Research results have been released indicating that body mass index (BMI) does not help predict health risks. There are also projections that the era of diagnosing obesity with BMI may come to an end.
A research team from the University of Florida published results on the 24th (local time) in the international journal "The Annals of Family Medicine" analyzing how accurately BMI and bioelectrical impedance analysis (BIA) predict health risks.
BMI has long been used as a diagnostic criterion for diagnosing obesity. It is a method of estimating fat mass by dividing weight by the square of height, first proposed in the 19th century, and began to be widely used when the U.S. insurance industry started measuring health risks with BMI in the mid-20th century. In the 1990s, the World Health Organization (WHO) began to utilize BMI for diagnosing obesity, leading to its extensive use in medical practice.
However, because BMI uses only weight and height, there have been ongoing criticisms that it is difficult to accurately diagnose obesity. For instance, athletes with high muscle mass and low fat content may be classified as obese simply because they weigh more. For this reason, the medical and insurance industries have been seeking various alternatives to establish new standards for obesity to replace BMI for a few years now.
The researchers at the University of Florida confirmed whether BMI actually helps predict health risks. They analyzed data from 4,252 individuals registered in the National Health and Nutrition Examination Survey (NHANES) in the U.S. from 1999 to 2004. The age range was 20 to 49 years. They analyzed how BMI scores changed over 15 years and how well these scores predicted health risks based on death certificates registered as of December 31, 2019.
The study found that BMI failed to accurately predict all kinds of health risks that can lead to death. The researchers noted, "BMI did not show a statistically significant relationship with mortality from all causes" and emphasized that, "BMI, calculated using an individual's height and weight, cannot distinguish between muscle and fat, and only indirectly estimates body fat, making it inaccurate for everyone."
In place of BMI, the researchers proposed bioelectrical impedance analysis (BIA) as an alternative. BIA is a method of measuring body composition based on the speed at which an electric current passes through the body. It uses the fact that fat tissue has higher electrical resistance than non-fat tissue.
In fact, BIA was found to provide more accurate assessments of health risks than BMI. Individuals identified as having a high body fat percentage through BIA had a 78% higher mortality risk from any cause compared to those with healthy body fat levels, and the probability of dying from heart disease was about 3.5 times higher.
The researchers stated, "BIA showed a statistically significant relationship with most causes of death or heart disease, excluding cancer deaths."
References
The Annals of Family Medicine (2025), DOI : https://doi.org/10.1370/afm.240330