Obesity Epidemic and BMI

Centers for Disease Control and Prevention (CDC) data shows an alarming increase in obesity among the United States (US) population, from 30.5% in 2000 to 42.4% in 2018. The combination of overweight and obesity comprises over 70% of many sex-ethnic specific populations. Obesity related diseases include heart disease, stroke, diabetes, and some cancers, which in turn are major contributors to preventable, premature death.² The COVID-19 pandemic brought in new challenges like curtailment of outdoor exercise, change of diet, surge of depressive illness, which may have accelerated the pace of the global epidemic of overweight and obesity.

Body mass index (BMI) remains a simple but unique anthropometric measurement that predicts metabolic well-being as well as cardiovascular (CV) outcomes, making it one of the most common tools used in clinical medicine. BMI measurement forms an integral component of basic vitals recorded at any clinical encounter.

One challenge is that the cutoff points of the data on BMI are mostly derived from the White population. In today's connected world, most societies comprise of a healthy mixture of people from every ethnic background. It is important to validate BMI for different ethnic groups and set appropriate cutoff levels. Moreover, BMI is a poor predictor of total body fat or distribution of body fat (visceral adiposity). In contrast, the link between waist circumference, visceral adiposity and cardiometabolic disease seems to be much stronger.

BMI and Ethnicity

In 1993 the World Health Organization (WHO) put a BMI normal cutoff level at 25 kg/m², labelling those above as overweight, while those above 30 kg/m² as obese and above 40 kg/m² as morbidly obese. These numbers were based on data obtained from mainly White populations.¹

It was soon realized that some ethnic groups such as those of South Asian descent need a lower cutoff of BMI, since they have smaller structure and a different distribution of body fat with a higher percentage of visceral fat compared to subcutaneous fat. National Institute for Health and Care Excellence (NICE) guidelines, published in 2004, reset the BMI levels of South Asian persons to 27.5 kg/m² to define obesity (instead of 30 kg/m²). Such arbitrary change was based on low volume, sparse data.¹

Ethnic-Specific Data

The analysis titled "Ethnicity-specific BMI cutoffs for obesity based on type 2 diabetes risk in England: a population-based cohort study"¹ published in Lancet, aims to bring clarity to this grey area of defining and categorizing BMI levels in ethnic communities such as South Asian, Black, Chinese, and Arab persons, looking specifically at the ability of BMI to predict future occurrence of diabetes.¹

These data are based on an electronic health record (eHR) database from primary care, hospital-based data from clinical practice research datalink (CPRD) and data from the National Health Service (NHS). Adults of 18 years of age or older with no baseline diabetes were included in the analysis. Data were collected from September 1990 to December 2018, with those with at least 1-year follow up. BMI <15 kg/m² or >50 kg/m² were excluded, as were those with mixed race or missing ethnic data. A CALIBER phenotyping algorithm was used to identify diabetes. The first recorded BMI was defined as an index BMI with a 1 year 'blanking' period to prevent reverse casualty (diagnosis of diabetes improves weight watching behavior). The risk of developing diabetes was compared with that of Whites who had a BMI of 30 kg/m² and 25 kg/m². The incidence of diabetes was predicted in the White population and then the BMI cutoff reverse calculated by a binomial regression model.

The data comprised of 1.3 million White adults of whom 97,827 people (6.6%) developed diabetes. There were 75,956 South Asian (5.2%), 49,349 Black (3.4%), 10,934 Chinese (0.7%) and 2,764 Arab (0.2%) persons included. The results show that to predict development of diabetes, the cutoff BMI of 30 kg/m² (obesity) for the White population was at par with 28.1 kg/m² for Black population, 26.9 kg/m² for the Chinese, 26.6 kg/m² for Arab and 23.9 kg/m² for South Asian adults.

A lower cutoff BMI of 25 kg/m² (overweight) in White adults, translated to a BMI of 23.4 kg/m² for Black, 22.2 kg/m² for Chinese, 22.1 kg/m² for Arab, and a much lower BMI of 19.2 kg/m² for South Asian adults is noted.

