The Doctor?

Are Pediatricians Hooked on BMI Scores for Kids?

When it comes to BMI, everyone is a hater. Even so, most people are content to keep using it as an easy number to screen adults for obesity. But for kids, everyone knows BMI just doesn’t work. In Pediatrics this month, Bill Dietz  describes “widespread agreement” that BMI scores for kids with severe obesity are inappropriate.

And yet, the question of a replacement measure remains “unresolved.”

BMI Percentiles and z-Scores

Dietz explains why BMI percentiles and z-scores don’t work in severe childhood obesity:

Extrapolation of z-scores or the use of percentiles to assess severe obesity is problematic because large changes in weight and BMI are associated with small changes in BMIz or BMI percentiles.

A new study by David Freedman and Gerald Berenson sparked this commentary. They found that BMI z-scores were actually quite misleading over time for children with severe obesity. For several children in their study, BMI z-scores went down over time while their obesity became more severe.

Sticking with Variations on BMI?

In the face of these problems, Dietz, Freedman, and Berenson all suggest that other variations on BMI might be the answer. Simply use different calculations for kids with severe obesity. Some of the possibilities include absolute BMI, BMI as a percent of the 95th percentile, or BMI as a percent of median BMI for age and sex.

We wonder if this solves the central problem with BMI in children. BMI does a lousy job of representing body fat percentage in kids. A recent study by Peterson et al in JAMA Pediatrics showed this pretty clearly. BMI z-scores misclassify children 20% of the time. A simple alternative measure, TMI, cuts error rates by 75%.

So why are we hanging onto BMI scores for kids? And why do some schools persist in sending home BMI report cards?

Click here for the commentary by Dietz and here for the study by Freedman and Berenson.

The Doctor? Photograph © AJ Schroetlin / flickr

Subscribe by email to follow the accumulating evidence and observations that shape our view of health, obesity, and policy.


September 6, 2017

2 Responses to “Are Pediatricians Hooked on BMI Scores for Kids?”

  1. September 06, 2017 at 11:35 am, Allen Browne said:

    Unfortunately I regard this as a smoke screen hiding the fact we are not taking treatment of children seriously.

    BMI percent over the 95th works well to follow the children with obesity from a clinical point of view.

    True – sending home BMI reports without providing accurate information and resources for treatment is wrong.

    • September 06, 2017 at 12:05 pm, Ted said:

      I agree, Allen. The overarching question should be about providing adequate care for children with obesity. And then when we start doing that, we need adequate measures.