Body fat %: Aria vs. DXA

[Dominique]

Has any Aria user done a D(E)XA scan in order to compare the body fat % obtained with both methods? While the Aria is convenient for home use on a daily basis and the body fat values I've got with it have been very consistent over time (have had the scale for 13 months), I understand it's not necessarily accurate in absolute terms and I've been considering a DXA scan to assess the discrepancy.

[Michael]

I haven't seen anyone that had a DXA scan done as a comparison but I did get a BodPod Express measurement done to comapre earlier this year. I posted my results here.

I would be very interested in the results if you do get a DXA scan done.

[Dominique]

Thanks, Michael, for the link to your older post! It was the first time I heard about the Bod Pod Express. They do say it's "available only for the US market" (I'm based in Europe). Was it expensive?

I'll definitely post results about DXA if/when I get a scan done.

[Michael]

No it wasn't expensive, $45 for a single test. (The local vendor here offers a discount if you want to pre-pay for multiple tests to track your progress.)

It looks like there are a few sites with the full BodPod in the UK, but availability over there appears limited.

[Nick17]

DEXA scans are the most accurate way to assess body fat percentages.  Instruments that use bioelectrical impedance can largely over-estimate or under-estimate the percentage of body fat, depending on a number of factors, one being hydration and the readings can even vary throughout a day if taken at different times.  Even body fat calipers that are used for measuring bodyfat by personal trainers tend to be off by up to 3%.

DEXA scans are definitely the gold standard.

[Dominique]

---

@Michael wrote:

No it wasn't expensive, $45 for a single test. (The local vendor here offers a discount if you want to pre-pay for multiple tests to track your progress.)

It looks like there are a few sites with the full BodPod in the UK, but availability over there appears limited.

---

Yes, $45 isn't too bad. I assume it only gives you a single value for overall body fat, like the Aria, right?

As a matter of fact, I will be in the US in late October, so I may give it a try, if I can find a suitable place.

[Dominique]

---

@Nick17 wrote:

DEXA scans are the most accurate way to assess body fat percentages.  Instruments that use bioelectrical impedance can largely over-estimate or under-estimate the percentage of body fat, depending on a number of factors, one being hydration and the readings can even vary throughout a day if taken at different times.

---

The method used by the Bod Pod mentioned by @Michael ("air-displacement plethysmography") doesn't appear to be sensitive to hydration, like bioelectrical impedance. The site includes links to studies validating it against reference methods, including DXA.

That said, I still intend to do a DXA scan, as I understand it's the "gold standard", as you said.

Btw, @Nick17, are you familiar with the BodyMetrix, by the company IntelaMetrix? I've seen it mentioned by a fitness professional as a reliable (though rather pricey) way for home assessment of body fat. It has good reviews on Amazon.

[GRRRRR]

I see the reference to the help page which says nothing of interest... thus I don't get the point?!

Re: accuracy vs DXA.

I got an Aria on February 25, and had a DXA scan on March 4.

Weight is accurate.  Aria: 203.5lbs, DXA 2 hours and 1x 12oz coffee later was 203.4lbs

Like most other scales Aria's biometrical impedence body fat measurements are next to useless for my particular body shape and I wish I could either turn them off and/or not have them show up on my dashboard.

Maximum BF as per Aria was 27.3% on March 1; and minimum BF was 25.4% on March 4.

It angers me to see these numbers because I wish they were true : - )

As per my Fitbit dashboard, Lean Mass on March 4 was 151.6lbs our of a total of 203.5lbs.

By implication, Fitbit claims 51.7 lbs fat (based on the %) and up to 51.9 lbs fat (based on simple subtraction) of total fat in my body--I wish!

DXA scan results the same day yield:

% Fat = 32.9%

Fat = 66.9 lbs

Bones = 6.3 lbs

Lean = 130.1 lbs

Your mileage may vary!

[Dominique]

---

@GRRRRR wrote:

Like most other scales Aria's biometrical impedence body fat measurements are next to useless for my particular body shape and I wish I could either turn them off and/or not have them show up on my dashboard.

---

I disagree with the fact body fat estimates are useless: so you know they are not accurate in absolute terms, but relative changes over time will still be useful. If you're getting leaner or fatter over a longer period of time, this is likely to be reflected in the body fat % values obtained with the Aria.

---

 Maximum BF as per Aria was 27.3% on March 1; and minimum BF was 25.4% on March 4.

---

A 2% fluctuation in just three days is a bit high and probably indicates the weigh-ins didn't occur at the same time of the day and in the same conditions. In my own experience with the Aria, body fat % fluctuations from one day to the next are usually less than 0.5%. I always use the Aria first thing in the morning after a trip to the bathroom, when my body is the least hydrated.

