Consider lower BMI cutoffs for diabetes, CVD risk screening in patients of Asian descent

Nationwide, there is significantly increased interest and focus on addressing racial and ethnic health disparities.

As Asian American and Pacific Islander (AAPI) Heritage Month comes to a close, it is critical that we raise awareness of practices that may perpetuate health inequities in the AAPI community.

It is important for us as physicians to be able to appropriately risk stratify patients to provide appropriate preventive care, treatment and care management, as well as identify risk factors that patients might have based on their race or ethnicity.

Many physicians are not necessarily aware that WHO recommends different BMI cutoffs for different groups. For example, for Asian populations in aggregate, WHO recommends a BMI cutoff of 23 kg/m² for those who are overweight (range of 22 kg/m²to 25 kg/m² depending on subpopulation) and 27.5 kg/m² for those who are obese (range of 26 kg/m² to 31 kg/m²depending on subpopulation), rather than what is usually taught in medical school and known in clinical practice in the United States, which is a BMI of 25 kg/m² for overweight and a BMI of 30 kg/m² for obese.

The reason why it is important for physicians to be aware of the different BMI cutoffs for patients of Asian descent is so we can appropriately identify patients who might need screening for additional risk factors associated with being overweight or obese: diabetes and CVD.

Patients of South Asian descent — those from India, Pakistan, Bangladesh, Nepal, Sri Lanka, Bhutan and the Maldives — have a fourtimes greater risk for atherosclerotic CVD (ASCVD) than the general population. They comprise 25% of the global population, but 60% of global heart disease patients. This shows that they have a disproportionately high rate of CVD. In addition, South Asians have very high rates of insulin resistance and diabetes as well as dyslipidemia. Those are certainly risk factors that predispose them to developing CVD, and in particular, ASCVD. Even within the different South Asian population groups, there is variable risk for developing CVD.

We need better risk calculators

CVD and diabetes are often initially without any symptoms but can be deadly. As primary care physicians, we are always trying to screen patients for these commonly asymptomatic conditions, so it is really important to know who needs a screening test. We use algorithms in our day-to-day practice, but those algorithms do not necessarily take into account the race and ethnicity of our patients.

As an example, in the U.S., we use the ASCVD risk calculator to determine the 10-year risk of heart disease or stroke in patients. The metrics that the ASCVD risk calculator require include age, diabetes status, sex, smoking status, cholesterol levels and BP. There is also a category for race. The options are “white,” “African American” or “Other.” There is a section within the calculator that explains the evidence behind how a patient’s risk is calculated and limitations of the calculated estimated risk. In that clarification, it states that the estimate that the risk calculator develops “may underestimate the 10-year risk for some racial/ethnic groups, including American Indians and some Asian Americans (eg, of South Asian ancestry).” It also goes on to say that the calculated risk “may overestimate the risk for some Asian Americans (eg, of East Asian ancestry).”

Basically, what this all means is that the main risk calculator that physicians in the U.S. use to determine a patient’s risk for ASCVD and whether a patient should be started on a statin, for example, for cholesterol control, might not accurately reflect which patients are at highest risk for developing CVD or which patients may benefit most from starting cholesterol-lowering treatment if the ASCVD risk calculator cannot adjust for non-white and non-African American patients to accurately estimate a patient’s CVD risk, which has the potential to exacerbate existing racial and ethnic health inequities.

In contrast, in the United Kingdom, physicians use a different risk calculator called the QRISK3 score. It is specific to the U.K. and looks at the risk of developing CVD over 10 years. They define CVD as a stroke, transient ischemic attack, myocardial infarction and angina. In the ethnicity section of QRISK3, the categories are “white or not stated,” “Indian,” “Pakistani,” “Bangladeshi,” “Other Asian,” “Black Caribbean,” “Black African,” “Chinese” or “Other ethnic group.” Of note, they have broken down not only different subtypes of the South Asian community, but also included some East Asian groups to reflect different risks among patients. The QRISK3 score also uses a patient’s postcode (ZIP code) because patients who live in certain sections of the country have better or worse outcomes than others. So, this score takes more variables into account to create a more individualized risk for patients.

In the U.S. it is important that we explore some of these differences in cardiovascular risk so that we are able to appropriately manage our patients. We need to develop updated risk calculators not just for CVD, but across the board, so that we have a better understanding of how to address and improve the health of all racial and ethnic communities in the U.S. This starts with the collection and analysis of detailed, disaggregated data.

In primary care, a lot of our work is data driven. For example, we follow the U.S. Preventive Services Task Force recommendations for screening guidelines to prevent different diseases, but again, all these recommendations are based on the data that are collected. If we are not collecting disaggregated data on race and ethnicity that most accurately represent all the people in our nation, then we will not know how to properly assess risk and develop guidelines that are specific to different racial and ethnic groups. When we are trying to address racial and ethnic health disparities, we must ensure access to disaggregated data, so that we may use the data to best identify and recommend interventions to improve health outcomes.

