Listen on Apple Podcasts | Listen on Spotify | Listen on YouTube

Are all body fat scales equally inaccurate? Are some accurate enough to actually be useful when tracking our body composition? That’s what Dr. Grant Tinsley wanted to test in the latest study out of his lab, and he’s discussing the surprising results in this interview. 

Many fitness enthusiasts want to track their body fat levels and body composition to keep tabs on their progress and ensure their training and diet are on the right track. So it’s no surprise that many have been tempted to buy relatively inexpensive body fat scales and home devices that claim they can assess your body composition just by stepping on a scale or holding special electrodes. I’ve never recommended these devices due to their inaccuracy, but are some bio-electrical impedance analysis (BIA) devices good enough to actually help people keep track of their body composition?

That’s exactly what Dr. Grant Tinsley wanted to find out and what he and his team put to the test in their latest study. Grant and his colleagues bought some of the most popular BIA devices off of Amazon and compared their results with current gold-standard methods and expensive devices to determine how useful these tools really are.

So, how accurate are BIA body fat scales, why do some work better than others, can they actually be useful for tracking your body composition, and how can you get consistent and accurate readings with them? You’re going to find out in this podcast.

In case you’re not familiar with Grant, he’s an Associate Professor at Texas Tech University, the Director of the Energy Balance & Body Composition Laboratory, and is a Certified Strength and Conditioning Specialist (CSCS) and Certified Sports Nutritionist (CISSN). He has over 150 peer-reviewed journal articles and abstracts, with his major research interests being sports nutrition strategies, body composition assessment techniques, and intermittent fasting in active populations. 

So, he’s the perfect guest to talk all about body composition measurement and discuss his new research on BIA devices. Press play to listen to our discussion! 


0:00 – Try Pulse today! Go to and use coupon code MUSCLE to save 20% or get double reward points!

4:30 – What is body composition and how accurately do home devices assess that?

21:09 – Do you recommend these devices?

30:17 – Did you come across any useful devices doing your experiments?

35:04 – What are other methods to accurately calculate body composition at home?

46:23 – What is three dimensional optimal imaging? 

47:37 – What are your thoughts on the expensive machines found at gyms?

51:01 – What is the accuracy of the army body fat equation?

56:31 – What are some controlled factors we can practice for consistent readings?

59:08 – Where can we find you and your work?

Mentioned on the Show:

Try Pulse today! Go to and use coupon code MUSCLE to save 20% or get double reward points!

Grant’s Instagram

Grant’s Website

Pre-registered study

What did you think of this episode? Have anything else to share? Let me know in the comments below!


Mike: Hey Grant, welcome to my podcast. Thank you for taking the time to come educate us on body composition.

Grant: Absolutely. Thanks for having me.

Mike: Yeah. Yeah. And for people who are watching this on YouTube, or I guess there’s video on Spotify as well. It looks like I’m recording this with a potato because I basically am.

This is my laptop. I think this laptop’s five years old. It still works fine, but the camera, this is what that camera looks like. And my normal web camera fancier one doesn’t wanna work today. So this is what we’ve. Anyway, we are here to talk about body composition. I’m excited to have you on the show to talk about this, because this is something that you and your lab specializes in, in your research.

And where we could start is maybe just a quick definition of the term body composition, just for people who aren’t familiar, it’s different than just body weight. And then I would love to hear about this study that I believe you said it’s submitted, so it. I think that’s what you said, but it’s the work of its complete on these.

Buy I don’t want to throw out terms. People won’t understand yet, but on these devices, like scales, handheld devices that you can just buy on Amazon. And how accurate are those really at assessing body composition? Yeah.

Grant: So I’d love to dive into all of that. As you mentioned. I think it’s always good to get terminology straight in the fitness community.

People usually think something when they hear body composition, probably often they’re picturing like a body fat percentage or the percent of your weight, that’s body fat. And that is one of the variables we often look at, but really from a broad perspective with body composition, we’re just talking about how you break up your body weight.

So most people have this understanding that body weight is a metric is not the most informative. Maybe you personally have been the same body weight at different times in your life, but you’ve looked a lot different. You’ve felt a lot different. You’ve been able. Warm a lot different. So really body composition, just broad principle.

How do we break up the body? At a simple level, most of the methods we have access to will be doing something like trying to split all of our body mass into fat mass. Fat free mass, which would just be everything else, but we can build more advanced models. We can look at things from more of an anatomical perspective where we look at true skeletal, muscle ATIP, post tissue, bone, things like that.

So it can get fairly complicated, especially in a research setting where we do have access to devices that can slice up the body in different ways, so to speak. But in terms of consumers, as you mentioned, many of these simpler device, Really are giving you a body fat percentage, which again is just related to the amount of fat you have fat mass and the amount of fat free mass you have.

So yeah, if you’d I’m happy to go ahead and dive in and elaborate a little bit more on our body fat scale study. Yeah, that’d be great. Yeah. So as you mentioned, we haven’t quite submitted it. If we were recording this in another week or two, it may be submitted, but we do have the research done.

We’re just finalizing a few parts of the manuscript. We have published a good chunk of this in abstract form at a few conferences over the last year or so from the outset. And I’ll probably say her name again. I wanna give a huge shout out to a doctoral candidate in my lab, Madeline seed. Who led the day to day of this study and has been involved from day one we had conversations in my office talking about this idea and she’s done a fantastic job rolling with it.

She’s starting the third year of her PhD here at Texas tech. From a big picture perspective, we were interested in taking kind of some of the rigorous methods we use for body composition assessment in the lab, but really producing results. That would be more useful. To consumers, just people interested in fitness even personal trainers or people trying to assess the body composition of their clients.

And as Mike, one of the really popular categories of body composition tools are these home body fat scales. So for anyone listening, if you’ve seen these scales, they often have like silver plates on them. Like little circles. You step on. Most of these scales are using a technology called bioelectrical impedance analysis, and they’re essentially passing very small imperceptible electrical currents through your body.

And the device essentially knows what it’s injected into you in terms of electricity. And then it measures what it receives back. It uses that information along with quite a few assumptions to estimate things like the fluid content of your body. And from there estimate your body fat percentage. So there’s error involved in every step of this process, but nonetheless, these are really popular devices.

