The right way to tell obesity is key to our well-being
BMI provided a shaky foundation for understanding obesity. A committed policy response must now accompany a shift to more accurate anthropometric measures
Over the past half-century, obesity has emerged as a vexing global health challenge — both because of the several adverse health effects and struggles with understanding the nature of a body’s “fatness”. Even as the prevalence of obesity was rising rapidly across the world, health professionals were unclear as to whether obesity should be labelled as a disease in itself or merely as a risk factor for other diseases. A distinction was notionally drawn between “overweight” and “obesity”. Recognition of amplified health risks posed by visceral or abdominal obesity, as distinct from the risk posed by general obesity, brought a new dimension to debates around obesity as an omnibus term.

As scientists sought to define obesity through quantifiable anthropometric measures, debates raged over whether the problem arose from personal behaviours of eating and exercise or from commercial forces that propelled changes in patterns of food consumption through their manufacturing and marketing practices. Attributing obesity to an individual’s gluttony or indolence led to body shaming and social discrimination. And now, with powerful new weight-loss drugs emerging recently, calls to treat obesity as a disease have become louder.
Amidst these debates, the Body Mass Index (BMI), proposed by Belgian mathematician Adolphe Quelet in 1832, became the anthropometric index considered cardinal for obesity identification. BMI is derived by dividing weight (in kilos) by the squared value of height (in metres). In western populations, a BMI between 25 and 30 was classified as overweight and values above 30 as obese. “Normal” BMI ranged between 18.5 and 25, while BMI below 18.5 was classified as being underweight. The World Health Organization (WHO) recommended these standards for all populations.
As BMI is a crude measure for identifying excessive body fat, it led to inaccurate labelling of individuals. This is because body weight is a composite of weights contributed by fat, muscle, bones, and body fluids. A person who is very muscular or has high bone density will have a high BMI even without a high-fat mass. Such a person would be falsely labelled as obese. Similarly, a bloated person who retained body fluids due to heart or kidney failure would be improperly labelled as obese even without excess adiposity.
Across different populations, high BMI is associated with a high risk of several health disorders. Obesity was a risk factor for diabetes, heart attacks, brain strokes, and cancers of several organs (especially the colon and breast), apart from joint disorders and impaired breathing. Even when diseases were not clinically manifest, abnormal patterns of blood pressure, blood fats, and blood sugar were seen in many obese and overweight individuals.
However, it also became clear that the western BMI scale did not apply to Asian populations while attempting to predict the risk of health disorders. This was initially observed among Indian, Pakistani, Bangladeshi, and Sri Lankan immigrants in the United Kingdom, the United States, Canada, Africa, and South East Asia. Persons of South Asian ethnicity were noted to be at a high risk of developing diabetes and coronary heart disease, even within the “normal” range of BMI. These findings were later replicated in their native countries. Similar observations came from studies on Filipinos and Chinese. The WHO then recommended lower BMI thresholds for overweight (between 23 and 27.5) and obesity (above 27.5) for Asian populations.
The objective of using BMI to identify individuals who are at a high risk of obesity-related diseases was ill-served even in non-Asian populations. Within western populations, there were individuals who met the BMI criteria for obesity but did not either manifest the disease or have the metabolic abnormalities associated with excess body fat. It also did not help that different types of fat cells (white, brown, and beige), with distinctive metabolic and inflammatory effects, were variably distributed across the body at different ages.
Why was BMI an inefficient predictor of diseases associated with body fat? The answer lies in the patterns of body fat distribution. It was observed that fat around the abdomen’s internal organs was associated with a higher risk of diabetes and heart disease than fat around the hips. Pear-shaped bodies were healthier than apple-shaped bodies. This led to the use of waist-hip ratio (WHR) as a measure of visceral adiposity.
Accurately measuring the hip circumference is not easy. In recent years, the waist-height ratio (WHtR) has emerged as a better measure. This ratio was originally used in children but was later validated for adults as well. A WHtR exceeding 0.5 predicts a high risk of disease. The Roundness Index (RI), a recently proposed measure, takes into account height, weight, and waist measurements to generate computed body images. These images are compared with the ellipsoid norm of the erect human body. This measure is more suitable for research.
A recent Lancet Commission report on obesity helps us to cut through the clutter and identify clinically useful categories based on risk assessment. The report identifies obesity as excess adiposity. Endorsed by 76 international organisations, it classifies obesity into pre-clinical and clinical categories. Body fat is measured by BMI, WHtR and/or X-ray techniques. Excess body fat is classified as “clinical” obesity if symptoms like breathlessness limit activity or laboratory tests show abnormalities of blood sugar, blood fats or liver function. Clinical obesity would be treated as a disease. Obesity, detected by anthropometric or radiological measures, would be classified as pre-clinical if there is no clinical or laboratory evidence of compromised organ function. Such persons would be periodically monitored to detect progression to clinical obesity. Any person with a BMI of over 40 would be treated as obese.
This classification will help guide clinical management by first identifying body adiposity and then assessing organ health. At the population level, a lot more needs to be done to promote healthy diets, regular physical activity, better sleep patterns, and less exposure to air pollution. This calls for a composite and committed policy response that goes beyond the clinic.
K Srinath Reddy is a distinguished professor of public health, PHFI, and the author of Pulse to Planet.The views expressed are personal
