EASD opens in Milan on Monday, September 28. On October 1, Lilly presents ACHIEVE-4 — the first cardiovascular outcomes data for Foundayo (orforglipron).

The press release already quotes a hazard ratio of 0.84. The framing in most coverage will be "16% lower cardiac risk."

Before that number runs everywhere, here's what it actually means — and why the full picture is different.

What a Hazard Ratio Is

A hazard ratio compares the rate of an event between two groups over time.

HR = 0.84 means: in the trial, the Foundayo group had events occurring at 84% of the rate of the comparison group. That's the "16% lower" framing.

But a hazard ratio is a point estimate. It's the center of a range. The range is called the confidence interval.

What the Confidence Interval Actually Says

ACHIEVE-4 MACE-4: HR 0.84, 95% CI 0.59 to 1.20

The 95% confidence interval means: if this trial were run 100 times, approximately 95 of those runs would produce an interval containing the true effect.

The interval here runs from 0.59 to 1.20.

0.59 would be a 41% reduction in cardiovascular events. That's a substantial benefit.

1.20 would be a 20% increase in cardiovascular events.

The interval crosses 1.0. When a confidence interval crosses 1.0, the result is statistically consistent with no difference between groups. It is also consistent with harm.

"16% lower cardiac risk" describes the point estimate. The confidence interval says we cannot rule out zero effect or a slight increase.

Why the Interval Is Wide

Wide confidence intervals indicate few events. In cardiovascular outcome trials, you need a large number of major cardiac events to generate a tight estimate.

ACHIEVE-4 enrolled adults with type 2 diabetes, overweight or obesity, and elevated cardiovascular risk. The trial was designed to demonstrate non-inferiority against insulin glargine — not superiority. Non-inferiority means "not worse than the comparator," which requires fewer events to demonstrate than superiority.

The trial succeeded at its design objective. It does not demonstrate a cardiovascular benefit in the way SELECT did for semaglutide.

The same release reports one interval that does not cross 1: all-cause mortality, HR 0.43, 95% CI 0.25 to 0.75, nominal p=0.002. It is a secondary endpoint with few events and a nominal p-value, so it does not carry the weight of a primary result — but it is the stronger number in the release, and it is the one almost nobody is quoting.

The SELECT Comparison

Semaglutide's cardiovascular outcomes trial (SELECT) showed:

HR 0.80, 95% CI 0.72 to 0.89

That interval does not cross 1.0. Both ends of the range show benefit. This is a demonstrated cardiovascular risk reduction.

ACHIEVE-4's interval (0.59–1.20) is three times as wide as SELECT's (0.72–0.89). That width reflects fewer events and a non-inferiority rather than superiority design.

These are not equivalent data even if the point estimates look similar.

What to Watch October 1

Full ACHIEVE-4 data presents at EASD October 1. What the press release doesn't include:

How many total MACE events occurred in each arm
Subgroup analyses (particularly people without established cardiovascular disease)
Secondary endpoints including body weight, blood glucose, and renal outcomes
Duration of follow-up

The cardiovascular headline will run before most of this is available. The confidence interval won't be in the headline.

The Broader Principle

When a trial reports a hazard ratio, always ask for the confidence interval before interpreting the finding.

For cardiovascular outcomes specifically:

Trial	Drug	HR	95% CI	Crosses 1?
SELECT	Semaglutide	0.80	0.72–0.89	No — demonstrated benefit
SURPASS-CVOT	Tirzepatide	0.92	0.83–1.01 (95.3% CI)	Yes — non-inferiority design
ACHIEVE-4 (topline)	Foundayo	0.84	0.59–1.20	Yes — non-inferiority design

The framing "GLP-1 drugs reduce cardiovascular risk" is well-supported by SELECT. It is not equally supported by every trial in the class.

Trial populations also matter: SURPASS-CVOT and ACHIEVE-4 enrolled people with type 2 diabetes and cardiovascular risk. Neither trial says anything about cardiovascular outcomes in people without those conditions.

EASD Coverage This Week

PeptidesGPT is covering EASD data in real time as it drops:

Sep 30: TRIUMPH-2 — retatrutide Phase 3, 80-week data
Oct 1: ACHIEVE-4 — Foundayo cardiovascular (MACE-4 HR 0.84, CI 0.59–1.20)

All coverage will include confidence intervals.

Source: Lilly press release, September 15, 2026. ACHIEVE-4, MACE-4 HR 0.84 (95% CI 0.59–1.20); MACE-3 HR 0.77 (95% CI 0.52–1.13). SURPASS-CVOT MACE-3 HR 0.92 (95.3% CI 0.83–1.01). SELECT HR 0.80 (95% CI 0.72–0.89).

PeptidesGPT is an educational platform. This content is for informational purposes only and does not constitute medical advice.

Before EASD: How to Read a Confidence Interval (And Why It Matters This Week)