They’re affordable. So in the study that we’re starting to discuss now, we looked at 15 different devices. 14 of these were consumer grade devices. We essentially went to, looked at some of the best selling home body fat scales, cuz we wanted these results to be applicable to a wide range of consumers.

And with those scales, the price range from about $24 up to $350. With the majority of those falling below a hundred dollars. So a reasonable price point for a consumer who’s really interested in this, or again for someone like a trainer who doesn’t may not have a large budget for body composition testing, but wants some device they can use with their clients.

As a side note, I’ve talked to others about this before, but one of the devices we use is one that you see sitting around on the counter. Half the gyms in America and that’s the little handheld body fat device where you grab it. And you’re just kinda like squeezing it as hard as you can.

And you’re. I think that was the

Mike: first I’ve of any of these devices or methods of assessing body composition that I that’s the first I can remember at least

Grant: yes. Yeah, no, it’s, it seems to be a common experience in the fitness realm that, there’s that one sitting on the counter and I don’t know, they use it with a clients as they’re getting a personal training program or something like that.

So we did actually include that, that device. But from the big picture perspective, we were looking at all these consumer grade devices. And the plan for this study was to bring in individuals, test them with all these devices, and then also test them with our laboratory methods, using our gold standard laboratory method, which I’ll describe in a moment.

And essentially look at a few items, one at a single point in time, how accurate are these devices compared to our gold standard method? But then even more importantly, how can these devices track changes over time? So these individuals after we’d assessed them in the lab left and continued living their life for a few months, and then we brought them back into the lab, repeated all the assessments.

Both the gold standard and these other body fat scales. And we were able to really see how well these devices could track changes over time. Cuz most people who have these scales, they’re assessing periodically over time. Maybe even more frequently than they should to see. And they’re hoping to see that body fat percentage number go down or that fat free mass number go up.

So just again, did the study a little bit more? We essentially recruited healthy adult females and males. We ended up with 73 individuals. With a really even split between males and females, these were all generally healthy individuals. We didn’t require them all to be physically active, cuz we wanted this to span the range of activity levels.

So we had some individuals who were just recreationally active and we had some individuals who were higher level athletes. So we got a pretty wide range of body compositions and activity levels represented in our sample. So prior to our first assessment of these individuals, we, of course, screened them for some eligibility criteria.

We had everyone abstain from exercise and vigorous physical activity for at least 24 hours before they came in. We had them fast from food and fluids. We recommended a certain amount of fluid intake the night before we did all these control measures so that when they report after an overnight fast, we were fairly confident their body was rested fasted.

There weren’t some, any strange fluid disturbances which is particularly important with this type of technology that relies on body fluids. So once we had ’em in the lab, We performed our gold standard method which is called a four compartment model. And really, we have to use multiple different devices to build this model.

So some people may maybe fill earth DXA, or dual energy. X-ray absorbed. Geometry is one of those methods. Another method we use is the bond pod. Or air displacement plethysmography, which just gives a view of the three dimensional space. The body takes up we used a calibrated scale and we got a body water measurement.

With a technology called bio impedance spectroscopy, which uses mathematical modeling testate body fluids. So we have all these kind of complicated research grade devices that we assessed individuals with and we pull different variables from each of those. Put ’em into this validated equation and we get what’s called our four compartment model, body fat percentage.

So that was our gold standard. So we assessed that. And in the same visit in a randomized order, we assessed individuals on all 15 of these bio impedance devices. So 10 of these were what we call foot to fit scales. So these are the ones that look just like a body weight scale, and they have. Typically silver electrodes on the surface, someone steps on there.

And in addition to taking a body mass measurement, again, these scales are injecting a little electrical, current quantifying what they receive back and then plugging that information typically into some type of black box proprietary algorithm and spitting out the body fat percentage. So we had 10 of those who were foot to foot.

We had four that were hand to foot. So devices that had both electrodes were in contact with your feet and your. So these would often have a little pullout handle where individuals are grasping onto this handle in addition to standing on the platform. And the idea here is that the current is being injected not only into the feet.

And completing that electrical circuit through the lower body, but also through the arms. And the idea is that you can get a better picture of the total body here. And then the last device, the one we were just joking about was the little handheld device looks, almost like an old, I don’t know, some type of video game controller or steering wheel or something like that.

So we had these 15 devices. We had everyone. Assessed by those devices two times. So back to just get a simple test, retest reliability, which is looking at how much can a value vary. If someone gets on the scale, steps off and steps back on and if you’ve ever looked at Amazon or reviews for any of these products, or if you’ve messed around with them yourself, there are some disgruntled users who will say I stepped on, I was 24% body fat.

I stepped off and I stepped back on and I was 26% body fat, they’re disgruntled about that. So it is an important point to consider just the test retest reliability. So that was one component. Again, the second component was just at this one point in time at this first visit, how accurate was each device compared to our gold standard.

And then after this visit again, we let people. Leave the lab. We did not control what they did. But they had the option to return 12 to 16 weeks later. And we had 37 individuals who agreed to come back. So they came back, they were tested. We didn’t know for sure what we get, but actually turned out really nicely.

Essentially. It was a. Almost perfect split half the individuals gained body fat by our gold standard method, half lost body fat which is really what we wanted to see. We just wanted the natural variability and body composition. People are changing lifestyle, exercise, different stress levels.

Some of these were college students. We just wanted natural variability to take place. And that’s what we. So they came back in, we performed all the assessments again, and then we were able to look at the actual change so we could look at it and say okay, for this individual, if our gold standard method said, they gained 2% body fat, what do all these other methods say hap happened?

So I’ll pause there for a second. That’s the data collection. And there, there were a lot of results. I’ll try to distill it down. But do you have any questions or comments at this point now that we’ve discussed the setup of the study. No, that was a

Mike: great introduction to the results.

And just a comment that I know just from many people reaching out to me. So the two questions I’ve been asked the most with these BIA devices are one, are they. Accurate. And then if they’re not, are they at least consistently inaccurate? Okay, fine. If they’re not perfectly accurate, then maybe the an individual reading doesn’t mean that much.

