Body Mass Index: Not the Best Marker for Obesity

Evidence-Based Diabetes Management, October 2014, Volume 20, Issue SP13

Obesity is no longer an epidemic restricted to the United States—it has become a global phenomenon. While 34.9% of the adult population and 17% of youth in the United States were estimated to be obese based on a 2011-2012 census,1 the World Health Organization (WHO) estimates that 11% of adults around the globe (>20 years of age) were obese in 2008. Additionally, WHO confirmed that nearly 40 million children under the age of 5 years were either overweight or obese in 2012.2

Obesity: The Source of All Diseases?

While the American Heart Association put forth a recommendation to classify obesity as a disease,3 the American Medical Association went a step further and adopted policy to recognize it as a disease in 2013,4 which can open doors for medical intervention to prevent or treat the condition. Why all the fuss over obesity? Because the condition has been identified as the root cause of numerous disease conditions: heart disease, stroke, type 2 diabetes mellitus (T2DM), kidney and liver disease, and cancer.5,6

The bottom line: extreme obesity can reduce life expectancy by up to 14 years, as a result of the secondary disease conditions. A study published in the journal PLOS Medicine, spanning 3 countries and 20 studies, discovered that adults with extreme obesity have an increased risk of dying early due to cancer, diabetes, heart disease, stroke, and kidney and liver diseases.7

All of these studies use a person’s body mass index (BMI) to determine whether a person is obese. Obesity, defined as “BMI greater than or equal to 30,” is an index of weight-for-height: it is the person’s weight in kilograms divided by the square of his or her height in meters (kg/m2).2,7 In children and teens, BMI is interpreted after considering the child’s age and gender, because the amount of body fat changes with age and gender.7

Is BMI the Right Indicator of Obesity?

That is the question being asked lately, and a universal BMI cut-off value for obesity—the one-size-fits-all attitude—is being examined. A meta-analysis of 37 studies, which included 53,521 children between 4 and 18 years of age, found that although highly specific, BMI was not very sensitive to identifying pediatric obesity. The authors point out that although adiposity measurements are driven by race and ethnicity, organizations like the American Academy of Pediatrics still recommend using BMI to diagnose pediatric obesity in the clinic. The meta-analysis recognized that the BMI measure has a 73% sensitivity, meaning that more than 25% of the children in these studies—who may have had excess adiposity—were not identified as being obese.

Acknowledging this finding, the authors recommend that other tools that can measure body fat, such as dual-energy x-ray absorptiometry (DXA) and air-displacement plethysmography (ADP), be used as secondary measures.8 Characterizing the composition of body fat is important for diagnosing obesity, an element that is disregarded in the BMI calculation. Normal BMI does not necessarily indicate normal body fat—a major flaw with BMI as a marker of obesity.8 The study concluded that if the BMI is higher than normal, the child is most definitely obese; but if the child has a normal BMI, secondary measures like ADP or skin-fold measurements should be used to confirm the results, especially in male children. The authors also point to other studies which indicated that the increase in BMI above normal is not proportional to increases in the percentage of body fat.9

An evaluation of the correlation between BMI and fat, muscle, and bone percentages of athletes found that BMI could not be used as a direct measurement of body fat content in athletes, who tend to have a lean body mass.10 Another study that compared the National Institutes of Health’s BMI based obesity classification with the WHO’s percent body fat (%BF)-based reference standard—in women of different racial backgrounds (namely Caucasian, African American, and Hispanic)—found that the BMI cutoff by the NIH failed to identify nearly half of the women who were classified as obese based on their %BF measure. This includes another variable in the BMI equation: race. The results beg the conclusion that BMI cutoff values should factor in an individual’s racial background for a more accurate classification of obesity.11 BMI could be used as a screening tool to determine the population’s risk of obesity, but there is a need to follow up with secondary measures to confirm the absence of risk.

The CDC agrees that BMI should be used a surrogate marker for obesity, since it only measures excess weight and not excess fat. In a statement directed to clinical practitioners, CDC acknowledges that BMI is widely used primarily because it’s a simple, inexpensive, and noninvasive test, the results from which can be used at the population level by public health professionals to generate models that span populations across time and geographic regions. Other measures of body fat, such as skin fold thickness, ADP, and DXA—although better indicators of body fat and risk of obesity-related health issues—can be expensive, intrusive, and not readily accessible. There’s also a need for specialized equipment and trained staff to conduct these measures, which can prove challenging in routine clinical practice.12

What Are the Alternate Measures of Obesity?

