One of the things I hate about obesity are statistics. It's practically impossible to put a figure on how many people are obese, as the definition and measurement techniques are not standardized.

I've discussed BMI and the issues with it before, but two new studies today show how silly trying to drill down to actual numbers can be for those in charge. 

The first study that really cuts to the heart of the matters comes from Australia. Their childhood obesity rates have been steadily rising, but they've found that the wrong criteria was applied. 

It turns out that the study didn't take into account the level of BMI in children taller for their age. Makes sense, right? Taller kids will be larger kids, which means their BMI threshhold will be higher. 

As Richard Telford says:

A taller child is biased towards being classified as being fatter than he or she really is and a shorter child the reverse. Now if that sort of thing is happening its going to affect our statistics so we have to go back to the drawing board and find a new measure that was not biased towards height

The second study shows the fallacy of obesity statistics when you rely on self reported statistics. The Canadian city of British Columbia is trumpeting their obestity statistics. It turns out they have dropped from 12.7% in 2005 to 10.9% in 2007. Althought it doesn't look large, it's 15%. 

Remember, these are self reported statistics with no control nor any proper method of ensuring the same people responded. Heck, you might have gotten the 15% who are thinner. 

Still, they want to celebrate. Mary Polak said:

We're committed to helping British Columbians live healthier,offering the tools that help people make the changes in their lives that can have a tremendous impact on their health.

It's a noble effort, but if you cannot reliably survey and access the correct rates it doesn't matter what you do.

Bookmark/Search this post with:
delicious | digg | reddit | furl | google | yahoo | technorati | icerocket