack, conclude authors of a study published this week. If obesity is redefined using waist-to-hip ratio instead of BMI, the proportion of people at risk of heart attack increases threefold. Heart-attack patients had a strikingly higher waist-to-hip ratio than people in a control group, irrespective of other cardiovascular risk factors.

According to the article, you’d calculate your waist-to-hip ratio by measuring the fullest part of your hip, the narrowest portion of your waist and then divide your waist measure by your hip measurement. The lower the number, the better.

While Dr. Yusuf says BMI leads to undercalculation of risk, a related article One Number Doesn’t Fit All states:

Other research has found that BMI often overestimates risk. African Americans and others of African descent as well as Polynesians generally have less body fat and more lean muscle mass than do people of European descent. For this reason, some scientists believe that for these groups a BMI of 26 should be the threshold for overweight and 32 for obesity.

Perhaps WHR will be better but my testing results were mixed. Dividing my 26″ waist by my 36″ hip (yeah, I’ve gained 10 lbs for winter, I was too cold last year) that gives me .72 WHR. While that number is good, if I compare my usual waist/hip (25/34), that gives me a .74 and I’d hardly think I’m at greater risk of heart attack when weighing my usual 106 lbs, especially when my BMI calculation puts me in the underweight category (17.6). You be the judge; drag out your measuring tape and you can find a BMI calculator here.

Get New Posts by Email


  1. jeanette says:

    Unfortunately the waist:hip ratio is mainly based on the male form, as is most medical research (the theorhetical 70kilo male). As you may know the male hip and waist are usually very similar numbers with an ideal of 1.0 for the ratio. For women (if my memory serves me correctly) 0.75 is the preferred number for women.
    BMI is really only useful for white adults. It also is not a great number to use for pregnant people, body builders and children.

  2. Marie-Christine says:

    No, that’s not unfortunate, it’s well known: waist/hip ratio should be no more than 1.0 for men, and 0.8 for women.
    A math refresher: multiply your hip measurement by the desired ratio to find your max waist size..

  3. LizPf says:

    I agree that BMI is close to useless, it’s just a simplified version of the life insurance company weight tables from the 1960s.

    But the WHR is suspect too.

    Women come in different shapes. I have a friend who is seriously pear-shaped. I haven’t measured her but she may have a 32″ waist and 48″ hips. Her WRH is 0.67 … but I doubt she’s in better health than a thin petite young woman with a waist of 23″ and hips 30″ (WHR 0.77).

    For health reasons, I like body fat percentages best.

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.