Most of us are no stranger to the body mass index: weight in kilograms divided by height in metres squared.
At a population level, research tells us that having a higher BMI is associated with a greater risk of certain conditions, such as type 2 diabetes and high blood pressure. Rates of obesity, according to the World Health Organisation, have tripled globally since 1975.
But despite being enthusiastically adopted in doctors’ rooms and also by average people to quantify their own body composition, BMI is much less useful as an individual health indicator.
Australian experts, who discussed the usefulness of BMI and possible alternatives at the International Congress on Obesity in Melbourne last month, say health judgments based only on BMI can be stigmatising and potentially harmful. They are now calling for more nuanced measures beyond simply taking a number at face value.
Risks affected by ethnicity
BMI is an imperfect proxy for body fat. “Risks of certain health conditions do change when people gain weight, but on its own BMI doesn’t tell us much about the health of the individual,” says Dr Priya Sumithran, who leads the obesity research group at the University of Melbourne. “It’s hard to define a BMI cutoff for optimal health.”
The concept of BMI originates from Adolphe Quetelet, a 19th-century Belgian astronomer and mathematician whose goal to describe “l’homme moyen” – the statistically average man – influenced the development of eugenics.
Initially called the Quetelet index, the term BMI was coined by researchers in 1972 who described the measure as “not fully satisfactory” but preferable “on the simplicity of the calculation” and “at least as good as any other relative weight index as an indicator of relative obesity”. Fatness as a medical issue arose in the west only in the past century, with American life insurance companies collecting data on weight and height and linking the information to mortality.
Under official guidelines adopted by the WHO, a BMI less than 18.5 is considered underweight. A score between 18.5 to 24.9 yields a “healthy” weight, 25 to 29.9 is overweight, and a score of 30 and above is considered obese. There are many caveats to this classification, including that it is less applicable for tall, short and muscular people.
The health risks related to obesity are also affected by ethnicity. In comparison with Europeans, people of Polynesian background have lower body fat levels at the same BMI. The metric also overestimates obesity in African-Americans. For Asians, the health risks linked to obesity occur at a lower BMI, which has prompted countries such as Singapore to revise their guidelines to classify 23 and above as overweight.
Obesity can also be defined as an excessive accumulation of fat that presents a risk to health, Sumithran says. “There will be people whose BMI is above 30 but who don’t have unhealthy fat accumulation.”
Another shortcoming of the index is that it does not take into account fat distribution. “If it’s more centrally distributed, if it’s around your internal organs, it’s going to have more health consequences than in the classic pear-shaped distribution around the buttocks,” says Prof Louise Baur, president of the World Obesity Federation.
Baur, who is also chair of child and adolescent health at the University of Sydney, says BMI is still a useful measure for illustrating population trends over time.
“We know, for example, that the percentage of people with a high BMI … has increased quite dramatically over the past few decades from what it was, in Australia and many other high-income countries,” she says. “In South Africa, women are more likely to have a high BMI than men … in China, boys are more likely to have a high BMI than girls of the same age.”
A Lancet commission for diagnostic criteria for clinical obesity – an international panel of experts including Baur – is currently assessing the use of BMI and discussing new ways to diagnose obesity, including whether there are better measures of body composition.
‘You can’t assess nutrition via body size’
Dr Alex Craven, an obesity surgeon from Austin Health in Melbourne, is concerned by what he sees as an over-reliance on BMI as a single indicator of health for individuals.
“For some reason with obesity, we accept that we can give people advice based entirely on one number … to say: your BMI is this, therefore you’re automatically unhealthy,” he says.
While BMI can be useful, using it in isolation “would be the equivalent of your GP taking your heart rate … and, not looking further, giving you a diagnosis and a medication based on that”, Craven says.
Calls to adopt other measures of body composition are not new, and the persistence of BMI likely owes to the simplicity of its calculation.
“BMI is pretty darn simple: a scale and a measuring tape is all we need,” Craven says. “Just because something is convenient, [it] doesn’t automatically infer quality.”
There are alternative metrics, such as waist-to-hip ratio, which has been found to be a better predictor of cardiovascular disease (the downsides: waist measurements are hard to take accurately, and Baur says the ratio doesn’t work for children), and waist-to-height ratio, which may better predict mortality risk.
The Edmonton Obesity Staging Score measures the impact of obesity and takes into account conditions including diabetes, high blood pressure and osteoarthritis.
“I can teach my patients to use it in under five minutes, I can teach junior doctors and nurses and physicians to use it very quickly,” Craven says. “The only drawback is that you have to approach your patients with a bit of curiosity and ask questions about things that are often more important to them than their weight anyway.”
Dr Fiona Willer, a dietitian and lecturer at the Queensland University of Technology, is frustrated that a “panic about body weight” and rises in obesity rates in recent decades have led to public health messaging that prioritises weight control.
“Dietary guidelines talk about weight before they talk about food,” Willer says. “You can’t assess nutrition via body size.”
Focusing so intensely on weight could mean, for example, that inborn errors of metabolism – rare genetic disorders – that affect weight are overlooked.
Willer’s PhD thesis, which studied eating behaviour and size acceptance, found that people whose focus was on a healthy lifestyle, regardless of BMI, had more nourishing dietary patterns.
“The poorest dietary quality – the narrowest diets least likely to meet nutritional requirements – were those who were weight-focused and didn’t care about health,” she says. “They also had the worst levels of body acceptance.”
Baur says some researchers recognise that “in a focus on obesity we unintentionally create an environment which may trigger unhealthy eating behaviours, or make people living in larger bodies feel uncomfortable”.
“There’s a lot of stigma from health professionals that does not help the situation,” she says. Even if a new definition of obesity were to replace BMI, Baur does not believe it would “magically change people’s views of weight stigma”.
“I suspect whatever word we use to describe people of larger body size who’ve got health complications from that, that word is going to be stigmatised in due course, unless we turn around some of the thinking around that at a societal level, and at a health-professional level too.”
