On October 1, 2026, Eli Lilly released a new set of analyses built on its ATTAIN-1 Phase 3 trial, projecting that its oral GLP-1 medicine orforglipron, sold as Foundayo, could meaningfully lower a patient's long-term predicted risk of developing type 2 diabetes and cardiovascular disease. The numbers are striking, but the method behind them matters just as much as the headline figures.

What Lilly Actually Published

ATTAIN-1 is a completed Phase 3 trial that enrolled 3,127 adults with obesity or overweight and at least one weight-related condition, such as hypertension, dyslipidemia, obstructive sleep apnea, or existing cardiovascular disease, across ten countries. Crucially, this population did not have diabetes at the start of the trial. The new release is a post-hoc analysis of that trial's data, not a new clinical trial and not a measurement of who actually went on to develop diabetes or a cardiovascular event.

Instead, Lilly ran participants' trial outcomes, primarily weight loss and changes in metabolic markers, through two established predictive tools: the Cardiometabolic Disease Staging (CMDS) risk engine for diabetes, and a BMI-based Framingham risk engine for cardiovascular disease. Those models estimated that orforglipron could lower a person's predicted 10-year risk of developing type 2 diabetes by 45% to 57%, depending on dose, and lower predicted cardiovascular disease risk by 13% to 18%, with the largest reduction at the highest dose tested, 17.2 mg.

Why the Distinction Between Modeling and Outcomes Matters

A risk engine is a statistical tool that estimates future disease probability from measured inputs like body weight, blood pressure, and lipid levels, using relationships derived from other, older study populations. It is a legitimate and widely used way to translate short-term trial results into a longer-term risk picture. But it is fundamentally different from a dedicated outcomes trial that follows real patients for years and counts how many actually develop diabetes or have a heart attack or stroke. Modeled risk reduction is a projection, not an observed clinical event rate.

This is a distinct data set from Lilly's earlier ACHIEVE-4 trial, covered elsewhere on this site, which did measure actual major cardiovascular events in a different population: adults who already had type 2 diabetes and elevated cardiovascular risk. ATTAIN-1's new analysis instead asks what might happen, over time, to people who don't yet have diabetes, based on how their measured risk factors improved.

Regulatory Status

Orforglipron, branded Foundayo, received FDA approval on April 1, 2026, for chronic weight management, and is separately being evaluated for type 2 diabetes. This new modeling analysis does not represent a new FDA approval, label change, or indication, and the agency has not reviewed a diabetes-prevention or cardiovascular-risk-reduction claim based on this data.

What Would Confirm It

To move from a projected risk reduction to a confirmed one, a dedicated long-term outcomes trial would need to track a comparable group of people with obesity but without diabetes for several years, with incident diabetes diagnoses and cardiovascular events as the measured endpoints, not modeled ones. Lilly's statement did not announce such a trial alongside this analysis.

The Bottom Line

The ATTAIN-1 modeling results are a genuine, data-driven signal that Foundayo's effects on weight and metabolic markers are large enough to move established risk calculators substantially. That is useful information for understanding the drug's potential long-term value. It is not the same as proof that a given patient's actual risk of diabetes or a cardiovascular event will fall by those exact percentages. Anyone considering orforglipron for weight management, whether for metabolic risk reasons or otherwise, should talk to a physician about their own individual risk factors rather than applying population-level projections to their personal outlook.