Modern medicine loves numbers. LDL. A1c. BMI. Tumor diameter. Pounds lost. They are easy to measure, easy to track, and easy to celebrate. But none of them are the disease itself.
As GLP-1 prescribing accelerates and insurers increasingly define success by weight reduction, we are repeating a familiar mistake: Elevating a surrogate marker to the status of a biological endpoint.
BMI was never designed to diagnose disease in individuals. Adolphe Quetelet created it in the 19th century as a population statistic, not a clinical tool. Yet today it shapes treatment pathways, insurance eligibility, and public health messaging.
The danger of surrogate markers
Weight correlates with risk at the population level, but it does not measure inflammatory activity, endothelial dysfunction, thrombotic instability, immune dysregulation, or tissue-level metabolic signaling, the processes that actually drive myocardial infarction, stroke, cancer progression, and severe infection.
Cardiology has already taught us this lesson.
The Cardiac Arrhythmia Suppression Trial (CAST) showed that suppressing premature ventricular contractions improved the surrogate marker but increased mortality. The ACCORD trial demonstrated that aggressively lowering A1c did not uniformly reduce cardiovascular mortality and, in some cases, increased it. Hormone replacement therapy improved lipid profiles yet increased thrombotic events in the Women’s Health Initiative.
More recently, CANTOS proved that targeting inflammation reduced recurrent cardiovascular events independent of lipid levels. Biology, not the surrogate, was causal.
The true mechanisms of GLP-1 agonists
GLP-1 receptor agonists produce weight loss and improve glycemic control. Some cardiovascular outcome trials show benefit in high-risk patients. But the mechanism is unlikely to be weight loss alone. These agents have pleiotropic metabolic and anti-inflammatory effects, including modulation of endothelial function and inflammatory signaling.
If pounds lost were the primary causal mechanism, we would see uniform cardiovascular benefit across all weight-loss interventions. We do not.
A patient can lose 15 percent of body weight and still have active vascular inflammation. Another may show improved inflammatory stability with minimal weight change. Without measuring biological activity directly, we risk attributing benefit to the wrong variable.
Precision endpoints in medicine
Public health messaging often equates obesity with disease causation. But social determinants, housing, sanitation, environmental exposure, and chronic stress, have historically exerted profound effects on inflammatory burden and mortality. Structural improvements reduced infectious and inflammatory mortality long before macronutrient debates dominated medical discourse.
This is not an argument against treating obesity. Excess adiposity can contribute to metabolic and inflammatory stress. But weight reduction should be viewed as a potential modifier, not the endpoint itself.
Precision medicine requires precision endpoints. We should prioritize measurement of inflammatory activity, endothelial function, metabolic signaling, and thrombotic risk, the processes that directly determine morbidity and mortality. Anthropometric change alone cannot serve as a proxy for biological suppression of disease.
Weight belongs in the chart. It does not belong as a stand-alone clinical endpoint.
Richard M. Fleming is a physician specializing in cardiovascular and inflammatory disease.
