By far the most common question I get asked by customers and others is about the Bio-electrical Impedance Scales.
You might know these scales in a couple of formats: either the home bathroom scales which will also tell you your body fat, sometimes a handheld device that you grip with both hands to get a reading, or you might have come across the more advanced versions which are springing up in gym changing rooms, or indeed out on the gym floor which include both a scale component and a handheld component.
Collectively these scales all work on the same principle and have some features in common. They all run a small current through your body, from one of the points of contact (there will be metal pads that you either stand on or hold or both) and based on the amount of electricity which is impeded (stopped) by your body, it calculates your body fat percentage. Sounds great right? Alas, as with all these more readily accessible methods, when you start to dig into the research, issues start to arise.
How do they calculate your body fat?
All the readings on these scales have been compiled by the equipment makers by testing a whole lot of people with other methods such as DEXA and Hydrostatic weighing, and running a test current through the same subject and equating the readings. For example, they take a reading from a study participant, a hydrostatic weighing measurement reported that the subject has 12.5kg of fat mass, and 60kg of lean mass. The machine makers would then run a current through the body of that subject – and find some of that current was impeded. They measure the amount of the impedance, and then, normalising for weight, sex, age, racial variables, be able to say that measurement of impedance means an amount of Total Body Water (TBW) and assuming that the fat-free-mass has a fixed hydration level, they can make the jump to say their measurement of the subject’s impedance equals 60kg on lean mass.

So what does the research say?
Over the last 20 years, a massive amount of research has been done into these scales. The medical profession, understanding that BMI is pointless, and that detailed and accurate body composition figures are what is necessary to review their clients’ health, have been searching for a method which is easily and widely available, relatively low-cost and can be done in doctors offices and the wider community. While some of the results have been promising, BIA scales aren’t the holy grail of mass-availability of detailed body composition that the industry has been looking for – at least not yet.
So what are the issues?
In coming up with a total body fat, the calculations start by calculating total body water. There are a lot of assumptions taking place in this step. Not everyone will be equally hydrated for example, but some assumptions are a factor in every type of test we do. More importantly, as the machines don’t measure anything detailed directly, all they can say reliably is that their data correlates very well with actual measures like DEXA. But this correlation is highly dependent on the population you are comparing it to.
A 2008 study found that BIA was reasonably accurate only when it was paired with a specific population that matched a given subject on attributes of sex, age, weight, height, activity level, historic activity level, proportional limb length and race (that’s right, even having slightly longer or shorter arms/legs will have an impact on the reading) It concluded that without this matching, it was not a suitable method to test someone in wide population level studies.

Another 2017 study found that its accuracy of testing progression was also questionable. It used BIA and other methods to see how the same group tested, as they engaged on a weight loss journey. They found that for neither sex did the BIA give accurate results of their fat mass lost or muscle mass.
All this bears out what I always say to my clients – maybe one day, but at the moment BIA just isn’t accurate enough. It is worth noting also that these studies were conducted using the very best BIA technology available. This is the big machines you might find in gyms which usually cost in the range of £5k plus. Virtually no independent scientific testing has been done on the £200 set of bathroom scales which you might be tempted to buy. These machines are a total unknown.

There are a couple more points to make about the BIA scales when compared to DEXA.
BIA is a 2-compartment model. That means it divides the body into two compartments: fat, and everything else. So wrapped into the assumptions made, is that you have a certain fixed density and quantity of bone. DEXA is a 3-compartment model: fat, bone, and everything else.
BIA can only collect information from 2 points, or 4 points on the body. If its just a scale-type then it will run a current from your left foot, and collect data at your right foot. If you are using a more advanced version it might run the current up your left leg and collect at the right hand, or vice versa. It doesn’t know where the fat is, or how much of it there is. It just knows that somewhere between your left leg and right arm there is x amount of fat.
By comparison, on a DEXA Scan, every pixel on that image has a value, and that value will tell us how much fat, muscle and bone is in that pixel. This is how we are able to tell you that you have a given amount of muscle in your left arm vs your right, or how much fat you have around your organs (Visceral adipose tissue). We have literally thousands of data points in a DEXA scan.
So, the only ready way to get accurate measures is to come in to have a DEXA scan. I would say you can throw your BIA device away, but I do think it is good for one thing – i’m sure it will still accurately tell you how much you weigh.
Any questions or comments about anything in this article? post them below, or send us an email, or indeed give us a call. Most importantly however, get your DEXA baseline booked in today!
