Ep 3. Body Composition Testing

It's important to know your body fat percentage, right?!

In this episode, we explore why all existing body composition methods fail to accurately give individual results. 

The source for much of the episode is this article by James Kreiger
https://weightology.net/the-pitfalls-of-body-fat-measurement-part-1/

Transcript

Copy Transcript Copied

Hi Bros and hoes and everyone else. This is Gay Bro Shit, a podcast for deep thoughts about shallow issues.

And you can call me Brosephine Baker, because I too was better received in Paris. 

Today is all about body composition testing and wether or not we can accurately assess how fat or lean we might be. 

So weight is often the single metric most people use to define fatness or leaness but of course its problematic because it doesn't really tell us much other than just literally how much mass someone has on that day according to that scale. 

Your weight includes all of your muscle all if your bones, all of your organs, your blood, the water stored in your muscles and organs, any food You've consumed, any waste products your building up, and all of you are microbiome. And so just stepping on a scale isn't going to tell you what that weight is actually made up of. 

Body composition is just the term we use to describe the ratio or percentage of your body that is stored fat tissue versus all other types of tissue and fluids and what not. 

So we use body fat percentage as the primary number for how lean someone is. 

While the scale goes from 1 to 100, realistically the leanest someone can be is Is about 4% body fat- and when you see someone who is this lean they kind of look really gross. Like you can see all of the little strands of muscle and it looks like that bodies exhibit where they show people who've been basically flayed and you can see

all the red muscle fibers. 

On the extreme high end, you can be up to 80% fat- but these are outliers. 

The average American male is somewhere around 22% body fat in their late teens and early 20s and then it slowly increases up to an average of about 30% in your 80s. But this is the statistical mean so even if your body is bigger or smaller than this mean you still probably have an average body just somewhere along the distribution scale.

In terms of societal expectations the "ideal" body fat range is usually between 10% and like 18% for men and obviously people idealize men outside of this range but this is the ideal as depicted in popular media. 

The phrase 'dad bod' usually gets applied to people closer to 18% or even 20% which is a bit wild when the average dad is probably more like 25-30%. I saw one video where these three girls on the street somewhere saw a picture of the reigning Mr. Olympia classic Chris Bumstead- so basically the world champion for an asthhetic body- was described as having a dad bod with his off season weight and he's maybe like 12% body fat but probably less 

So let's save that term for people who actually have a bit of a belly on them. Let's keep our wildly unrealistic standards at least within the realm of possibility

Now finding out your body fat percentage is a fairly tricky thing as even the best methods have a degree of error of 2% or more percentage points across a population and up to 10% on an individual level.  

Going into this I felt I had a pretty good grasp on body composition testing and how current methods are limited Because I’ve been working in fitness since 2012- but I had an experience that sent me down the rabbit hole and led me to really digging into the research behind each of these methods and how good they might be-

Spoiler alert: none of them are good. 

A lot of what I’ll be telling you today was sourced from an article by a researcher named James Kreiger out of Washington state who did a really good job pulling data from multiple sources- and who also has a lot of experience doing body composition testing for corporate wellness clients.  There will be a link to this article in the show notes so just giving credit to him and his team for collating this information and summarizing it in a really impactful way. 

https://weightology.net/the-pitfalls-of-body-fat-measurement-part-1/

It turns out there is only one way to get a truly accurate measurement of body composition- and in order to do it you have to be already dead- because basically they totally dissect a body and carefully separate out all the component tissue and then measure that. So you can only do that once- and you definitely won't be having an open casket funeral afterward. 

So because we can't do that to people who are still alive and want to stay that way we have to use other methods. 

Scientists who study this then use what is called a Four-compartment model- where in they use one test to find out your bone slash mineral mass, one for body density, and another for measuring hydration of the body.  The numbers are then plugged into a formula that gives us a reasonable estimate of body composition within about 2%. So this test would say you’re 20% body fat and most likely you’re somewhere between 18 and 22, with 20 being the center of this range. 

 And because these all need to be done one after the other to get accurate results, it's very expensive and generally only available in a lab setting for research purposes.  So when we compare the accuracy of any body composition testing we are comparing it to this four-compartment number. But this too is just an estimate, an informed one, but still an estimate.  So we’re comparing the accuracy of something that is also not completely accurate so we get compounding errors. But for the purposes of clarity and simplicity, we are going to say the four-compartment method is accurate, especially at the population level. 

