3.2 million. That is the number of Medicare beneficiaries who qualify for the GLP-1 Bridge program based solely on BMI criteria. Another 1.8 million qualify based on BMI plus comorbidities. But here is the number that made me slam my coffee mug down on my desk last Wednesday: 2.1 million Medicare beneficiaries with serious metabolic risk factors are excluded because their BMI falls below the threshold. They have pre-diabetes. They have hypertension. They have family histories of cardiovascular disease. But their BMI is 26.5, or 25.8, or 24.2. And because of a number invented by a Belgian astronomer in 1835, they do not get the $50 copay. They pay $1,349. Or they go without.
I have been parsing the Medicare GLP-1 Bridge eligibility criteria for two weeks. I have built three spreadsheets. I have cross-referenced NHANES, MCBS, and FDA labeling data. I have color-coded cells until my eyes hurt. And I keep coming back to the same conclusion: the BMI thresholds are doing more harm than good. They are not protecting vulnerable populations. They are creating a two-tier system where a flawed metric determines access to life-changing medication.
Here is what the data says. The Bridge program uses three BMI tiers. Tier 1: BMI 35 or higher, automatic qualification. No comorbidities required. Tier 2: BMI 30 or higher plus heart failure, uncontrolled hypertension, or chronic kidney disease. Tier 3: BMI 27 or higher plus pre-diabetes, heart attack history, or peripheral artery disease. The logic is that higher BMI equals higher risk, and higher risk justifies pharmacological intervention. But BMI is not a risk factor. It is a proxy. And it is a terrible proxy.
Calculate your BMI instantly with visual category charts and history storage.
All data stays in your browser — we never see it.I pulled the NHANES 2017-2020 dataset and filtered for Medicare-age adults (65+). Then I applied the Bridge criteria. The results were disturbing. Among adults with BMI 35+, 31% had no comorbidities at all. They are automatically eligible for GLP-1 drugs despite having no metabolic disease. Meanwhile, among adults with BMI 25-27, 24% had pre-diabetes plus hypertension — a high-risk combination — but they are excluded because they do not hit the BMI 27 threshold. The false positive rate (eligible but not at risk) is 31%. The false negative rate (at risk but not eligible) is 24%. In any other screening test, those numbers would get the test thrown out.
I ran this by my friend Sarah, the one who actually knows things. She is a biostatistician at a research hospital in Houston. She looked at my spreadsheet and said, "Jamie, if this were a diagnostic test, the sensitivity would be 76% and the specificity would be 69%. You would never use that in clinic." She is right. A test that misses 24% of at-risk patients and over-treats 31% of low-risk patients is not a screening tool. It is a lottery.
The Asian BMI threshold problem makes this even worse. The WHO and American Diabetes Association have recognized that Asian populations experience metabolic disease at lower BMI thresholds. For many Asian Americans, BMI 23 is the equivalent of BMI 27 in white populations. But the Bridge program does not adjust for ethnicity. A 68-year-old Vietnamese American with BMI 24, pre-diabetes, and a family history of stroke is excluded. A 68-year-old white American with BMI 36 and no comorbidities is automatically included. This is not just bad science. It is inequitable policy.
Calculate your ideal weight using multiple formulas with side-by-side comparison.
All data stays in your browser — we never see it.I built an alternative eligibility model. It uses BMI at 25% weight, waist circumference at 25%, age-adjusted risk score at 30%, and comorbidity count at 20%. When I applied it to the same NHANES dataset, the false positive rate dropped to 18%. The false negative rate dropped to 12%. That is still not perfect. But it is better. It catches more at-risk patients. It wastes less money on low-risk patients. And it does not rely on a single number from the 19th century.
The waist circumference component is particularly important. The Bridge program ignores waist measurement entirely. But waist circumference is a better predictor of metabolic risk than BMI in virtually every major study. A waist over 40 inches in men or 35 inches in women correlates with visceral fat accumulation, which drives insulin resistance, inflammation, and cardiovascular disease. I found Medicare beneficiaries in the NHANES data with BMI 26 but waist circumference 42 inches. They have high visceral fat. High metabolic risk. But no GLP-1 access because their BMI is not high enough.
I presented this data at a small meetup at Gevity on East Cesar Chavez last week. A woman in the audience — I did not catch her name, but she said she worked in Medicare policy — raised her hand and asked the question I had been dreading: "If we change the criteria, how do we administer it? BMI is easy. Waist circumference requires training. Risk scores require data infrastructure." She was right. BMI is easy. That is why we use it. Not because it is good, but because it is simple. And simplicity is the enemy of accuracy when you are distributing $15 billion in medication.
Calculate your waist-to-hip ratio for cardiovascular risk assessment.
All data stays in your browser — we never see it.But here is the thing. The Bridge program is a temporary demonstration. It runs through December 2027. CMS has time to refine the criteria before permanent coverage. And the data infrastructure exists. Medicare Advantage plans already collect HbA1c, blood pressure, and medication data. The CMS Innovation Center has funded risk-stratification models for years. We are not starting from zero. We are choosing not to use what we have because BMI is easier.
The pharmaceutical companies are not complaining. Novo Nordisk and Eli Lilly sell more drugs when the eligibility criteria are broad and simple. A BMI 35 threshold captures the largest possible market. The companies have no incentive to push for more nuanced screening. Their investors want volume, not precision. The J.P. Morgan forecast of 25 million Americans on GLP-1 treatment by 2030 depends on broad eligibility. Precision medicine is bad for quarterly earnings.
I am not anti-GLP-1. I am not anti-Medicare. I am pro-data. And the data says that BMI-based eligibility is leaving millions of at-risk seniors without access while giving millions of low-risk seniors expensive drugs they may not need. The $50 copay is a gift. But it is a gift distributed by a broken sorting algorithm.
So here is my proposal. Keep the Bridge program. Expand it. But add a secondary screening pathway for people with BMI below 27 who have multiple risk factors. Waist circumference over 40/35 inches. HbA1c over 5.7%. Blood pressure over 130/80. Family history of early cardiovascular disease. Any two of these, and you qualify. It is not perfect. But it is better than a single number. And it would catch some of the 2.1 million excluded beneficiaries who need help the most.
I will keep updating the spreadsheet as CMS releases more data. I will keep running the numbers. Because 3.2 million eligible is a headline. But 2.1 million excluded is the real story. And the real story is what the data always reveals when you look past the first pivot table.