See this post for intra-month fluctuations observed over 20 months.

[GRRRRR]

Dominique, I do appreciate that your Aria seems to be able to measure your (less than 20% fat & relatively stable weight) body relatively accurately, this has not been the case for me during the 8 days I've had it.

I use my scale every day, after eliminating and before eating or drinking anything.  I do not start my day at a set time, so during the past 7/8 days I've weighed in during a 4 hour period starting at 7am.

25.4% was at 7:51am and 27.3% was at 9:08am.  My weight was 203.5lbs and 204.2lbs respectively.

These % body fat calculations fail a basic sanity test: they imply a 3.86 lb weight change as opposed to the observed 0.7 lb one.  This is an order of magnitude off.  And to Fitbit's credit if they wanted to fudge the results they could have easily taken this into account.

You imply in your previous post (and in another thread) that over-time the Aria may help me figure out if I am gaining, or losing fat. 

Yet, I see threads where people who believe in the accuracy of the Aria's % fat calculation and are trying to make decisions about their weight loss based on the results they're "seeing" as opposed to realizing that they have to seek accurate measurements in a different method (the Aria is just as accurate as any other scale that uses 2 electrode biometrical impedence to measure % fat).

Unfortunately individual weight loss results vary over an extremely wide spectrum.  People often experience relatively common 3:1 or 4:1 fat to musle losses. But they also experience extremes such as "OH CRAP" 1:1 losses, or even "OGM, OMG" 20:1 fat to muscle losses!

The Aria has claimed that I was 26.5% fat at 204.8lbs and 25.7% fat at 205.3lbs.  Luckily these were consecutive day measurements and since I knew what I'd eaten I was well aware that I hadn't actually gained lean mass and that the sanity test had failed.

If I was only weighing myself once a week, or once every 15 days, I could have EASILY believed that I gained some muscle because of my exercise program.

I suppose that if you have no other option you can try to dissern a trend out of Aria's results.

Thankfully I do have another option and Aria's % results are only a distraction to me which is why I said  that I wish I could turn them off (FOR ME).

[TandemWalker]

I don't have an Aria scale, but I do have a digitial scale that uses similar technology to measure body fat. And yes, it too seems to vary widely, not only day-to-day, but I've noticed that the body fat ratio varies a lot if I weigh myself many times during the day.

As @Dominique  pointed out, I think that the trend over time is where these 'gadgets' find their niche as opposed to looking for accuracy in specific readings.

I liken this to the sleep tracking functionality available in many Fitbit trackers. Some folks are looking for a sleep lap encapsulated! It's not, of course; and it's not very accurate. But over time, it will indeed offer some valuable insights about one's sleep quality and improvement. The same is true with the Aria or similar scale that offer body fat 'estimates.'

So used in that context, to gain insights and observe trend over time, these functions do have a lot of merit in my view. I know my scale is not accurate, but I make an entry into my spreadsheet for every reading; and once in a while, I look at the graph and see if the trend is going north or south.

Over the past 18 months or so, my body fat went from 28 down to 25. Sure, this last reading hovers between 24-26; but on average for the last week, it's 25; and it's been 25 for the few months or so. That's the kind of insights I get from it. Never once has the scale showed me that 28 number over the last three months or so. 

One more comment regarding the 'fat to muscle losses.' Muscle loss, in day-to-day life, simply does not happen. You cannot burn muscle. You can make muscle grow and shrink, but you don't burn muscle unless you have reached a state of starvation. The body goes for carbs first, then fat, then muscle. So it's only when your body has completely exhausted carbs and every ounce of fat that your body will switch to survival mode and start using muscle to sustain life.

My take.

TW

[GRRRRR]

Ha! I just realized how much this thread has drifted (sorry Dominique as you're the OP!)

TW I was with you till you discuss fat vs muscle loss. Perhaps just a definition issue as you also talk about muscles shrinking and burning? Do you exclude weight loss from your defnition of "day-to-day" life?

There is conflicting information on just about everything when it comes to weight loss; but, I don't see much disagreement that during weight loss you lose both fat and fat-free mass.

If anyone has more appropriate references than the ones below, please chip in. I don't have any background in any of this other than what friend Google provides (or stops providing) as her whims dictate!

"However, a concomitant decline in lean tissue can frequently be observed" http://www.ncbi.nlm.nih.gov/pubmed/16526835

"The degree of caloric restriction, exercise and rate of weight loss influence the proportion of weight lost as FFM after non-surgical interventions" http://www.researchgate.net/publication/6720115_Changes_in_fat-free_mass_during_significant_weight_l...

"These data provide evidence that exercise training reduces the amount of body weight lost as fat-free mass during diet-induced weight loss" http://www.scopus.com/record/display.url?eid=2-s2.0-0027958215&origin=inward&txGid=42B04201BADEF9EBF...