So we also need to clarify how something could be accurate on a population level but not on an individual level- 

The example is if you have a test method and half the people who get tested it measures out as 10% less than what they really are, and the other half measure out to 10% more than what they really are- that still averages out to 0% average error rate meaning for the whole group is completely accurate but for each individual its still off by 10%. And this is going to be a recurring theme here. 

So let’s cover all of the methods of body composition testing in the real world

Hydrostatic weighing

For a long time, and still to many people this considered the gold standard for determining body composition. 

If you’re not familiar sometimes people refer to this as the dunk tank and its definitely not something most people have access to as you find these at medical schools and other reserach instituions. 

The gist of this method is that they weigh you out of water, and then you get totally submerged and they weight you again underwater. Because fat floats the change in weight will tell you about how much fat mass you have. and they also measure the volume of water you displace, so the goal is to determine your body volume and body density. 

They also have you blow out all of your air and hold your breath underwater- so its probably going to be an uncomfortable process for most people. 

So one source of error is that you may still have air inside your lungs, but you also have air inside your digestive tract or even trapped in your hair so you might float a bit more or less depending on how much gas is inside you. 

The blogger source of error is that in order to determine your body fat percentage they rely on the assumption that fat free mass has a set density- but, muscle density is not the same for every person, its also not the same for people of different racial backgrounds, for example people of african descent have up to 10% denser muscles. But because your density can also be changed by hydration levels, this means the formula gts thrown off. 

Related note is that the easiest way to change your scale weight rapidly is to change your hydration levels- which is why people who have to cut weight for sports with weight classes like wrestling or combat sports usually do this by severely dehydrating yourself. So you can wind up with a much lower scale weight even though your fat mass is still the same. Its not unheard of people to lose like 20 or more pounds in a few days by just drastically cutting water. And in a lot of extreme diets with rapid weight loss this is what the scale is actually showing. 

So while on average this method is accurate within a few percentage points for any given individual the error margin is 6%. 

Visual

This is where you look in the mirror or take photographs and give an estimate on what your fat accumulation looks like and or what muscles are visible. If you go on your preferred search engine and do a image search for body fat percentage you’ll see a number of photo montgaes that show what people are supposed to look like at a certain body fat percentage and then you compare yourself to that. 

It should be no surprise that this method is wildly speculative because 

a. We have to assume the people in the photos are actually that exact body fat percentage 

B. not everyone stores fat in the same way or same places, and our muscles all look different so 

C. Were biased- sometimes we want to be overly optimistic and want the number to be lower, or maybe we have body dysmorphia or other negative self image issues and we see an exaggerated image of ourselves 

Sometimes on the internet people will post pictures of themselves and other users will try and give them a body fat estimate based on these images but these are all prone to the same erors. 

Provided you dont have dysmoprhia- You can definitely measure progress visually because you can see how your muscles change in relation to your softer areas, you can see increased muscle definition or how they just look bigger or how your clothes fit differently- so its an ok method for tracking progress but its just a guess. 

Another issue with pretty much all of these methods here is that they usually use gender as an input so were assuming people are cis-gendered when we talk about the accuracy of these results and I cant even imagine how inaccurate they are for people who  are transgednered- because do you use their birth sex or their lived gender- or what about people who are intersex or non-binary or any other variation where the sex characteristics are less easily diferentiated? 

Calipers (Skin-Fold)

Skin-fold measurements involve using a measuring tool that pinches your skin together and measures how big the tissue fold is, hence the name. Usually you do this in a number of spots- between 3 and 7. 

So you add up all the measurements and plug it into an equation with your age and are supposed to get a body fat reading. 

So the first issue with this is that the person administering the test may or may not know what theyre doing- or they might change their technique to get a more favoreable result. You have to pinch the skin in the exact same way everytime and at each site you measure you have to do it in the exact same spot. So if you’re doing this periodically over a period of time evry time you retest you have to find the exact same spot and measure the exact same way to be able to measure changes consistently. Its unlikely you can guarantee this to be the case.