Maybe you can understand it more as a range, but if they were at least consistently. Inaccurate then if you wanted to watch it, you could look for trends, just how you could watch your body weight and pay attention to trends. But of course, if they are just inconsistently inaccurate then that would make them just useless.

Like why even

Grant: bother. Absolutely. Yeah, you hit the nail in the head of what we were trying to get at here. So yeah, that one point in time, and that’s where most studies stop and we’ve done studies like this in the past. It’s a lot easier to get someone to come to your lab. One time.

Them. And then you never see ’em again, rather than come to the lab and be like, okay, please come back in 16 weeks. We really wanna see what happened to you. But yeah, that’s what we’re trying to get at with that second visit is just what you said. Some of these devices, at the group level, maybe they’re just systematically overestimating body fat, but like you said, if you’re doing that consistently, we might still be able to answer the more important question of, has my body fat increased or decreased over time.

So from the big picture perspective and I’m gonna do my best to summarize here because we have 15 specific devices. Once this study’s published, I’m happy to share. And I could even list them out now, but many of the best selling body fat scales on Amazon we have all these devices.

We have lots of metrics. We looked at lots of different things. So all. Paint with kind of some broad strokes here from a Bing picture perspective. And this is exciting to me as a researcher. We did see a lot of separation really between the performance of devices. So there were some that are definitely in the category of that across the board.

They’re inaccurate, maybe they’re inconsistently inaccurate that they’re not really useful. The magnitude of errors are so large that you would need someone. Have their body fat before it’s oh yeah, you saw a change in body fat. So that’s about half the devices.

The other half had performance that depending on the context could be acceptable. So there were devices, even cheap devices say sub a hundred dollars devices that. That performed surprisingly well, I was actually a little bit shocked we had for comparison, we had a $12,000 bioelectrical impedance device in there, as a lab laboratory grade comparison that was thrown in with all these consumer devices.

And there were some of these cheap consumer devices that performed nearly identically to this $12,000 lab, great device. So that as a, as a general. Finding is encouraging for these home body fat scales, but it’s very important that the results are looked at on a case by case basis. Cuz we, we had some devices that if you look at them, if you look at the price, if you look at how they’re described and marketed they seem like they should perform similarly.

And we had some that were incredibly different. In their performance. So there were several ways. We looked at the performance, some were what are called, like group level metrics. We’re just saying overall, did this device overestimate or underestimate body fat that’s often where studies stop. And that, that can be useful.

But again, we want this to be maximally benefit the consumer and most consumers, most users of these devices, they don’t care if we track. 20 of them doing the same nutrition exercise program. They care about the one and equals one. They care about, themselves. Did I gain earlys body fat?

So we also looked at a variety of individual level metrics and something that, again, my doc student, Madeline had this very good idea was to generate all these rankings. So we have all the devices ranked first through 15th in different categories. So like individual level accuracy at one point in time, individual level accuracy.

If you’re trying to actually track a change in body composition over time. So all this will be available in the manuscript and hopefully it’ll be very helpful to consumers where they can find their particular device. Or if they’re in the market for device, they can look and say okay, the category I care about most is accuracy in tracking in one individual.

Cause I’m gonna use this in myself or my clients. And they can look through and find, say the top ranked device at the price point that works for them and use that device. I would say in general the devices that have electrodes in contact with the hands and feet, those tended to usually be in the top half performers.

But surprisingly there were some just foot to foot devices that just looked like the scale you step on that were also near the top in terms of performance. I know these are all broad statements and for those who have a specific scale, I definitely encourage you to follow up on this.

We’ll try to share. Widely as soon as it’s published, because we think it will be of interest to, to really anyone using these scales. But again, from the big picture, some of these, not worth using will probably add more confusion to benefit, but there is actually promise in this category of technology, even on these relatively cheap consumer grade devices, that if you pick the right device, you could actually get some useful.

Mike: Interesting. And are there any details you can share on why you think the, some of these devices? Cause I’m a little bit surprised to hear that. I basically wrote these devices off some time ago and didn’t bother using them myself and didn’t recommend them. I tell people, just pay attention to your weight.

Maybe take, pay attention to the, your waste circumference. There are a few things if you wanna take some measurements, but from the research that I had seen, I, again, I haven’t looked into this in some time, but from the research I had seen, I was like these things are all over the place. So I just I just kinda wrote them off.

And so I’m a little bit surprised to hear that, that some of these scales in particular I was expecting the four points to, to outperform for obvious reasons. But do you have any thoughts to share as to why some of these devices actually performed well and maybe better than other scales or just two point

Grant: devices?

I do have a few thoughts and I’ll start out by saying that I completely agree with what you said about tracking body mass and circumferences some of those other metrics, even with the best performing devices here. I think for the longitudinal tracking, one of the best performing devices had, if there are any like, Stats nerds in there.

It had 95% limits of agreement of about three and a half percent. For others, this would essentially mean if you saw a change in body fat, it’d be reasonable for the real change in body fat to be about three and a half percent higher or lower than that. So that’s a pretty large range. If someone’s interested, did I gain, or did I lose 2% body fat?

And you go, three and a half percent above or below that. You’re clearly looking at some reasonable values that could be negative or positive. Like you actually lost body fat. You’re actually gained body fat. So that’s a big caveat. Even the better performing devices have a margin of error. It’s incredibly difficult to be very precise and accurate when you’re tracking body composition of one person over time.

So that’s a caveat. So I’d completely recommend what you were saying, Mike.

Mike: Yeah. That’s a good point, especially for fit people, even for women. Yeah. Even for, if women let’s say, they’re fit women and they’re around 20% body fat give or take and they want to, you’re still talking about enough error to.

Confuse somebody, where , they’re like looking in the mirror and they’re feeling their clothes, but this thing says they got

Grant: fatter. Like how does that work? Yeah, no, you’re absolutely right. In a good proportion. It was depending on whether it is the cross section, like the one point in time or the tracking analysis, 40 to 50% of the systematically.