Make culturally appropriate recommendations

There are also other important considerations when assessing a person’s risk for CVD. If someone says they are vegetarian, for example, that does not necessarily mean that they follow a healthy diet. In addition, when we make certain dietary recommendations for patients, it is important that they are culturally appropriate dietary recommendations so that individuals can modify their diet without completely changing the types of foods that they eat. If we approach a patient and ask them to completely change their diet and to eat foods that they have never eaten before or that they do not frequently cook or have access to or cannot afford, then it is going to be much more difficult for that person to make changes to improve their health through diet modification. To improve outcomes, we need to have culturally specific resources in place, available in multiple languages.

In addition, when it comes to assessing obesity and risk for metabolic syndrome in the South Asian community, studies show that South Asians have a higher amount of visceral fat, which contributes to their predisposition for metabolic syndrome. It is therefore important not to just assume a patient is at low risk for diabetes, high cholesterol or CVD because they look thin or appear to have a normal BMI, but to fully assess that patient’s risk, because if conditions like diabetes or high cholesterol are not diagnosed early, then the risk for subsequent complications such as CVD increase over time. In the current COVID-19 pandemic, for example, we know that patients who are obese or have diabetes or heart disease are at increased risk for severe complications of COVID-19. So, being able to appropriately risk stratify our patients is important. Also, as we have seen throughout the COVID-19 pandemic, there are much higher rates of complications and death among persons of color. As a result, again, as we continue trying to address the health inequities that exist in our country, we must take the entire picture into account, and develop systems to appropriately risk stratify, manage and treat individuals to eliminate disparities based on race and ethnicity.

Additional considerations

When the USPSTF updated its diabetes screening guidelines in 2021, it recommended “screening people ages 35 to 70 with overweight or obesity for prediabetes and diabetes.” Since this is a U.S. guideline, the definitions of overweight and obese use the BMI cutoff of 25 kg/m² and 30 kg/m², respectively. Although international BMI cutoffs for overweight and obesity depend on an individual’s race or ethnicity, if a clinician in the U.S. decides to screen a patient of Asian descent with a BMI of 24 kg/m² for diabetes, under current policy, insurance carriers in the U.S. would not consider that patient overweight and would likely not provide the screening as a USPSTF-recommended screening that should be covered without cost sharing for the patient. This could result in undiagnosed diabetes in a patient with a U.S.-defined “normal BMI,” who is ineligible for diabetes screening, despite being considered overweight if using race/ethnicity-based BMI cutoffs, leading to a later diagnosis of diabetes and a risk of developing more severe complications of diabetes due to a delayed diagnosis. This, in turn, would perpetuate the racial/ethnic inequities in health that are present in the U.S., and highlights the importance of not only revising definitions used by clinicians to determine which patients require preventive screenings at what time point to prevent inequitable health outcomes, but also emphasizes how critical it is that we ensure that policies that determine how screenings are covered and paid for by insurance carriers align with clinical criteria/definitions that may vary based on race and ethnicity, in order to create policies that promote health equity.

References:

American College of Cardiology. South Asians and cardiovascular disease: The hidden threat. https://www.acc.org/latest-in-cardiology/articles/2019/05/07/12/42/cover-story-south-asians-and-cardiovascular-disease-the-hidden-threat. Accessed May 25, 2022.

CDC. Diabetes and Asian Americans. https://www.cdc.gov/diabetes/library/spotlights/diabetes-asian-americans.html. Accessed May 25, 2022.

ClinRisk. Welcome to the QRISK3-2018 risk calculator. https://qrisk.org/three/. Accessed May 25, 2022.

Congress.gov. H.R.3131 - South Asian Heart Health Awareness and Research Act of 2020. https://www.congress.gov/bill/116th-congress/house-bill/3131/text. Accessed May 25, 2022.

Havard T.H. Chan School of Public Health. Ethnic differences in BMI and disease risk. https://www.hsph.harvard.edu/obesity-prevention-source/ethnic-differences-in-bmi-and-disease-risk. Accessed May 25, 2022.

Hsu WC, et al. Diabetes Care. 2015;doi:10.2337/dc14-2391.

Joslin Diabetes Center. What is Body Mass Index (BMI)? https://aadi.joslin.org/en/am-i-at-risk/asian-bmi-calculator. Accessed May 25, 2022.

MDCalc. ASCVD (Atherosclerotic Cardiovascular Disease) 2013 Risk Calculator from AHA/ACC. https://www.mdcalc.com/ascvd-atherosclerotic-cardiovascular-disease-2013-risk-calculator-aha-acc#evidence. Accessed May 25, 2022.

Scroll.in. The struggle to understand – and combat – heart disease among South Asian Americans. https://scroll.in/global/983518/the-struggle-to-understand-and-combat-heart-disease-among-south-asian-americans. Accessed May 25, 2022.

USPSTF. Final recommendation statement: Screening for prediabetes and type 2 diabetes. https://uspreventiveservicestaskforce.org/uspstf/uspstf/announcements/final-recommendation-statement-screening-prediabetes-t2dm. Published Aug. 24, 2021. Accessed May 24, 2022.

Volgman AS, et al. Circulation. 2018;doi:10.1161/CIR.0000000000000580.

WHO Expert Consultation. Lancet. 2004;doi:10.1016/S0140-6736(03)15268-3.