The second issue is that the formula used to get the body fat number is itself based on a formula that is itself an estimation so its an estimation based on an estimation. And the original equation did not account for racial background and the only people tested were white people. They used hydrostatic weighing to determine the body fat and since that method is also error prone theres compounding errors going on here. So individual; error rates are as high as 15% for this method meaning you could be 20% body fat and it would say that youre 35%. 

Biolectrical Impedance

This method sends a weak electrical signal through the body and assumes thst because fat is hydrophobuic- meaning it doesnt contain water it wont transmit electricity as well as the rest of your tissue that does. This is the method you see on bathroom scales or hand held devices. 

With the scales the electrical current goes up from one foot and then through the body and is measured back out the other foot.  With handheld devices it goes through one hand and up the arms and then back down the other arm into the other hand. 

So the problem right there with these devices that only have two points of contact is that the electricity is going to follow the shortest path through the body, so if you step on a scale its goignt to send the electricity up one leg, and then its going to come right back down the other so its not going to go through your torso or upper body at all. If you use the handheld one it skips your lower body entirely. Because these machines need water to transmit the electricity your hydration levels can definitely effect the results. 

There are other machines where you have four points of contact- both hands and feet is this is theoretically more accurate, but probably not in practice because…

Just like with fat claipers the datat they use to determine body fat is obtained by testing people another way- usually hydrostatic weighing and so again its an issue of compounding errors. 

When compared to the 4 compartment model we talked about earlier- The error rate is about 8% in either direction- So someone tih 20% body fat could read as anyhting between 12 and 28%

It’s also not good for tracking change- in one study it was found to be even less accurate than using plane old hateful BMI- the error range again was about 8% so someone could reduce their body fat by 4% and it could say they gained by 4% or vice versa. 

BodPod

This is an egg like capsule that you sit like some weird scifi space travel cocoon and then it measures how much volume you take up by air deplacement. 

It should be as accurate as Hydrostatic weighing but a lot of things can throw it off including body hair, moisture, body temperature and how your swimsuit fits- So individual error rates can be as high as 15%. So this is fairly garbage

3D Scan

This is a device where you stand on a turntable and it shoots laser at you to determine how much volume you take up. It has most of the same issues as the BodPod so it too is not reliable. 

Dexa scanner- gold standard ?

The Dexa scanner is a fullbody x-ray that uses two different wavelengths to determine what type of tissues your body has. You lie on a table and try and hold still for 10 minutes while it shoots beams through you. Because od the dual length x ray beams it can hopefully tell the difference between bone, organs, muscle and fat. HOPEFULLY

In General people have concerns about the use of XRays as they are a from of radiation and in high doses this could be really bad- but the DEXA radiation dose is equal to about one day worth of background radiation (it's about 1000 times less than a full body CT scan).

Many people consider this the most accurate method along with the dunk tank- but- of course like everything else weve talked about it has issues. 

A lot of the issues have to do with the machines themselves as different manufacturers use different formulas, and so the machines can be different model to model and machine to machine. Th software also gets updates from time to time and the machine may or may not be up to date.  The way the xray beams themselves are shot up also makes a difference as they can create whats called a ‘maginification error’ where the result gets skewed because of the angle of the beam

The machine also will get different results based on different hydration levels- so a 5% variation in hnydration can change the body fat reading by as much. The accuracy can also be affected by biological sex, size, fatness degree….Which if the whole point is to accurately measure body fat and it gets skewed with a higher body fat percentages than we really have to ask what the fucking point is? 

In studies the individual error rate compared to the four compartment model is anwyehre from 4 to 10%- so again if someone has 20% body fat the machine might say they have up to 30, and as low as 10%....which is a very very wild spread. These studies are usually done on bodybuilders

Ok this cracked me up- they did a study where they tried to measure how the device measured changed in body composition- and what theu did is they literally strapped lard around the legs of bodybuilders- and, while the machine did measure an increase in fat- it also interepreted a loss in bone density- on the exact same leg on the exact same day.

But I’m just trying to visualize how they strapped lard to peoples legs- like did they spread it on cplastic wrap like peanut butter and then just slap it around with some tape? How did they do this? I need to know!

Anyway, what prompted this research and in fact this entire episode is last week I went and got a Dexa scan.

So I had intended to include it as part of something I’ll be talking about in the next episode- but the rabbit hole gaped for me and I fell in fists first.