Overestimated body fat and lean individuals and underestimated body fat and those with higher body fat. So essentially there were a lot of devices that were pulling people towards the middle. So you might be really lean stepping on there and it was showing you at a higher body fat than you really are.

Conversely, if you have a higher body fat presented, you might be stepping on there and it’s like. Pulling the body fat percentage down. So there were whole category devices that displayed this characteristic I’m describing which is called proportional bias. It’s bad. It’s something we don’t want.

So again, many of these devices you look at and be like, okay, no, this is bad. This is bad. It’s Whitling it down to a select number that we’re potentially useful. And again, then I would still pair with what you said with body mass, with circumferences and say, does this make sense? Like you were saying, if.

I look better, my clothes fit better, whatever my performance is up, but this scale showed body fat change in the wrong direction. Then I would not put too much, weight on that. Yeah. In terms of the reasons something we saw emerging is some of the manufacturers that have been around longer.

So like Tanita is an example manufacturer that’s been around long time. They’ve had a lot of research and development personnel look at their equations over the years. Some of those seem to perform much better than some of these newer companies. There are lots of kind of companies that just do consumer tech type products that they’re like, oh yeah, we can do this body fat scale.

And they produce a device. And the device itself maybe is similar to say like a cheap to needed device or one of these other devices. But I think there are some cases probably where the people putting the equations together. I wanna say, didn’t do it right. I don’t wanna be too negative but they might not have known the best way to put the equation together to where it would scale outside of their validation population.

And they also might have validated the scales in a very homogenous, like small group with some of those issues. Like I was just saying where lean individuals and individual with higher body fat were systematically miscategorized that could speak to an issue with how well the equations were validated.

So I think part of. The scale itself is one component. Some might use better quality components. They might be getting cleaner data. But the other component that I also strongly suspect is how good of a job their scientist did developing these body fat equations. And as I mentioned a bit earlier, the hard thing here is that.

None of the manufacturers will disclose equations. We have a table put with article about whatever information we could glean about each device. Madeline I know contacted a number of companies trying to get information and most of the time, the person you can get in contact with one probably doesn’t know, but also won’t disclose.

And they’re like, oh, this is proprietary. From these hints, I think that some equation. Some people maximize the utility of that bioelectrical data they get and others are just like, oh yeah, we’ll throw this in equation. Like this seemed to check out in our sample of 20 people.

We tested this on, let’s go ahead and mass produce this and sell it. And the consumer won’t know any different

Mike: and brands. I would guess at least in many cases, brands are actually not doing any of that work. They’re just going to a manufacturer and the manufacturer says, oh yeah, sure. We can make that for you.

And. If they do even ask questions beyond if that, in many cases that’s probably the, okay, here’s my, here’s the order. And we’re gonna slap our brand on this then. Yeah. Maybe they would ask for some validating information, but I’m speaking from experience, maybe more in the supplement world, but especially if you look around on Amazon, I’m sure you guys and gals ran into this that you have a lot of.

There are a lot of Chinese companies now that they’ve spawned. I’ve seen so many random brands, especially in electronics, and these are not brands you would recognize, but there are a lot of these products now that have just flooded Amazon, particularly with electronics. And so if you don’t recognize the brand name, if it’s just, it might even sound like something legitimate, it might just be some random company over in China.

That’s just throwing their stuff up on Amazon. They might not care about the quality of the thing. That’s a growing problem on Amazon.

Grant: Yeah, no, that’s that’s certainly fair. And, there were, some trends emerged like the more reputable companies, but there, there were the occasional oddballs where there’s a scale where yeah, this scale, we’re not too optimistic where it actually came out more highly ranked than we’d expect.

We saw some of those trends. But again, I’d go back to, it’s a little bit of a case by case basis based on what we saw. Even ones that appear similar. So yeah, lot, lots of interesting kind of wrinkles here. And, I wish every company was completely transparent with what they did or didn’t do to develop their equations and just would show us the equation, but that’s not the nature of it.

We just wanted to, test what we could test and report what we found. So consumers and practitioners can make more informed decisions. Yeah.

Mike: And for people listening who want to read the paper and they wanna see the ranked lists, do you know approximately when it will be available?

Grant: Yeah, I would be I want to give an estimate. I would say a few months, the peer review process is not super speedy. I think this is a really clean manuscript. It’s really tight. We’re submitting to a fairly reputable journal. Very reputable journal, a very highly ranked journal. So we hope that they’ll expedite everything.

There are a few abstracts that are already. Available that I know I’ve posted about my social media, if any of that is linked, I’ll grab them and

Mike: I’ll put them in the show notes. So whatever’s available.

Grant: Yeah, no, that sounds great. And I’ll definitely share widely once this is fully available, we also actually already have a public listing.

This study was not funded by any of these companies. We just used. My own research funds from Texas tech to purchase these devices. And there’s nothing inherently wrong with industry funded research, but for this particular one, none of these companies played any role. So we have this registered on clinical

So if individuals are curious, I can also send a link to that. The full list of every device we looked at is already posted publicly. And we did that just for openness and accountability. So it’s not like we’re dropping some device off the list because the manufacturer, whatever.

Pressured us or anything like that. So the fullest is out there if you’re someone that uses one of these devices, and you’re curious, you can look at the list and if you see your home body fat scale on there, then you can, wait with baed breath until it’s published and you can see

Mike: all the results.

And last question about that paper before we move on. Are there are are there any devices that that emerged from the analysis that you think could be useful? I you implied that this was the case, but then you had mentioned these margins of errors. So I just wanted to follow up with a question of, for people who are into their fitness enough to wanna track their body composition.

Were there any devices in the 25 or so that you purchased that you feel. Would be

Grant: useful for that. Yes, I think there are. And I’ll I’ll consult my spreadsheet over here, cuz there are so many different rankings. So we did ultimately get to an overall ranking like considering the cross sectional.

So the one point in time, accuracy, both for groups and individuals and the long term tracking accuracy for groups and individuals and their reliability.

Mike: Which is what most people that at least who have reached out to me, that’s what they care the most about is of course, it’s, it would be great if it were very accurate in an individual reading, but minimally, it needs to be consistently accurate ish enough so that they can just, throw it in their spreadsheet.

