What is it?
A great example of a surrogate-gone-mad is the body mass index (BMI). It is a medical screening tool for certain health conditions; it has become the standard indicator for obesity. Colloquially, it is understood as a measure of “fatness”. That is not quite correct, in most people, BMI only correlates with body fat.
The popularity of BMI is related to its easy derivation, it involves only two variables, a person’s weight and height, both of which can be measured precisely. With weight measured in kilograms (kg) and height in meters (m), the formula is simply \[
\text{BMI} = \frac{\text{Weight in kg}}{(\text{Height in m})^2}
\]
Notice that the height is squared in the denominator. If you prefer to work in U.S. pounds and inches, the calculation is \[
\text{BMI} = \frac{\text{Weight in lbs} \times 703}{(\text{Height in inches})^2}
\]
In my case (6’3” tall, 208 lbs), the BMI is (208 )/(75^2) = 25.9. According to BMI charts such as this one at the Cleveland Clinic, I am in the overweight range.
BMI is used widely by medical professionals. People with low BMI values might be at risk for developing anemia, osteoporosis, infertility, malnutrition, and a weakened immune system. High values can indicate a higher risk for heart disease, high blood pressure, type 2 diabetes, gallstones, osteoarthritis, sleep apnea, depression, and certain cancers.
Geez. It seems that unless you are in the optimal BMI range you are either bound for osteoporosis or osteoarthritis. No wonder folks are obsessing over their BMI.
The Cleveland Clinic is quick to point out:
It’s important to remember that you could have any of the above health conditions without having a high BMI. Similarly, you could have a high BMI without having any of these conditions.
…
It’s important to remember that body fatness isn’t the only determiner of overall health. Several other factors, such as genetics, activity level, smoking cigarettes or using tobacco, drinking alcohol and mental health conditions all affect your overall health and your likelihood of developing certain medical conditions.
And
The standard BMI chart has limitations for various reasons. Because of this, it’s important to not put too much emphasis on your BMI.
People who are muscular can have a high BMI and still have very low fat mass. The BMI does not distinguish between lean body mass and fat body mass. BMI charts do not distinguish between males and females, although females tend to have more body fat (says the Cleveland Clinic!). People today are taller than when the BMI was developed. The BMI charts do not apply to athletes, children, pregnant people, or the elderly.
Even though the BMI chart can be inaccurate for certain people, healthcare providers still use it because it’s the quickest tool for assessing a person’s estimated body fat amount.
Ah, so it is used because it is easy to calculate, not because it is particularly useful or accurate.
A January 2025 article in the medical journal The Lancet states (Rubino et al. 2025)
Current BMI-based measures of obesity can both underestimate and overestimate adiposity and provide inadequate information about health at the individual level, which undermines medically-sound approaches to health care and policy.
Based on this article, Business Insider went a step further, calling BMI bogus. Having “obesity” according to the BMI scale does not mean a person is unhealthy. In fact, BMI does not tell you anything about the health of a person.
Origin of BMI
It is worthwhile to examine how BMI came about. The comment above about the increasing height of people suggests that BMI was developed some time ago. Indeed. It was invented for an entirely different reason, to describe a population average man in Western Europe in the 19th century.
To make this connection we need to introduce Adolphe Quetelet (1796–1847), who invented the BMI to quantify a population according to the weight of its persons. It was initially called the Quetelet index. Quetelet was a Belgian astronomer, statistician, and mathematician—not a medical professional. He studied the distribution of physical attributes in populations of French and Scottish people. Quetelet determined that the normal, the most representative value of an attribute, is its average. Prior to Quetelet, the idea of “norm” and “normality” was associated with carpentry and construction. The carpenter square is also called the norm and in normal construction everything is at right angles. The classical notion of ideal as an unattainable beauty up to this time was reflected in great works of art.
Quetelet focused on the middle of the distribution as the “new normal” and saw l’homme moyen, the average man, as the ideal (Grue and Heiberg 2006).
There is no association with health, and there is no association with the individual. The BMI as developed by Quetelet was supposed to describe the average in a population, not obesity of the individual. The population it was intended to describe is French and Scottish of the 19th century. Leaping from that application and that data to near universal measure of obesity since the 1970s is quite the stretch.
