Measuring Body Fat
by Michael Lombardi, NIST - October 2010

People are sensitive about their weight. I learned that lesson the hard way in the third or fourth grade, when our regular teacher was sick. The substitute teacher seemed nice, but she was much older and bigger than our regular teacher, who was a very small woman who had probably just left college. At recess, the boys in my class were discussing the “differences” between the two teachers, and one of them dared me to ask the substitute teacher how much she weighed. I knew she weighed a lot and must have known there was danger involved, otherwise it wouldn’t have been a “dare,” but I was too young and naive to realize what a rude and insensitive question that was. After all, the only women I had ever talked to were my mom and my grandmothers; I didn’t have a sister, and for the most part freedom of speech was practiced in my Italian-American household. Thus, I ran over amiably with a foolish smile on my face to address the teacher. About one second after I popped the question, the teacher popped me with a right hook that would have made Joe Frazier proud. When I finally stopped crying, it was clear to me that asking people how much they weighed was hazardous to your health.

To this day, it seems to me that weight is a dangerous topic of discussion, perhaps more dangerous than religion or politics. So against my better judgment, and with great fear and trepidation, I’ve decided to tackle a very controversial topic – the metrology of body fat. This could turn out badly, but I’ve always been a sucker for a dare.

Body fat is measured as a percentage of a person’s total weight. The idea is to find out how much of the body is fat and how much is lean body mass, which basically consists of muscle, bone, and tissue. My substitute teacher didn’t appear to believe it, but fat isn’t necessarily bad. In fact, some body fat is essential for a person to stay alive, or to maintain reproductive functions. Woman require an essential amount of body fat ranging from about 10 to 13 %, whereas men only require about 1 to 3 %. However, not even the most athletic people have body fat percentages that are that low, meaning that almost everyone is carrying more fat than they absolutely need to survive. Today, the average woman has body fat ranging from about 25 to 31 %, and the average man has body fat ranging from about 18 to 24 %. A woman or man who is considered to be in top physical condition would have a body fat percentage about 10 percentage points lower than the average, typically under 20 % for a women or under about 13 % for a man.

Excess weight can cause a lot of health issues, and thus body fat measurements are an important medical topic. A great deal of research and effort has gone into the measurement of body fat percentage, and the amount of available literature exceeds that of many other areas of metrology. Any type of body fat measurement will provide a general idea of a person’s body composition, and more importantly, will inform a person who is trying to reduce their body fat of their progress. Even so, all methods are just approximations, with fairly large uncertainties. It seems that the four most popular measurement methods for determining body fat percentage are hydrodensitometry weighing (underwater weighing), anthropometry-skinfold (calipers), bioelectrical impedance analysis (BIA), and dual energy X-Ray absorptiometry (DXA). Let’s take a brief look at each of these methods.

Hydrodensitrometry, or underwater weighing, is based on the fact that fat floats and lean tissue sinks. That fact is easy to believe, as many of us will note that floating is easier now than it was when we were kids. The equipment available for this measurement ranges in sophistication from the standard stainless steel tank with a chair or cot mounted on an underwater scale, to a chair and scale suspended from a diving board over a swimming pool.

The person under test (PUT) is first weighed outside of the water. They are then placed in a chair and lowered into the water until they are completely immersed, with all of the air exhaled from their lungs, and weighed again. To get an accurate measurement, the process is typically repeated at least three times. A lower immersed weight indicates a higher body fat percentage. For most PUTs, this method has a fairly low uncertainty, typically 2 or 3 %, and for years was considered the best way to measure body fat percentage.

Underwater weighing has several disadvantages, however. For one thing, the equipment is highly specialized and not always available, and not everybody is OK with the idea of total submersion. The method also assumes that the densities of fat mass and fat-free mass are constant for all PUTs. Not true. For example, athletes tend to have denser bones and muscles than non-athletes, which may cause their body fat percentage to be underestimated. In contrast, the body fat of elderly patients suffering from osteoporosis may be overestimated. For an accurate measurement, it’s also important to know if the PUT has expelled all of the air out of their lungs, and for the water to be completely still, without any wind or movement.