They can take their readings once a week

Grant: or. Yeah. No, absolutely. Yeah, looking at my list here for the individual tracking. So again, just tracking changes, looking at that one domain rather than our overall rank. The top few were those. What we call Octa polar, or they have eight points so that the hand electrodes and the foot electrodes some of the best performing ones, one was actually an Amron device.

So not just the handheld one, but the Amron HBF five 16. So it’s still very affordable. I think it was under $60. But has hand in foot electrodes. The next couple, there was a Tanita device that also had the Tanita BC five 60. Inner scan. So another one that utilized those had the hand electrodes also.

One other one that I thought was really interesting that was actually ranked in the overall rankings. Second highest out of all of them in the first highest was our $12,000 device. So I, we were a little bit glad to see okay, that at least holds.

Mike: Yeah. Imagine this the $60

Grant: device beats that, okay. Yeah. I’d be like, oh didn’t spend that money. But one device that’s really interesting. That was a little bit higher price point is $350. So it was the most expensive of the consumer grade devices, but it was new and it’s from a very good manufacturer. Was the InBody H two O N. So those involved in researcher who have been in like a bunch of supplement stores, have them, some gyms maybe familiar with these InBody devices embody, it’s a good manufacturer.

I know a little bit. Small amount about what they do under the hood from communications, with their science staff. And I’d say it’s a reputable company. So they have this again, price point’s a little bit higher at $350, but this InBody H two O N, which was another one that had both the feet electrodes and hand electrodes.

It performed well, and I’m actually involved in some other data analysis from a separate university looking at the same device and it’s performing there also. So I think it’d be a nice one. That’s a little bit higher end, maybe a, personal trainer, something that’s wanting to have these reports for their clients.

Have a little bit of a nicer device, a, not that looks or everything, but a classy, nice, solid looking device. Yeah, it performed really well. It was in those top few for tracking in visual changes. And again, it was ranked second out of all of them. Second just behind the research grade device and the overall ranking.

That’s what I like. I’ll disclose that I embody had no involvement in this study. Several years ago they loaned a device to our lab just as a loan not research funding or anything that we used for a study or two. And then it’s been long since return, but I’ll give that just as a disclaimer, because I have had interactions with that company before.

But anyways, so yeah, across Tanita, Amran and embody, like we were talking about common thing is that all those are pretty well established body composition manufacturers. Each of those had both a hand and foot electrodes. So those performed really well, but once you go past there, you do get some that are just foot to foot, including one of the, to needed devices.

But yeah, it’s initial glimpse on specific results. Those three are ones that performed pretty well for the individual tracking. That’s great. That’s great.

Mike: Let’s shift gears now and talk about some other methods of assessing body composition. So something that I’ll I will get asked fairly often is just what’s the best method.

What’s the most accurate method, right? The four compartment analysis is. Is something that’s done in a lab. So we can probably exclude that you described that. And but for what, for something that, people can do at home, people will ask me about CAS. How about that? Or, yeah. They’ll ask about these devices that you discuss.

They’ll ask about different measurements that you can take and plug them into to calculators. Can you talk a little bit about and, even though it’s not accessible, people will ask me about DEXA or, somebody will show, maybe it’s a video of somebody getting a DEXA scan and it’ll say 10%.

And the dude looks absolutely shredded. Maybe even Strad glutes shredded. I’m exaggerating a little bit, but where, yeah, there just seems to be a little bit of a disconnect, 10 10% when he is covered in veins.

Grant: Yeah, I’ll answer the main question, but since you’ve finished on it, I’ll mention the DXA question.

One important thing to realize is, I think sometimes from the outside Dex is a perfect example because, the evidence space community they’re familiar with DXA, they’ve either had a DXA done or seen it. They’re aware it is a really good method. When you’re on inside, I’ve used multiple different manufacturers of DXA across multiple universities, multiple different software versions.

And you realize that there, there are quite a few things you can do in the software that change the results. And I’m not talking about willful manipulation, but things like settings, you can change different settings that could reliably change the body fat percentage by say five. And

Mike: 5% absolute. Just so people understand.

Grant: Yeah. So really, meaningful changes where you could have the same shredded strated glutes person, jump on one DEXA, get it, get the scan done, they have the results. You could change some of the settings and reprocess the same scan and you would have quite different results.

So that

Mike: could be the 5% turns into 10 or something. Yeah.

Grant: Yep. It certainly could be. So there, there are some of those settings there’s some error in every device. Even like the pre-assessment standardization, we talked about a little bit with fasting fluids, all this, we’ve done studies where you can trick DXA, so to speak by carbohydrate and creating and fluid manipulation.

So depending on where someone. Assessed or how depleted or replete they are at the time of assessment that, that could affect their readings beyond just what you would see visually.

Mike: And I’ve seen what I think is some manipulation as well. On the other side I’ve seen I can think of a fitness influencer or two has, who has used DEXA scans to try to promote their programs.

And yeah, Dexus scans are often represented as the gold standard infallible. And it’ll hop on the DEXA and get a reading of five and a half percent. It. No, he dude, he’s lean, but he’s not five and a half percent. That’s a nice maybe eight or 9%, but that is not getting ready to step on stage.

And so that’s just a good thing to point out that for people listening don’t just take DEXA. Readings at face value, they might be right. They might not be right.

Grant: Yeah. And the same principles we were talking about with the BIA apply that it’s really hard to accurately track an individual. So even in that case, maybe the person’s not willfully manipulating, but they might be that person who, because of some of the characteristics in their body, they’re systematically underestimated by Dexus.

So they’re like, they’re happy. They’re not gonna argue. If it’s saying they’re five and a half percent body fat, even if they maybe in their heart of. Think they might not be five and a half. That would be that they

Mike: would’ve to step on stage at 1%. But yeah.

Grant: Yeah. So anyways, just yeah.

Point being, there’s air involved, there’s settings, this and that transitioning, to the main question you’re asking about best method I, I think you alluded to this already, but I’d say the best method, certainly depends on the context. If you’re talking to someone who has a flexible tape measure and some $10 Calper, and you’re saying, oh yeah, go, pay $200, you’d go drive five hours and pay two.

