I’m aware of the accuracy problems in this method such as multi-regression analysis for the Jackson and Pollock density equations applied to a general population. You article brought to light other issues. I use the measurements, in particular the trends in changes when I’m attempting to lost body fat for a contests as a benchmark. I participate in natural bodybuilding for men over 60.

http://gallery.rxmuscle.com/index.php?contest=97&year=94&bodybuilder=6303#4

My bodyfat percent measured 6.5% in the photos using the body composition number from my website. I have found estimating my body composition to be an additional tool in my bodybuilding training. I am not limited to the weight scale and mirror.

- Kirs

Joe: thank you for the clarifiying points. And I don’t think I said DEXA was the best body composition tool, just pointing out that it does avoid some of the issues inherent to calipers. And given the cost of DEXA, it’s still generally unrealistic to use that frequently anyhow.

DEXA does a very good job of assessing bone mineral content, but when bone is present in a DEXA pixel it can not determine fat from lean mass. It can only assess bone from non-bone. It makes assumptions about the tissue surrounding the bones. Since bone is present in approximately 50% of the DEXA image it is making assumptions on up to 50% of the body composition of the subject. Assumptions definitely impact accuracy.

The other component about DEXA to consider is the attenuation strength of the device. The thickness of the subject can significantly impact the devices ability to measure body composition. An obese subject with 8 inches of tissue depth or more on their body will not get an accurate body composition assessment. The strength of the radiation is not high enough to image the soft tissue.

Last, how DEXA validated? Was it validated against hydrodensitometry (under water weighing)? If so then to body density issues discussed above would apply to DEXA. If validated against MRI, there are problems with MRI. MRI looks at slices not the entire body. Assumptions are made from slice to slice.

In reality the best means of assessing body composition is via multi-compartment measuring. To perform multi-compartment testing you need to be able to measure body density (body fat), bone density (DEXA), and total body water (double labeled water which iss assess with a mass spectrometer). DEXA measures bone the best but is not the best body composition tool.

Very hypothetically speaking, let’s say a person had regular access to DEXA testing, would there be a limit in terms of how often this test could safely be done in a given time frame or would that be a non-issue?

I’m also curious as to what you think about Charles Poliquin’s BioSignature course, since it is based so heavily on specific site values from caliper testing. It seems like in this case, mere consistency of testing wouldn’t be enough, since supplement protocols are recommended on the basis of skinfold ratios between specific sites.