So I drove out to a random business park in the suburbs and went to this training facility where they had one of these machines. They had it parked in a van outside because apparently they drive it around to places including the Nike campus...So I got in dark windowless  van with a strange man. Which I never felt unsafe but I imagine maybe if I was a woman or femme or someone other than a muscular cis man I would feel really uncomfortable getting into a van alone with someone and lying down on this machine. 

And this man was a trainer so he spent the ten or so minutes while I was laying there explaining how it worked and what the numbers meant and how I should be strength training and it was clear he wanted to sell me on training out of habit- or was just filling the time- but I didnt interject to say ‘hey actually dude I know all of this and probably a lot more…’ so he kept talking.

And we wrapped up and he told me I would get the results in half an hour- but apparently there was something wrong with the machine so I didnt get the results back until the next day- which I’m not really bothered by now in hindsight makes me more suspicious of the results

And yeah- I got the results back and they were really rough- They had my Body fat as way higher than I thought it should be- and my muscle mass lower than it should be, and my bone density was just average when as someone whos been lifting for 20 years it should definitely have been higher. 

This machine would have me classified as Obese…which doesnt make a ton of sense to me because I have some ab definition and thats not something you see in someone who is actually obese. 

Though of course the term Obese is very problematic because the original definition was just the top 5% of a given population- so by the original definition its impossible for 36% of americans to be Obese when the term should only apply to the top 5%. And thats also flawed because its always based on comparison to others and not how healthy or not that individual actually is. 

To be very honest this completely popped my bubble. All the progress I thought I was making completely disappeared. The places I thought I saw improvements suddenly became flaws and I felt really compelled to restrict my diet because clearly I was still eating too much and not working hard enough

And that was freigtening because I knew better. I actually have a ton of knowledge and experience and have done a shit ton of work on myself to stop falling into these traps and I still found myself falling into one

So I think doing this is potentially really dangerous because I could see the machinations of how eating disorders arise in myself - someone who knows what to watch out for and usually had a healthy self image- so I can't imagine how bad this would be for someone who is more impressionable and hasn't done mindfulness and hasn't had therapy and is just listening to the voices of popular culture that they're not good enough and this machine is confirming it. 

Fortunately I was able to get out of this hole really quickly by digging into that actual research on how accurate these testing methods are- and just going back to the basics of

-Do I see progress in the mirror? Yes

- is my scale weight trending the way I want it to? Yes

- are my circumference measurements trending the way I want? Yes

- do I like the way I feel? Yes 

So I cant say how accurate these results were- but knowing that this machine gets drive around town and our roads are shit- I can only imagine it gets thrown slightly out of alignment every time it goes anywhere…and then the fact there was something wrong with the machine for those 24 hours also makes me suspicious as I said- and now knowing about the lard study and everything else about the individual error rates and I’m going to assume that these results were off by at least a few percentage points and potentially as much as 8% though I suspect more like 4 or 5. 

So for you out there I really want you to just try and put all this aside and stop trying to measure your body against a set standard and just measure it against your standards for your body and what feels healthy and happy for you. 

In Conclusion-

1. There is no reliable way to measure body composition accurately so dont worry what body fat percentage you are or might be
2. None of the available methods accurately track changes over time 
3. If you feel like using a testing method remember it is at best an estimate and assume that the answer you receive is off by at least a few percentage points
4. If you want to track changes just take your weight and circumference measurements and if those are trending in the direction you want (i.e. going down for fat loss, going up for muscle gain) than you are accomplishing your goals. 

Im actually going be somewhat contradictory here and give a soft recommendation for a simple way to get a body composition estimate thats is completely based on your weight and some body measurements. 

This method is called the US Navy method and if you do an internet search you can find some free calculators where you input your numbers and it will give you a rough body fat estimate and also roughly how much lean mass and fat mass you might have. 

Im not recommeding this because I think this number is more accurate- I think it is still ver much an estimate, but I think it might be useful for you 

a.this is a nonivasive way that doesnt involve any equipment other than a scale and measuring tape

B. its measuring variables- weight and neck and waist circumference that do directly correlate to body composition. So if those numbers are changing then your body composition is likely changing as well. 

So dont take this measurement as the truth, but use it to affirm whatever progress you might be making. 

Thanks again to James Kreiger for writing up this research- you can find his website at weightology.net