Or she had a Dexus scan done somewhere. That’s the best method, that’s probably not the best method for their context. It’s certainly not the best method for repeating frequently over time. I’m a big proponent as others are. I know bill Campbell often says this, which coincidentally enough, Madeline Seeler who helped so much and really led the.

The BIA study. She studied with bill Campbell before she came to study with me. So she’s had a lot of body comp stuff, but I’m a big proponent of those practical measures and not doing too much with them. So for example, if you’re interested in fat loss and you can take these. Raw skin fold thicknesses at different sites on your body.

It might be best to just keep those raw and interpret them as much as you want to plug those into equation and distill all that down to 13% body fat. It might be more informative to watch those over time and say okay, I have a lot of skin fold sites that are not changing, but I. See like this abdominal skin fold, this is actually decreasing over time.

That’s something that I want. That’s consistent with my goals that might get washed out when you’re pooling all of these together and throwing it into a body fat equation. Similarly circumference, like you said, if you were seeing your waist circumference, go down over time. The scale’s not budging maybe it’s even a home body fat scale and that number is not changing or it’s just randomly oscillating around that’s an encouraging sign okay, waste circumference is decreasing.

Maybe my upper arm circumference is increasing. Maybe I have some amount of, recomposition going on right now. I’ve, increased my protein intake recently. I’ve, increased the intensity of my training and so on. And as the more trained someone is, you’re dealing with small margins and.

You’re not gonna see large changes and it’s hard to track those small changes. You do see cuz of the, little bits of error, but I’d say there’s certainly nothing wrong.

Mike: I’ve stopped bothering with that because take calibers. I wouldn’t say that it requires a lot of skill to take caliber readings, but there are right and wrong ways to do it.

Yeah. Even the amount of skin that you pinch, if you’re not consistently pinching the right amount of skin in there are spots too. Are hard to do yourself and really it’s probably better if somebody else does them. And then, like you said, also the margins of improvement are so small that they overlap with the margins of error.

And so I’ve found just as a random comment that what has been more. Useful for me, if we’re talking about tracking muscle gain is just using performance in the gym as a proxy and watching my performance with given working weights progress. First in reps in reserve, where okay. That weight felt a little bit heavier a month ago.

That’s a good sign. Eventually that turns into being able to get an extra rep or two with the working weight. Eventually that turns into some extra weight on the bar. And of course. I just go, Hey, if that’s happening I’m gaining, I know I’m gaining at least a little bit of muscle and it might not show up if I were to take measurements because we are talking about such small amounts.

And if I mess it up a little bit, it’s not gonna show.

Grant: Yeah, no, I think that’s completely fair. And I don’t think there’s, I certainly don’t think there’s anything wrong with a method like that, even if you want to periodically. And even if it’s every several months, every six months, every year, if you’re just one big picture I’m true.

Yeah. I’m, I’m willing to maintain this, whatever I’m at for the next decade, two decades, three decades, I think a periodically check and say okay, within, reasonable margin of error, is there anything I’m just not realizing that I didn’t realize cause general population wise, maybe not.

Fitness community quite as much, but general population wise, things tend to drift over the years and decades. Yeah. In a direction you don’t want. So even something as simple as that, I ideally your general practitioner, someone might do it, but maybe not, simple waste circumference every six months or something, just to make sure something’s not creeping way you don’t want to.

And again, I’m getting more general population there unless, like peak fitness, but. Yeah,

Mike: no, that’s a good point. But I guess you could, like you said, you could do, even in the case of someone like you or me, maybe we would, we could do check-in measurements every so often if we were so inclined, for me, it’s to that point of maintenance where I’ve accepted that this is about.

As big and strong as I’m going to be. No matter what I do, if I really wanted to push for more, of course, there’s a little bit more to be had, but I’m pretty happy with where I’m at. And this is I’m getting ahead of myself because something that you know, I’ve gone through the whole experience of body fat percentage and being really into, trying to get to the number and eventually.

It’s just do I like what I see in the mirror, if I feel like I’m too fat. Cool. I’m just gonna restrict my calories for a month or two. And now I like what I see in the mirror. Okay, great. What number is that? I don’t know. It’s in the range of what I like the most is probably between eight and 11% or something like that, and exactly where I’m at in that range.

I don’t care. As long as I’m in that range. That’s what

Grant: I like. Yeah, no, that’s completely fair. And it’s funny as a researcher who does a ton of work in this area, there are definitely times I zoom out and think about that and does this matter? And it does. And some of the research we do is for other contexts and, there are lots of health and disease implications, other things we do with body com, but yeah, at certain at certain fitness level, you definitely can legitimately question like, okay, how much does.

Truly matter at this point, what does the number I’m getting matter? Yeah, I would say, circling back around for anyone, interested in body comp assessment, certainly, there’s nothing wrong in being interested in them, but I’d interpret everything’s grain of salt. I’m always a proponent of multiple metrics, especially, if you’re doing some of these potentially less accurate tests, but if you were, if you do have even one of these better home body fat scales, pairing that with tracking your weight, considering your performs in the gym, baby periodics circumference or skin folds, kinda getting that holistic.

Cuz again, just, I think I saved this earlier, but say again, if you’re getting a, some number that’s tracking the wrong way, but everything else is good. You feel better, clothes fit better. Say, like your method with performance is better. All of that. You certainly wouldn’t wanna make some large programming change, say okay, I need to scrap everything and start over because my home body fat scale said I went up 2% or get, just get

Mike: yeah.

Or just be demotivated or discouraged by it. Yeah. And I understand that again that, that. Be the case now, but there was a time when I just wasn’t informed enough. And maybe would’ve been discouraged because it’s just conflicting information. And if you place too much credence in, it could be a device or a method like, skin fold.

I know a lot of bodybuilders were, will swear by the skin fold method. And and if that gives you a number. Is discouraging, might lead you to question some of these other methods of assessment.

Grant: I agree. As one random side note, I just mentioned, cuz it’s another class of technology we haven’t talked about, but that our lab has done a lot of work in and we’re really interested in is I think there’s a lot of promise in terms of just reliable assessments and easy assessments.

