Diabetes Screening Thresholds Should Be Lowered for Racial Minorities

The body mass index (BMI) threshold for diabetes screening should be lower for certain racial and ethnic groups, researchers suggested.

In a cross-sectional study, the predicted prevalence of diabetes among 35-year-olds with a BMI of at least 25 — the cutoff for overweight — was much higher for Black, Hispanic, and Asian adults in the U.S. compared with white adults:

• Asian: 3.8% (95% CI 2.8-5.1)
• Black: 3.5% (95% CI 2.7-4.7)
• Hispanic: 3.0% (95% CI 2.1-4.2)
• White: 1.4% (95% CI 1.0-2.0)

These predictions were based on national survey data on the prevalence of diabetes among adults ages 35 to 70 — the group eligible for diabetes screening, noted Dhruv S. Kazi, MD, MSc, MS, of the Smith Center for Outcomes Research in Boston, and colleagues in the Annals of Internal Medicine.

Current 2021 guidelines from the U.S. Preventive Services Task Force (USPSTF) recommend that adults ages 35 to 70 be screened for diabetes if they have a BMI of 25 or higher.

Among people in this age group, 12.4% of white Americans had diabetes compared with 17.3% of Asian Americans, 20.7% of Black Americans, 20.6% of Mexican Americans, and 16.4% of other Hispanic Americans.

Racial and ethnic minority groups also had a much higher rate of undiagnosed diabetes: 27.6% of Asian Americans, 22.8% of Black Americans, 21.2% of Mexican Americans, and 23.5% of other Hispanic Americans compared with 12.5% of white Americans.

Assuming that a BMI of 25 or higher among white adults “reflects the implicit societal standard for diabetes screening,” the equivalent BMI thresholds for diabetes screening would need to be lowered for other racial and ethnic groups, Kazi’s group suggested. Specifically, the equivalent BMI thresholds would need to be 20 for Asian adults, <18.5 for Black adults, and 18.5 for Hispanic adults.

“The simplicity of a single screening threshold for all Americans is alluring, but it is deeply inequitable,” Kazi explained in a statement.

“If the current thresholds are universally applied, without accounting for differential risk in racial/ethnic groups, clinicians may underdiagnose diabetes in Asian, Hispanic, and Black Americans,” he added. “On the other hand, applying a more tailored approach may allow reduced rates of undiagnosed diabetes and produce population-wide improvements in diabetes care.”

In order to achieve more equal diabetes screening, age would also have to be lowered as well, the authors noted. Assuming an age of 35 represents the standard for white adults, they suggested that the equivalent age threshold for screening would need to be 23 for Asian adults, 21 for Black adults, and 25 for Hispanic adults.

“Delayed diagnosis and inadequate treatment of diabetes can produce catastrophic consequences, jeopardizing one’s heart, kidney, eyes, and limbs,” said co-author Rahul Aggarwal, MD, of Beth Israel Deaconess Medical Center in Boston, in a statement. “But it doesn’t affect all of us equally — there are striking disparities that are largely the legacy of structural racism.”

“Fixing the health disparities for Americans with diabetes will require a range of strategic investments in health care and efforts to reduce structural inequities. Making screening more equitable is a place to start, as it ensures that individuals with diabetes can receive preventive care and treatment in a timely manner and avert the most catastrophic consequences of diabetes,” he added.

The researchers noted that these findings built on prior USPSTF recommendations that suggested lowering the BMI threshold for screening in Asian, Black, and Hispanic adults.

In an accompanying editorial, Quyen Ngo-Metzger, MD, MPH, of Kaiser Permanente in Pasadena, California, agreed that the USPSTF should indeed change its diabetes screening recommendations “to be consistent with its own stated aspirational goals.”

“The USPSTF already recommends risk-based preventive services for other conditions, including cancer, cardiovascular disease, and infectious disease,” Ngo-Metzger pointed out. “To address the current inequity in diabetes screening, the USPSTF should apply the same consideration to its diabetes screening recommendation.”

For this study, Kazi’s group used National Health and Nutrition Examination Survey data from 2011 to 2018, including responses from 19,335 nonpregnant U.S. adults ages 18 to 70, comprising 6,319 white adults, 2,658 Asian adults, 4,597 Black adults, 4,998 Hispanic adults, and 763 adults from other racial/ethnic populations.

Diabetes was defined as an HbA1c level of at least 6.5% or if the individual responded “yes” when asked if a healthcare professional ever formally diagnosed them with diabetes. Undiagnosed diabetes was defined as criteria being met for the condition but with a response of “no” or “don’t know” regarding a professional diagnosis.