I am a South Asian physician — Pakistani, to be specific. By the American Diabetes Association’s (ADA) own rule, my doctor should begin screening me for diabetes once my body mass index (BMI) passes 23, because Asian Americans develop the disease at lower weights. And they did just that. Yet, when I asked about a GLP-1 receptor agonist, the answer was no. Eligibility for FDA-approved GLP-1s starts at a BMI of 30, or 27 with at least one weight-related condition. At my BMI of 26.5 but without diabetes, I meet neither threshold.
The ADA guideline that flags me for annual diabetes screening and the prior authorization form that denies me the drug are describing the same 26.5. Only one of them was written with an Asian body in mind.
Healthcare coverage for GLP-1s is shrinking. Medi-Cal dropped coverage for GLP-1s for weight-loss in January; MassHealth followed in July. When a benefit contracts, the patients whose eligibility was always marginal lose first. Meanwhile, in June, the ADA released Asian-specific waist and BMI thresholds for diagnosing obesity. The clinical definition moved in one direction while access has moved in the other.
A Closer Look at the Screening Guidelines
At the same BMI, people of Asian descent carry more visceral fat and less muscle than people of European descent, the body composition that drives insulin resistance. That fat is invisible to the scale, so BMI may read lower: the 2025 iCARDIO global obesity guidelines put an Asian person’s BMI about two to three units under a white person’s at the same level of body fat. This can manifest as chronic diseases.
Type 2 diabetes and cardiovascular disease reach Asian adults at lower BMIs and, across every subgroup the American Heart Association has examined, at higher rates than white adults. The World Health Organization (WHO) saw this in 2004 and set lower action points for Asian populations, at BMIs of 23 and 27.5.
How large is the mismatch? In a study of nearly 1.5 million adults in England, researchers asked at what BMI each ethnic group reaches the diabetes risk that white adults reach at 30 BMI. For South Asians, the answer was 23.9. For Chinese adults, 26.9. A South Asian at 24 BMI carries the risk that earns a white patient the diagnosis of obesity, yet treatments that come with it are denied until 30.
Why Hasn’t Drug Eligibility Caught Up?
The pivotal trials were designed around Western definitions of obesity, and the labels copied the trial entry criteria. A 2026 analysis of 352 GLP-1 trials found that of every inclusion criterion these studies used, BMI was the one that shut Asian patients out. A clinical trial inclusion cutoff of 30 resulted in 41.5% of non-Asian American adults being eligible, compared to 13.5% of Asian Americans and 7.2% of Koreans. Blood sugar, kidney function, and liver criteria admitted more than 93% of every group. Trials have under-enrolled non-white patients for 30 years, and each under-representative trial has become the reason the next label stays where it is.
More recent research invalidates that excuse. Published in 2025, STEP 11 tested semaglutide 2.4 mg in South Korea and Thailand using the local definition of obesity: a BMI of 25 or higher. Participants lost 16% of body weight compared to 3.1% on placebo. A post hoc analysis looked at 68 participants who started below a BMI of 30. Within that group, the 45 on semaglutide lost 17.6%, compared with 4.4% for the 23 on placebo. Under the American threshold, none of these 68 would qualify on BMI alone: most of them would need a second diagnosis before an insurer would cover the drug. In their conclusion, the authors highlighted the importance of population-specific approaches to inform reimbursement policies and public health strategy.
Meanwhile, across eight cardiovascular outcome trials, GLP-1 receptor agonists prevented a major cardiac event in 2.9 of every 100 Asian participants and 1.4 of every 100 white participants. The patients with the most to gain are the ones the threshold has kept waiting for access.
The Labels Already Changed. Who Noticed?
Current labels — including the label for semaglutide (Wegovy) as revised in 2026, and the one for tirzepatide (Zepbound) — now state the indication without a number: adults with obesity, or adults with overweight plus a weight-related condition. Yet, the 30 and the 27 survive in the trial descriptions and in payer prior authorization forms.
So, the question has shifted from arithmetic to definition. Which definition of obesity applies?
The ADA’s 2026 Standards of Care tell clinicians to confirm excess adiposity with cutoffs appropriate to ethnicity, and name South Asian patients specifically. Britain’s National Institute for Health and Care Excellence sets a higher starting line than any American insurer: a BMI of 35 with a comorbidity for tirzepatide; but they still tell prescribers to lower the threshold by 2.5 units for South Asian, Chinese, other Asian, Middle Eastern, and Black patients. A system that rations more tightly than ours built in the adjustment. Ours did not.
When Is Race-Based Medicine Appropriate?
Medicine spent the last 5 years pulling race out of certain clinical calculators, and it was right to do so. The race coefficient in the old eGFR equation multiplied a Black patient’s estimated kidney function by roughly 1.2 for no measured reason, which meant Black patients reached the thresholds for nephrology referral and transplant listing later than everyone else. The National Kidney Foundation and American Society of Nephrology retired it in 2021, and after transplant programs recalculated, more than 6,100 Black candidates had their waiting times backdated by a median of 1.7 years.
An ethnicity-adjusted BMI threshold runs the other way. This adjustment reflects a difference measured directly, in visceral fat and lean mass at a given BMI, since the WHO’s 2004 review. Factoring race into BMI would widen access to care instead of narrowing it.
What Should Payers Do Now?
Insurers and pharmacy benefit managers should adopt the ethnicity-adjusted BMI bands their own clinical societies already publish, or accept an ethnicity-appropriate waist circumference as an alternate qualifier. Both documents exist. Both were written by the organizations these payers cite whenever it suits them.
The screening threshold was set low to find metabolic disease early in Asian patients. Yet, payers set the treatment threshold high. That decision lives in prior authorization criteria, and it contradicts the guidelines those same payers rely on. The fix needs no new law and no new trial. It needs someone at the payer level to read what is already on their desk.
