Share on Pinterest - Researchers investigated the accuracy of Body Mass Index (BMI) in predicting obesity among adults aged 20-59 years old.
- They found that BMI predicted less than half of people with obesity as predicted by body fat percentage.
- The findings indicate that supplementary measures to BMI should be taken to calculate healthy weight.
Body Mass Index (BMI) is an indicator of weight categories ranging from underweight to obese and is widely used to assess the risk of various conditions. It is calculated by dividing a person’s weight by the square of their height.
However, increasing research suggests that BMI is insufficient for predicting body fat percentage. This comes as it does not account for fat distribution around the body, muscle mass , bone density , and racial and sexual differences.
Further investigation of the limitations of BMI as a health indicator could inform more effective strategies for assessing weight and overall health.
Recently, researchers investigated the accuracy of BMI measures in predicting obesity.
Whereas only 36% of participants were classified as obese by BMI, close to 3 in 4 were considered obese according to measures of total body fat percentage.
Dr. Wang Lushun , Senior Consultant Orthopedic Surgeon in Singapore, not involved in the study, told Medical News Today :
“The study underscores the limitations of BMI as a singular measure of health, highlighting a possible need for a more comprehensive system. Such a system should factor in potential racial or ethnic disparities for a more accurate health assessment of each individual.”
Researchers presented the findings recently at ENDO 2023 , the Endocrine Society’s annual meeting in Chicago.
Using BMI as an indicator of obesity
For the study, the researchers analyzed data from 9,784 participants with an average age of 39 years old from the 2011-2018 National Health and Nutrition Examination Survey (NHANES).
Data included BMI and body fat percentage calculated by dual-energy x-ray absorptiometry (DEXA). BMIs were grouped into ethnic-specific categories.
Ultimately, 36% of participants had a BMI of 30 or above, indicating obesity. However, when assessed according to body fat percentage, 74% of participants were considered obese.
Many participants with a BMI indicative of healthy weight- 18.5- 24.9- were classified as obese when accounting for body fat percentage, including:
- 27% of non-Hispanic Black people
- 44% of non-hispanic white people
- 49% of Hispanic people
- 49% of Asians
“It was very interesting to see that non-Hispanic Black people with overweight had 58% lower odds of obesity compared to non-Hispanic White people. This could be due to typical genetic weight distribution,” Dr. Erkeda de Rouen , a board certified physician in family, diversity, and lifestyle medicine, not involved in the study, told MNT .
Study limitations
MNT spoke with D r . Aayush Visaria , Internal Medicine Resident Physician and Public Health Researcher at Rutgers Robert Wood Johnson Medical School, one of the study’s authors, about its limitations.
“While DEXA scans are quite sophisticated, they cannot differentiate subcutaneous from intramuscular from visceral fat . Visceral fat is generally the most harmful so future studies would need to selectively measure that using MRI or CT scans,” he said.
“In our analysis looking at racial/ethnic differences, we did not account for possible confounders such as dietary habits, physical activity patterns, sleep health, all of which can increase risk of obesity and may drive the differences seen between races/ethnicities,” he noted.
Dr. de Rouen said: “[The researchers included] a wide range of participants, ranging from 18-59, yet did not reveal their age distribution. As age increases, the metabolism does the opposite. The gender disparity could also have affected the body composition, with women typically having more adipose tissue in the thigh and waist areas.”
Dr. Yvonne Covin, an Internal Medicine physician in Dallas, Texas, not involved in the study, told MNT :
“While the gender distribution was balanced, the sample size for race/ethnicity was skewed, with non-Hispanic whites representing 61% of the sample. This raises concerns about the generalizability of the findings to the larger US population, particularly for ethnic minorities. Therefore, we must exercise caution when extrapolating the study’s findings to underrepresented ethnic groups.”