There’s a lot of promise for three dimensional optical imaging. Oh, that’s good. I wanted to ask you about that. Okay. Great. Yeah. So there are lots of these standalone scanners like St. U fit, 3d size stream. And so on that you could actually like here in town, our crunch fitness, for example, has a fit 3d, and those are showing up, I think in more and more facilities, but with phone applications, we’ve recently been evaluating some of these they’re, phone cameras are good enough.

And the technology is getting good enough on 3d scanners, that there are some free phone apps where you can set up your phone and rotate in, in front of your phone. And it builds a little 3d avatar of your body. And the nice thing there is that it can pull circumferences very reliably. So it’ll have some definition based on the pixels it’s seeing on where your waste circumference is.

So there’s no. Guesswork really of am I measuring this in the same way? Or can I really measure this particular circumference? If I’m trying to measure my chest circumference, do I have it, this flexible tape measure at the same point in my back, if I’m measuring it by myself. So I think there’s some promise there for just like simple things, even if it’s her periodic check in where technology can’t assist, even in a free capacity without like purchasing a particular device, but could assist with tracking some of those basic metrics.

What have you

Mike: seen with some of the more expensive devices? What you’d find in a gym? My gym has one. I haven’t bothered with it, but

Grant: yeah. So the test three test reliability is very good in general from our lab and the literature. Most devices would be comparable to a trained tailor in terms of the reliability.

So the only times I’ve seen like a manual assessor outperform, the reliability of these circumference assessments is if the study is using yeah. Again, someone trained. Equivalent to a tailor. So if you’re talking about, your random gym buddy or something, who’s just like slapping on and measuring might be better quality than that.

This is another example where I think the raw inputs, the circumferences in this case would be potentially more useful than plugging into a body fat equation. The body fat equations have a little bit of work to do. And it makes sense. They’re trying to estimate something internal your body fat, just based on surface metrics and not even getting.

Skin fold or something like Calper. So there’s some challenges there, but in terms of the circumferences, if you’re like, okay, I wanna track my upper arm circumference, my thigh circumference and my waist circumference over time. I want my arms and thighs to get bigger. And I want my waist circumference to get smaller.

They’re sufficiently reliable that they. They, if you assess the same way not after a big meal one time and fasted the next time and all of that if you assess the same way they do have pretty small margins there where they could track small, but real changes in these circumference.

So I would say I’m generally optimistic on them. But again, with some caveats that like if it’s spinning out the body fat percentage, say the one at your facility I wouldn’t put, too much weight on that, but if you’re interested in tracking, some individual circumferences, I would say it’s a decent technology for that.

If it’s

Mike: consistently tracking those circumferences even if the body fat percentage were wrong, I suppose if it. Consistently wrong at least could show you a trend, right?

Grant: It could. And the hard thing on there is it’s similar to BIA, but I’ve seen a little more under the hood cause I’ve been involved with a couple companies helping them develop some of these equations, but it just depends on what circumference they’re pulling.

So I was even advocating, there’s some, they’re even different versions of the waist circumference, depending on if they’re doing the narrowest or the Naval or, halfway between where they think the ribs are and the hips are, all this it depend different companies pull different circumference inputs, cuz it, it can measure anything on the body.

As a 3d avatar. There are all kinds of weird ones. It’s like I can measure from like the tip of this shoulder down to like your opposite butt cheek, just all this random stuff. They can pull any input they want, cuz they’re not actually there with the tape measure. It’s just, based on the avatar.

I think there’s a lot of room for growth for what inputs are pulled, how they’re weighted and equation and all that. Yes, you’re right. That I think you could track there. Hard thing is it’s such an early technology that some of. Devices are like pushing a body fat equation update like every few months, which is not helpful to the consumer.

We’re it’s all of a sudden it’s different by 5%. They’re like, oh yeah, we changed the equation. We don’t use like this circumference anymore. We use this one. So that’s another reason why they don’t do that as much with actual raw circumference. It’s usually just what they’re doing with the circumference.

So yeah, the relative youth of the technology is another reason why I am a more of a proponent of the circumferences. Now you.

Mike: Spoke, I think implicitly about this already, but I think it would be worth just explicitly commenting on it. And that is the. Accuracy or the usefulness of, oh, what’s the name of the method.

So you’re supposed to take some measurements. This is for body fat percentage. And I don’t, I haven’t done it in a while. So you could tell me, I don’t remember the exact measurements, but you take a few measurements. You plug ’em into a calculator and it spits out. I believe it was it. Is it a Navy thing or a military thing?

Grant: Oh yeah. Like the army, body fat equation.

Mike: Yeah, I know there’s one of these things out there that makes it seem very simple. And I know then many people turn to it because it seems so simple.

Grant: Yeah, that, that’s a really interesting question. I had another student just defends dissertation and it focused a lot on military and military relevant populations with that equation you’re describing and 3d scanners and four compartment, all the things we’ve talked about rolled up into one.

Okay. So it’s not perfect. We know that if you’re just looking at like next circumference and waist circumference in males and like neck waist and hips and females, this is certainly not a perfect measurement. Like we were just talking about 3d scanners. We’re not measuring anything internal to the body, but we’re trying to measure something internal with body.

So you already have a little bit of a disconnect there, but surprisingly, when you look at there’s a reason why it’s hung around and of course from say the military perspective, they need something simple. They use like height for weight tables, and then this body fat equation, like as needed.

But they need something really simple. We’ve had some discussions with personnel there about whether even just moving to 3d scanners, they could at least eliminate some of the like tester to tester, variability and how these are assessed. But the equation does. Okay. It’s not as bad as you’d expect.

We went into this big project. And over a hundred individuals looking at this equation in many others. And we expect it to be like, oh, there’s the antiquated, Navy army, military equation. And it did pretty well for the people who developed it were, I don’t know, onto something like this scaling with the Nexar conference.

It’s weird, but like Nexar conference matters for like just a natural build, like correcting the waste circumference for someone’s natural build. So anyway I certainly wouldn’t take it and have someone plug it in there and, and. Funny, but an anecdote, even one of the military people were talking about is how people would train their neck really aggressively and try to get all this neck hypertrophy to trick the equation.

