A new study published the Open Access journal Nutrients by researchers from Italy and Lebanon find that over a third of the general population’s weight status in Northern Italy was misclassified when using the body mass index (BMI) classification system. The results of the study demonstrated the significant limitations of the nearly 200-year-old method of determining weight status.

The research is important to re-assess both the accuracy and usefulness of the widely utilized clinical measure of weight, considering it does not factor in age, sex, ethnicity, or body composition.

What is body mass index?

The BMI measure by a Belgian statistician Adolphe Quetelet as an anthropometric measure of the average person across the general population, regardless of age, sex, or ethnicity. The index works by using a simple equation of dividing weight (kg) by height (m). This scale enables the population to fit into an approximate ‘bell-shaped’ curve based on the value obtained, with the middle portion occupied by the most common values, and both left and right ends occupying the least common values.

BMI is now a prominent measure used in clinical settings as a convenient method to determine whether an individual is underweight, normal, overweight, or obese. These four parameters were defined by the World Health Organization (WHO) in 1993 and is based on data mapping and other epidemiological studies on disease morbidity and mortality risk.

BMI is widely used by practitioners in many countries, such as the UK and the US as an initial screening tool to determine weight status and the risk of developing various diseases. However, research demonstrates how BMI presents a skewed reality of actual weight status at an individual level across populations.

BMI misclassifies people’s weight status

The observational study was carried out to assess the BMI and weight status of participants across Northern Italy. The participants were randomly selected to participate in the study, which took place between 2013 and 2024. A total of 1351 participants were included, consisting of mixed genders (60% females, and 40% males), and age groups above 18 years old.

The total and segmental body composition of each individual was measured using a gold-standard dual-energy X-ray absorptiometry (DXA). This measure determined each individual’s body fat percentage.

When comparing the two systems, they found that a significant proportion of individuals in each BMI category were being misclassified. The results showed that:

• 4% of people within the underweight BMI category were misclassified and were considered as normal weight according to DXA.
• 22% of people with normal weight BMI were misclassified, with 9.7% having underweight, 11.4% having overweight and 0.8% having obesity according to DXA.
• 53% of people within the overweight BMI category were misclassified, with three quarters of this percentage being of normal weight, and one quarter having obesity according to DXA.
• Lasty, 34% of those within the obese BMI category were misclassified and were considered as overweight, and a small minority as normal weight, according to DXA.

The results display the striking misclassifications of weight status and body composition across individuals in across the population when using BMI to evaluate weight status. Professor Chiara Milanese, co-author of the study, highlights:

“(…) Even though both systems identify a similar overall prevalence of overweight and obesity, we are talking in some cases about different people – or in other words the individuals identified by DXA are not all the same as those from BMI classification.

This is due to the disagreement between WHO BMI and DXA-derived body fat percentage classification systems in determining weight status in the general population among body weight ranges and age groups of both genders.”

In light of these findings, the authors discuss how global health bodies and governments should review the use of BMI in determining weight status and its role in making key clinical decisions.

The pitfalls of using BMI

BMI does not consider key factors, such as sex, age, and race differences. Importantly, it does not distinguish between muscle, bone density, fat, types of fat, or its distribution. This raises alarm bells on the reliability and even risk of using the nearly 200-year-old medical model.

With BMI used as an important medical guideline globally for around 2 centuries, it is now deeply ingrained into social perceptions of what healthy looks like. Hence, it has played a crucial role in dictating eating habits, levels of activity, general lifestyle and body image for many.

Many researchers and health professionals now raise the question of its usefulness in categorising weight status and hence, its role in screening for risk of disease. In light of these important discussions,

The research presented here confirms that BMI may misclassify weight status and body fat composition, when compared to the gold standard for assessing body composition.

Future research would still be needed to see if BMI misclassifications occur within populations from different ethnic backgrounds and general populations.

Extensive work across medical health boards would also be needed to elucidate more efficient alternatives to classify people’s weight and thus, determine disease risk when making key clinical decisions, such as diagnosis and treatment options.

Read more about research like this published in the journal Nutrients. Alternatively, you can access the full journal list.