Issues with using only Body Fat % to track your progress

This article is about one number, and one number only – your body fat %!

Obtaining an accurate Body Fat % is a lot more complex than you may think. It is currently the most over-emphasized number in the body composition testing industry.

This article will address:

  • How your body fat % is developed
  • The main types of body fat % testing offered to the public
  • The issues with body fat % as a number
  • Fit3d and it’s body fat % algorithm

 Key points:

  • Stop obsessing about your body fat %
  • ALL Body Fat % measures have some element of error
  • Fit3d girth measurements (esp waist and hips) and Body Shape Rating are a better way to track health progress
  • Photos in your underwear are a good way to supplement your Fit3d Body Scan
  • Don’t forget to track other areas of health too (strength, fitness, power, sleep, nutrition etc)

There are over 50 ways to obtain a body fat % reading. The only way to truly measure your body fat % would be to cut out all the internal and external fat from your body and weigh it. You would need to be dead for this to occur. Every other measurement method in the world is some kind of best estimate. Some estimates/machines are significantly better than others.

In order of accuracy, here are some of the main types (including brief comment) of body fat % testing offered around the world:

1. Full Body Autopsy

Full body autopsy can only occur if someone is dead

2. Full Body MRI Scan

A Full Body MRI costs thousands of dollars per scan, is not mobile, has some radiation risk, and is not available to the public in NZ

3. Full Body DEXA Scan

Aside from MRI, DEXA is quoted as being the Gold Standard for body composition testing in the world. Unfortunately, DEXA has limited availability in NZ, it is more time-consuming, costs a couple of hundred dollars per scan, does not have 24/7 online results access, is not mobile, and generally requires a referral from a health professional. DEXA algorithms themselves were initially designed to track changes in populations/groups of people (not individuals) over time. The standard of error in a DEXA machine has been reported to be approximately 5% e.g. if your DEXA Body Fat % reads 25% then it could actually be anywhere between 20-30%.

4. Fit3d Body Scan

Fit3d Body scanning is quick, low-risk, mobile (e.g. machines can be easily transported), takes multiple measures aside from Body Fat %, and gives you 24/7 online access to results. A more in-depth analysis of Fit3d will be discussed below.

5. Multi-frequency Full Body Bioimpedance Body Scan

There are various bioimpedance scanner models in the world. Bioimpedance scanning involves several currents running through your body. Generally, the higher the number of electrical currents travelling through your body the better accuracy. Essentially these currents travel at different speeds, depending on how much fat, bone, muscle/water you have in your body. One key flaw in ALL bioimpedance testing is that it becomes significantly less accurate when your hydration status changes or if when you are at a different stage in the female menstrual cycle. ALL bioimpedance body fat % algorithms are correlated with DEXA Scan results. Therefore, there is likely to be a measurement error of at least 5% per individual scan (see comments re DEXA scan above).

6. Skinfolds 8-site ISAK method

Until 2010, skinfolds were the most popular method of estimating body fat % in the world. Just to confuse people, there are over 10 different types of skinfold callipers, and over 10 different skinfold techniques and equations. To add to this, the skill of the therapist taking the skinfold measurements can vary greatly. Therefore, there are many areas where measurement error can occur (e.g. the equipment, the therapist, the body fat % algorithm used etc). Therefore, skinfold measurements cannot be accurately relied upon.  

7. Other infrared Body Scanners without fixed handles

Fit3d uses infrared cameras to take over 1200 photos of someone standing in their underwear. The accuracy of each scan relies upon the client standing still throughout the 40-secon test. To assist someone standing still, Fit3d uses FIXED handles. If a machine does NOT have fixed handles, it is impossible for the client to stand still during the testing process. Movement during a scan leads to a blurred/less defined 3d image of the client and less accurate girth measurements. Since infrared body scanners heavily rely upon accurate girth measurements for their body fat % algorithm, if they are not accurate, your body fat % will not be reliable/accurate.  

8. Stand alone bioimpedance scales with no handles

Stand alone scales take an impedance current up one leg and down the other without passing through the upper body and arms. Therefore, they make upper body bone/fat/muscle assumptions based on speed the current travelled throughout the lower body without measuring the upper body at all. Combine this with the other issues discussed above leads to some extremely unreliable test results.

Accuracy is great however time, cost, mobility of machines, accessibility, access to data, and risk all need to be considered when deciding which method to track your body composition.

It should also be noted that you cannot reliably compare body fat %’s between the different measurement techniques. For example, if you have skinfolds, a DEXA, and a Fit3d scan all within 2 minutes of each other, they will all give you different body fat %’s. To minimise confusion, our advice is to stick to Fit3d Body Scanning (or DEXA if it's available to you).





fit3d v dexa graph

Fit3d and Body Fat %

Fit3d has developed their own body fat algorithm, based off data they’ve gathered from their medical research partners that use both the Fit3d ProScanner and a DEXA body fat machine. Fit3D body fat is based off anthropometric measurements (hundreds of body measurements) of more than 1,000 scans for which they also have DEXA scan data. Your Fit3D body fat % is essentially calculated based on your girth measurements, sex, muscle volumes, height, and weight. The exact weighting of each of these measures within their algorithm is unknown.