That’s neck

Mike: training sees a little bit of a thing. I’ve been asked more about it recently than in the past. It just seemed the, in the fitness world, people asking how do I make my neck as thick as possible? Like

Grant: why, what do you mean? Why

Mike: just do more

Grant: squats? Come on. What are you doing?

Yeah. Yeah. So anyway, it, it certainly isn’t like the best method, but it’s all, contact dependent if you’re screening thousands of people and just trying to like have some rough cutoffs and this and that, it does pretty well. We’d never say oh yeah, you don’t need a DXA scan ever.

Just yeah, gimme your next circumference, your waste circumference. And that’ll tell you like, For sure. Your body fat percentage. So I don’t know. Those are, that’s just a random collection of thoughts. Yeah. Yeah.

Mike: It sounds like it’s maybe similar to BMI in that regard, but for body comp where yeah, sure.

They’re at a population level, it’s pretty useful at an individual level. It can it can underestimate it can overestimate body fatness depends on how fit you are,

Grant: blah, blah, blah. How much you’ve trained your neck, all that good stuff. .

Mike: So then I’m assuming the thicker, the neck, if you’re training your neck explicitly to try to trick it, then that brings your body fatness down.

That’s the point. Yeah. Yeah. So you can be overweight, like if I just get a really thick neck, then I’m gonna pass

Grant: this test. I think that’s the idea. How successful people have been when they try to do that? I’m not sure. Yeah.

Mike: Yeah. How much neck hypertrophy does it take? That’s funny, but no, that’s great.

And that that’s all the questions that I had for you. Is there anything that is still sitting in the back of your head that we haven’t discussed that you think we should cover

Grant: before we wrap up? I don’t think too much. I just reiterate that. I don’t think there are many methods that are just.

Complete trash. I think every method you should take with a grain of salt, you can maximize the benefit of any method you’re using by being rigorous with how you standardize before assessments, not just doing these haphazardly, even in our paper about submit, we advocate for things like taking multi-day averages.

So if you’re able to perform the same assessments on a couple days in a row and average across those days that can help reduce some of that day to day random error. So yeah, I think whatever method you use, if you use. Just apply it consistently. Ideally look at a couple different metrics. Don’t read too much into each one.

Just take it with a grain of salt, consider performance. What are your goals? Is body comp your number one goal and you’ll sacrifice performance for it or is it performance? And you’re hoping body comp comes along for the ride. So yeah, for anyone who’s like really in the weeds on body comp or obsessing over it, I.

Think about how you could use it to benefit you in some way without again basing all your decisions on it, or placing all your value in a number being spit out by a $35 scale that you got from Amazon. And again, for anyone interested in the consumer grade BIA devices, we’ll certainly share that, but hopefully that will help inform decisions.

Cuz some people might be using a scale where they’ll look at this and say, okay, we actually need to get rid of this. Worse than nothing, but some people might have scales where it’s okay, this isn’t everything, but it’s a piece of the puzzle that can help myself or my clients or so on. Yeah, I think those would be my only parting thoughts.

Mike: And one follow up question on that actually is if you are gonna be using one of these BIA devices, let’s say you have a good BIA device. You had mentioned previously a couple of the things that a couple of the factors that you controlled for to get consistent readings, but maybe we could just wrap up if you wanna share with people at home.

All right, here are the important things to make sure that you’re getting consistent readings and, or, and also accurate reading.

Grant: Yes. No, that’s a great question. So I’d say at a minimum you want an overnight fast from food and fluid. Sometimes I won’t get into all the weeds here, but we’ve done some studies just looking at simply someone comes in the lab and ingest water.

And how much does that affect their bio measurements? And it it can be meaningful depending on the type of technology. So I’d say at a minimum eight plus hour, overnight faster in food and fluids. So ideally, the morning is the ideal time to do. Void your urinary bladder, if you need to defecate or if you have a normal morning bowel movement, go ahead and do that.

Just everything you can standardize. Of course rested overnight ideal will be longer. Like in our study we had a 24 hour abscess from exercise. So if you do incorporate rest days or active rest days assessing body comp, the following morning might be an ideal time to do. Doing that consistently. So you wouldn’t necessarily want to assess sometimes the morning after a heavy leg day, the previous evening and other times assess after rest day.

So consistency, even on what you did the day prior is important general awareness of dietary patterns. If you track and if you’re, or if you’re just consistent on your diet over time, you may have this addressed, but just be aware, especially if you’re doing carbohydrate manipulation or if you have, you’re changing up your diet in a way that dramatically affects the water content of your foods.

Some of those things, even the preceding days can affect measurement. So I’ll just be aware if you are tracking, try to. Think not only overnight, but the previous day and kind of habitual diet, what things could you be doing that might be influencing this? And I’d say the more, those things you button down, the more noise you eliminate and the cleaner your signal or the actual value you’re trying to get comes through.

Those would be the main things, food and fluid rest kind of awareness of your overall dietary pattern exercise patterns, and so on. And I think if you do have someone else involved, for example, helping with skin folds, circumferences and so on having the same person do that over time and apply that consistently would be the ideal scenario.

Mike: Great tips. Great tips. This was a great discussion grant. I really enjoyed it. Very informative. And why don’t we wrap up with where people can find you’re active on social media share a lot of great educational information there, and then anything else you’d like people to know about?


Grant: let them. Yeah. So I’d say if you want information on me, I have a personal website. That’s just my name, grant, I have my academic CV up there. I have information about our lab equipment, our research team, my home gym, just all that type of stuff on the personal website. In terms of social media, I’m probably most active on Instagram and my handle.

There is just grant. Underscore Tinsley, T I N S L E Y, underscore PhD. And I typically share new research on Instagram in my website also has links to PubMed and other websites that kind of keep track of all the research output from our lab.

Mike: Awesome. Thanks again for doing this grant and I look forward to the next one.

We should brainstorm something else we can talk about.

Grant: Absolutely. Thanks for having me on Mike.

View Complete Transcript