Anthropometry-skinfold measurements are more convenient. They are done with hand held calipers that exert a standard pressure. Measurements are made by grasping the skin and underlying tissue, shaking it to exclude any muscle, and pinching it between the jaws of the caliper. The test is performed at three to seven body locations on the PUT where fat normally accumulates, such as the back of the arm, waistline and hip. Skinfold measurements assume 36 Metrologist : October 2010 that the fat found under the sampled body locations accurately represents the proportion of fat found throughout the rest of the body. Each body location is typically measured from two to three times to reduce the uncertainty. A simple, less accurate variation of this test is known as “pinching the love handles.” It is performed daily by nonmetrologists in front of hundreds of thousands of bathroom windows, worldwide. Most practitioners declare themselves to be out of tolerance.

After the skinfold measurements are recorded, a calculation is used to derive a body fat percentage based on the sum of the measurements. Different prediction equations are needed for children and specific ethnic groups, and believe it or not, over 3,500 equations have been validated! The underwater weighing method was used as the reference for many of the prediction equations.

Skinfold measurements have several advantages: they are easy to do, inexpensive, and the equipment is portable. The quality of the calipers affects the measurement uncertainty, they should be accurately calibrated and have a constant specified pressure. Inexpensive models sold for home use are usually less accurate than those used by professionals. The final uncertainty is typically about 2 to 4 %, and depending on the skill of the technician performing the measurement, can be about as reliable as underwater measurements.

Bioelectrical Impedance Analysis (BIA) measures body fat percentage by sending a low-level, safe, electrical current through the body. The current travels at a different rate through the various body tissues. It passes easily through muscle tissue, which contains a large amount of fluid, but it travels more slowly as it passes through fat tissue. The resistance encountered when the current hits the fat tissue is called bioelectrical impedance.

With the exception of the informal “pinching the love handles” method, the BIA method is the most common type of body fat measurement performed at home. Commercially-available devices that perform this measurement first appeared in the mid-1980s. Today, so many have been built and sold that they can sometimes be purchased for less than $10. All you have to do is enter your gender, height, weight, and age, and these devices estimate your body fat percentage. BIA analyzers are even built-in to many digital scales.

The BIA measurements are simple, painless, and fast, but may not be especially accurate. Accuracy is often compromised with extremely lean or obese people, and to get a good reading you have to pay special attention to the level of fluids in the body. For example, measurements should not be made if you have had a meal in the last two hours; or if you have just exercised vigorously or taken a bath or shower.

The most accurate method of measuring body fat is normally dual energy X-Ray absorptiometry, or DXA. These measurements can normally only be done in a medical clinic, since they involve a whole-body scanner that aims two X-ray beams with differing energy levels at the patient’s bones. The scan lasts for about 20 minutes. When soft tissue absorption is subtracted out, the bone mineral density can be determined from the absorption of each x-ray beam by bone. This technique is typically used to diagnose and follow osteoporosis, but the absorption readings are so accurate that the DXA has probably now surpassed underwater weighing as the new “gold standard” for body fat measurements, with lower uncertainties for most PUTs. Since it involves a whole body scan, DXA also can tell us where our fat is located, on the off chance that we don’t own a mirror.

Since I am too cheap or too scared or too much in denial (take your pick) to get my own body fat measured, allow me to close with another pointless anecdote. A friend of mine grew up in a modest house in the Denver suburbs that his dad originally bought for about $10,000. A few decades later, the average house in Denver had become much larger, and my friend’s house was considered even more modest than it had been originally. But now it was worth a lot more. While I was visiting my friend’s dad a few years ago, he told me that “I always dreamed of living in a $100,000 house. I just never thought it would be this one.”

Such it is with weight. As a scrawny 130 lb teenager, I dreamed of being a big, strong 200 lb man. A few decades later, I finally obtained that 200 lb body. I just never thought it would be this one.