Unlike some flawed algorithms, your age is NOT included in the Fit3d algorithm. 90% of the time (see graph above), the Fit3D body fat algorithm correlates with DEXA body fat data. Why correlate to DEXA? Aside from costly MRI scans, DXA scans are the gold standard for body composition data, common in medical research facilities, yet not easily accessible for consumers. 

Correlating with DEXA is important however what may be more important is Fit3d's intra and inter rater reliability. Fit3d has found that when scanning the same person 30 times in a row with the same testing technician (intra rater), 98% of the time all metrics were within 1% or 1cm of each other. They also found this to be the same even if there were different testing technicians administering the Fit3d Body Scan (inter).


Why is my Body Fat % increasing?

Your body fat percentage is your fat mass divided by total mass. Lean mass is made up of everything that is not fat, your muscles, bones, and organs. So, if you’re noticeably decreasing your total mass, while your fat mass stays the same, your body fat percentage will increase.

Even if your fat mass (kg) decreases, if your total mass is decreasing faster, this will result in an increase of body fat percentage.

What can lead to measurement error during my Fit3d scan?

Your Fit3d scan operator will do everything they can to ensure you have an accurate test, including sending you pre-test instructions, using an appropriate scanning environment, and continually calibrating the Fit3d machine. However, each individual needs to take some responsibility to scan correctly. The following things can lead to some measurement error:

  • Movement during your scan
  • Tensing or not relaxing during your scan
  • Wearing bulky underwear/clothing
  • Talking during your scan
  • Performing large amounts of exercise in the 3 hours pre scan
  • Eating/drinking large amounts in the 3 hours pre scan

In our own study, we tested 4 people 20 times within one hour and found that all body fat %’s were within 2% of each other and all waist and hip girth measurements were within 1cm of each other.

Extremely Lean and extremely Obese Clients

As with most forms of body composition testing, extremely lean athletes and the obese population are  underrepresented in body composition studies, as is the case with Fit3d. While Fit3d do a "good" job of assessing body fat for these populations, their current algorithm will overestimate body fat % in lean bodies and underestimate body fat % in obese bodies (on  average). This is further exacerbated because of a misunderstanding of how much body fat is necessary for sustaining life (according to the American Council of Exercise) and how much fat actually exists on someones body.   




bfp ranges chart 2019

I’m keen to get my body fat under 5%!

A female requires 10-13% of essential body fat to survive. A male requires at least 2-5%. Unless you are sick, it is unlikely you will ever have a body fat % lower than this. Some people may disagree with this. I would encourage you to ask them if they have ever had all their internal and external fat ripped from their body and weighed. Word-wide less than 0.5% of males/females tested have body fat %’s less than 10% and 15% respectively.


If body fat % is not as accurate as everyone would love, how should I track my body composition over time?

  • Use Fit3d waist and hip measurements. These are raw data with an accuracy rate higher than any tape measure or bioimpedance machine on the market
  • Use the Fit3d Body Shape Rating (BSR). In conjunction with their medical university research partners, Fit3d looked at what measurements on someone’s body puts them at increased risk of cardiovascular disease. The higher your body shape rating, the less cardiovascular disease risk, based on body shape alone. The good thing about the BSR is that it does not rely on an algorithm based on top of another algorithm (like what happens in all body fat % algorithms that are compared to a DEXA machine). Generally, less around your waist and strong less will give you a better BSR
  • Look into the ‘additional measurement’ section of your Fit3d dashboard and compare your trunk volume over time. Generally, a smaller trunk volume will represent less visceral fat and thus decreased cardiovascular disease health risk
  • Your Fit3d machine owner will have access to an ‘overlay’ feature that will be able to compare the outline of your body over time. Visually, this can show you how the shape of your body changes
  • Get someone to take photos of you in your underwear from the front, back, and side
  • Track your clothes sizes on common clothing brands that you wear


How else should I measure my progress?

Body composition is just one small aspect of overall health and well-being. It is recommended that if you are over 30 that you track your diabetes, cholesterol, blood pressure, sleep, stress etc. At any age, you may also want to track your strength, cardiovascular fitness, flexibility, power, agility etc. Numbers are great to put an objective measure on your health however being mindful of how you are feeling should not be underestimated. Remember, you cannot accurately measure everything!


Written by

Peter Halstead

MPhty, PGDipSportMed, BSpEx

Director / Physiotherapist

Health and Fitness Testing New Zealand, Fit3d NZ