"strength training significantly reduced the loss of FFM during dieting but did not prevent the decline in RMR"  http://www.ncbi.nlm.nih.gov/pubmed/9280173  

"the diet plus strength training group showed a trend toward preservation of lean mass, and the diet plus aerobic group in men resulted in increased testosterone concentrations" http://www.ncbi.nlm.nih.gov/pubmed/25599089

Even if some of the fat free mass is glycogen and water which is regained once the caloric deficit goes away when you discuss several kg losses some of it has to be lean muscle tissue!

If nothing else, even if none of the muscle tissue is metabolised to provide energy, what happens to your muscles when you no longer carry an extra 50lbs 24/7/365?

[TandemWalker]

@GRRRRR - Perhaps it is a matter of definition. However my opinion remains the same.

You can grow and shrink muscles; but you do not BURN muscles unless your body is completely depleted of nutrients (carb, fat and protein). Once you get there, you are in fact in a state of starvation. Nature will then follow its course and you will indeed start chewing away at the muscle mass, the last and only source of energy left. A person will die before that's depleted.

The statement "There is conflicting information on just about everything when it comes to weight loss; but, I don't see much disagreement that during weight loss you lose both fat and fat-free mass." taken as is without knowing the context in which it was formulated, is just wrong. I lost over 50 pounds over the last 18 months; and I can ASSURE you that I gained muscle mass - arms, legs, chess, abs ...

So yes, I gained muscle mass in tandem with weight loss. Why? Because I follow a balanced diet that is  commensurate with my energy needs. I also follow a balanced exercise regimen, with a good mix of cardio and resistance training.  But my intent was never to become a bodybuilder - I just wanted to change my bowling-pin physique to something a little more attractive!

Interesting discussion. And for the record, I agree with the quotes you have there, at least for the most part, when taken in context as opposed to stand alone statements.

TW

[GRRRRR]

TW: congrats on the body change and you are absolutely correct: I failed to fully qualify my statement : - )

What I INTENDED to say is: "There is conflicting information on just about everything when it comes to weight loss; but, I don't see much disagreement that during weight loss you lose both fat and fat-free mass UNLESS YOU MAKE AN EFFORT AND HIT ON THE SECRET FORMULA to avoid that!".

It SEEMS from what I've read to date that the "effort" should consist of eating sufficient amounts of protein (insert discussion as to what is sufficient, and whether the type and timing matters) and engaging in exercise (insert discussion as to what type, how much, intensity, relative risks of injury and non-compliance, existing injuries or impediments, etc).

And insert into the discussion the level of caloric deficit, the weekly % weight loss, the amount of fat available to be lost, metabolic damage from previous dieting, disease or physical impediments, age and gender, etc, etc.

Even insert into the discussion whether you should engage in exercise while catabolic to maintain muscle mass, or whether to wait for weight loss to concluded to benefit maximally from newbie gains in an anabolic state.

Or to complicate things how about intermittent fasting? Fuel your workouts for maximal muscle gain and use short term fasting for caloric deficit!

And, THREE, FOUR, or FIVE years from now, which method will have generated the best results for you?

http://www.nwcr.ws/research/

http://www.nwcr.ws/research/published%20research.htm

But this is morphing more into something I would expect to find in MFP's as opposed to Fitbit's boards : - )

For whatever it's worth I am happy as a clam because my last 3 months and 21.3 lbs were 20.2 fat and only 1.2 lean.  I wish I had known DXA scans existed when I first started all this a year ago!

[DaffyLisa]

The trend will be down.

For it to be meaningful it would have to be down by a similar amount of
body fat as your dexa scan will show you as losing.

Aria's 20% = certain lbs of fat.
Dxa's 27% also = certain lbs of fat

My own experience is that the aria does not give me meaningful results as
to the amount of fat in my body at a point of time or relative to previous.

If I believed it, for example, there would be times I lost 10:1 fat to
muscle when I would have thought I was losing way more muscle. The
opposite is also true when i lost 1.8:1 based on DXA at some point.

I don't need a confirmation from body fat % to indicate that I've lost an
unknowingly imprecise % of fat...

The scale by itself tells me that much on its own.

Even pushing the results to trendweight or weightgrapher says nothing that
ever corresponded to a series of 5 dxa scans over 10 months.

[BrentTX]

I weigh myself every morning at same time, same conditions, etc. (day to day it varies usually +/- 1% or less)   Aria showed me 20.5% body fat on the morning I got a DEXA scan. I had DEXA scan that afternoon and it showed 27%.  

Aria now shows me at 15.9%.  I'm planning another DEXA scan so we'll see if margin of error is still ~7% and/or if at least the trend downward is consistent between Aria and DEXA.