BMI, Diabetes and South Asians – More than Skin Color

The 1.9 billion South Asia population (India, Bangladesh, Pakistan, Nepal, Bhutan, Sri Lanka, Maldives) accounts for 24.89% of the total world population. There are 5.4 million South Asians in the US while United Kingdom is home to 3.2 million South Asian persons.³ So, it is important to focus on this large group and define their anthropometric criteria, like BMI, more accurately.

South Asian people have a smaller body frame, with variations of distribution of body fat (compared to White persons) with more of visceral adiposity, while a higher propensity to develop diabetes mellitus despite BMI values considered 'normal' for White persons. Studies from India⁴ have consistently focused on a lower BMI predicting diabetes in this population. Moreover, estimation of body fat by measurement of waist circumference and waist-to-hip-ratio (WHR) have shown to be a stronger predictor of cardiometabolic dysregulation and diabetes than BMI among South Asians.

The issue of BMI as a predictor of morbidity and mortality is more than skin deep. The Global Burden of Disease data published in Lancet in 2010 shows that despite being correlated with hypertension and diabetes, a high BMI was at best a weak predictor of mortality among the South Asian persons.⁵ While a BMI of more than 30 kg/m² (compared to BMI < 25 kg/m²) had a hazard ratio (HR) of mortality at 1.08 in the South Asian persons, the HR in White persons of similar BMI levels is substantially higher at 1.99.⁴

A larger BMI thus translates to more diabetes and hypertension but surprisingly, no increase in mortality in this population. Many explanations, including a larger pool of fat accounting for resilience to infections (a common morbidity among the South Asian population) is offered, but there is no concrete evidence.

Ethnicity to Personalized Medicine

Estimation of body fat in predicting cardiometabolic derangement has always been challenging. Despite availability of various simple methods ranging from measurement of waist circumference or WHR to more sophisticated bioelectric impedance, dual-energy X-ray absorptiometry (DXA) and magnetic resonance imaging (MRI), BMI remains as one of the most common clinical models in predicting emergence of future diabetes. Sharpening and refining this common clinical tool by identifying different cutoff values of BMI for different ethnic groups would make it more effective to predict diabetes in a much wider population, with improved accuracy. It is one step forward to attain the 'dream-goal' of universal personalized medicine, using traditional clinical tools.

References

1. Caleyachetty R, Barber TM, Mohammed NI, et al. Ethnicity-specific BMI cutoffs for obesity based on type 2 diabetes risk in England: a population-based cohort study. Lancet Diabetes Endocrinol 2021;9:419-26.
2. Overweight & Obesity (CDC website). 2021. Available at: https://www.cdc.gov/obesity/index.html. Accessed 07/02/2021.
3. Demographic Information (South Asian Americans Leading Together website). 2019. Available at: https://saalt.org/south-asians-in-the-us/demographic-information/. Accessed 07/02/2021.  
4. Snehalatha C, Viswanathan V, Ramachandran A. Cutoff values for normal anthropometric variables in Asian Indian adults. Diabetes Care 2003;26:1380-84.
5. Gajalakshmi V, Lacey B, Kanimozhi V, Sherliker P, Peto R, Lewington S. Body-mass index, blood pressure, and cause-specific mortality in India: a prospective cohort study of 500 810 adults. Lancet Glob Health 2018;6:e787–e794.

Clinical Topics: Cardiovascular Care Team, COVID-19 Hub, Noninvasive Imaging, Prevention, Magnetic Resonance Imaging, Diet, Hypertension, Diabetes and Cardiometabolic Disease

Keywords: State Medicine, Body Mass Index, Waist Circumference, United Kingdom, Waist-Hip Ratio, Adiposity, Intra-Abdominal Fat, Skin Pigmentation, African Americans, Electronic Health Records, Precision Medicine, Absorptiometry, Photon, Arabs, Mortality, Premature, Diabetes Mellitus, Type 2, COVID-19, Electric Impedance, Follow-Up Studies, Global Burden of Disease, Obesity, Morbid, Diet, Asian Continental Ancestry Group, Magnetic Resonance Imaging, Hypertension, Subcutaneous Fat, Neoplasms, Stroke, World Health Organization, Epidemics, Centers for Disease Control and Prevention, U.S., Primary Health Care, Cardiovascular Diseases, Heart Diseases, Hospitals, Primary Prevention, Race Factors

< Back to Listings