Body adiposity index (BAI) is a measure developed by the laboratory of Richard Bergman, PhD, director of the Diabetes and Obesity Research Institute, Biomedical Sciences, at Cedar Sinai. BAI accounts for hip circumference and a person’s height, without adjusting for gender. The study, which included Mexican American and African American men and women, identified %BAI as a direct estimate of the percentage of body fat.13

The index was also validated in a Caucasian population in Newfoundland. An evaluation of 2601 individuals of both genders identified BAI as a better index than BMI, as it reflected the gender difference in total %BF between men and women, correlated better with DXA, and performed well in normal weight and overweight subjects. Surprisingly, however, it was less accurate than BMI in the obese population.14

Biomarkers of Obesity

Disease biomarkers that are reliable and can be identified using minimally invasive techniques are an ideal alternative to BMI or BAI measurements. An ideal biomarker would identify a high-risk obese individual, who might progress to develop cardiovascular disease or T2DM.

Partnering on one such project, scientists at Nestlé and General Electric studied the metabolic and lipid profiles in the plasma and urine of 40 overweight and obese females (25 to 45 years of age, with BMI between 28 and 40 kg/m2). Additionally, their body composition and visceral fat distribution were analyzed using DXA and computerized tomography. On integrating the data from these various sources, a distinct pattern of amino acids and diacyl and ether phospholipids was identified in women with high visceral fat.15 Such metabolomic profiles could vastly improve diagnosis of the disease.

What Are the Advantages of Early Detection?

Being able to identify individuals with a normal BMI as obese, using some of the secondary methods or %BAI (once validated), would open a window of opportunity for early intervention—clinical as well as lifestyle. In children, for example, parents could influence a child to maintain healthy eating habits and to exercise.

Obesity in adults can lead to health issues such as T2DM; cardiovascular problems and stroke, as well as influence breathing, resulting in sleep apnea and asthma. Additionally, obesity can lead to anxiety (see Commentary on page SP375) or more serious mental health conditions. The least we could expect is an accurate measure or test to diagnose obesity, with the hope of preventing a lot of these comorbidities, which are a huge drain on the healthcare system.References

1. Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of childhood and adult obesity in the United States, 2011-2012. JAMA. 2014;311(8):806-814.

2. Obesity and overweight. World Health Organization website. Updated August 14, 2014. Accessed September 3, 2014.

3. Treating obesity as a disease. American Heart Association website.

UCM_459557_Article.jsp. Updated April 14, 2014. Accessed September 3, 2014.

4. AMA adopts new policies on second day of voting at annual meeting [press release]. Chicago, IL: American Medical Association; June 18, 2013.


5. Adult obesity facts. CDC website. Updated July 9, 2014. Accessed September 3, 2014.

6. Kitahara CM, Flint AJ, Berrington de Gonzalez A, et al. Association between class III obesity (BMI of 40-59 kg/m2) and mortality: a pooled analysis of 20 prospective studies. PLoS Med.


7. Healthy weight — it’s not a diet, it’s a lifestyle. CDC website. Updated July 11, 2014. Accessed

September 4, 2014.

8. Javed A, Jumena M, Murad MH, et al. Diagnostic performance of body mass index to identify obesity as defined by body adiposity in children and adolescents: a systematic review and metaanalysis [published online June 25, 2014]. Pediatr Obes. doi:10.1111/ijpo.242.

9. Wells JC, Coward WA, Cole TJ, Davies PS. The contribution of fat and fat-free tissue to body mass index in contemporary children and the reference child. Int J Obes Relat Metab Disord.


10. Garrido-Chamorro RP, Sirvent-Belando JE, Gonzalez-Lorenzo M, Martin-Carratala ML, Roche E. Correlation between body mass index and body composition in elite athletes. J Sports Med Phys Fitness. 2009;49(3):278-284.

11. Rahman M, Berenson AB. Accuracy of current body mass index obesity classification for white, black and Hispanic reproductive-age women. Obstet Gynecol. 2011.115(5):982-988.

12. Body mass index: considerations for practitioners. CDC website. Accessed September 4, 2014.

13. Bergman RN, Stefanovski D, Buchanan TA, et al. A better index of body adiposity. Obesity (Silver Spring). 2011;19(5):1083-1089.

14. Sun G1, Cahill F, Gulliver W, et al. Concordance of BAI and BMI with DXA in the Newfoundland population. Obesity (Silver Spring). 2013;21(3):499-503.

15. Martin FP, Montoliu I, Collino S, et al. Topographical body fat distribution links to amino acid and lipid metabolism in healthy obese women. PLoS One. 2013;8(9):e73445.