How accurate are InBody scans?

chromilo

I get measured using the InBody scan weigh scales that get you to step on the metal plates barefooted and hold on to these metal bars. What has your experience been interpreting the results and its accuracy? My recent one is attached. Thanks

livingleanlivingclean

In my experience, not very.
I had one done a couple of weeks after competing, and it spat out a result of 5% bf. I was lean, but that was ridiculous... It made me question every other result that comes out of those machines.

PowerliftingMom

I think the most accurate is supposed to be the Bod Pod

chromilo

I thought I attached scan. Here it is again. I am comparing it to a bf assessment done last July 2017 using harpenden skinfold caliper method and the lean mass and bf % are not too far apart. Mind you stepping on a weigh scale is easier and less invasive so perhaps InBody scan is just as accurate? Hard to tell as the caliper assessment was 9 months ago.

AnvilHead

https://weightology.net/the-pitfalls-of-bodyfat-measurement-part-4-bioelectrical-impedance-bia/

sijomial

Some commercial grade BIA devices such as InBody or BodyTrax are a lot more sophisticated than home devices that have such a poor reputation. Using whole body analysis as opposed to just hand or just feet sensors is also a big plus point.
However, they are still susceptible to the same pitfalls due to them measuring electrical resistance. A recent meal, recent exercise, variations in hydration are all going to have an impact.

My personal view is that they may provide a reasonable trend and on average a reasonable estimate when used regularly under consistent conditions but can't be used as a one off measurement. My gym has a BodyTrax unit free for members to use and overall it seems reasonable, mostly fairly consistent numbers but with some impossible anomalies on individual scans.

PS - caliper accuracy can be very varied too depending on number of sites used and the skill/experience of the person wielding them.

Azdak

We have an InBody 570 that I have used on myself and with our clients for the past 3 years. So I have a decent amount of experience with the machine.

I have also tried to research any studies on InBody machines, but either I suck at search terms or there isn’t much there. From what little I have found, I think the InBody tends to underestimate body fat by a couple of percentage points.

However, my experience has also been that, if done under the right conditions, test-retest reliability is excellent and the machine can detect small changes in fat and lean mass.

The key is strict consistency of testing conditions. Ideally that means same time of day, same general eating, drinking, and sleeping patterns, no exercise for at least 24 HRs before, no food or any beverages (except water) 2 hours before the test, and no drinking a volume of liquid of any kind within 30 min or so of the test.

Under those conditions, I have found that the InBody can pick up relatively small changes, and the numbers I saw matched other observed changes (circumference measurements, appearance, clothes fitting, etc).

The one episode that impressed me the most was with a client who was reasonably fit, but who had not lifted weights or run on anything but a treadmill for a couple of years. The client lifted weights one day and also ran 3 miles on an indoor track. I had just measured him the day before. The day after the workout, clients weight went up 4 pounds. For fun, I had him come in 2 days later and we redid the test—it showed 4lbs of water weight gain. After a couple weeks we tested again and it was gone.

I recently was approached by another member who was disappointed because, after 2 years of working out, she did a repeat assessment and had not gained any muscle. I looked up her results and saw that everything had remained almost exactly the same—body water, muscle—but that she had gained 5 lbs of fat. (Can’t out-exercise a bad diet). Again, the numbers matched her overall appearance and reflected her diet and exercise program with good accuracy, IMO.

SO, yeah, one offs can be tricky, and so is comparing bio-impedance with other methods. But, overall, given the cost and the convenience, I have come to see our InBody as a very good tool that provides a lot of useful data.

chromilo

sijomial wrote: »

Some commercial grade BIA devices such as InBody or BodyTrax are a lot more sophisticated than home devices that have such a poor reputation. Using whole body analysis as opposed to just hand or just feet sensors is also a big plus point.
However, they are still susceptible to the same pitfalls due to them measuring electrical resistance. A recent meal, recent exercise, variations in hydration are all going to have an impact.

My personal view is that they may provide a reasonable trend and on average a reasonable estimate when used regularly under consistent conditions but can't be used as a one off measurement. My gym has a BodyTrax unit free for members to use and overall it seems reasonable, mostly fairly consistent numbers but with some impossible anomalies on individual scans.

PS - caliper accuracy can be very varied too depending on number of sites used and the skill/experience of the person wielding them.

The caliper body assessment I had done used a 7 Site system. I had to strip down to my boxers for this and was not quick so I hope to not do this again for some time. I also used a tanita body composition analyzer Tbf-300 to compare but I think I have to get measured under consistent conditions like you said, especially when using BIA devices.

giantrobot_powerlifting

The only completely-ish accurate body fat measurement is to go under the scalpel and have the body fat cut off your dead *kitten* and weighed. Other than that, all forms are inaccurate and are estimates, and as long as you are consistent with the tool you use, the differences will remain constant.

I just do a three-point skin fold because A: Cheap; B: Convenient.

chromilo

I just had another scan done using InBody under similar conditions from before, same hydration levels, and the results appear close to the one done in March
Looks promising.