GLP-1 Clinical Trial Eligibility: Who Gets Left Out and Why

GLP-1 clinical trial eligibility criteria decide who could join a study, and therefore who the results describe most directly. If you have diabetes, a history of pancreatitis, a recent psychiatric illness, are under 18 or are planning a pregnancy, there is a fair chance the big headline trial you just read about did not include anyone quite like you.

That is not a scandal. It is how trials are built. But it is worth knowing, because "the study showed 20% weight loss" quietly means "in the people who were allowed in."

Why do GLP-1 trials exclude people at all?

Think of a trial as a carefully designed question. Researchers narrow the group so that:

• the answer is clean. If half the group had diabetes and half did not, a weight result would blur two different populations together.
• known risks are kept low. People with conditions the drug might worsen are often kept out until there is more safety data.
• the drug, not something else, explains the change. People whose weight was already swinging, or who take medicines that change weight, add noise.

Each of those is reasonable. Each also shrinks the slice of real life the results speak for.

What did real GLP-1 trials actually exclude?

You do not have to guess. Every trial registered on ClinicalTrials.gov lists its criteria. Here is what two landmark obesity trials posted.

SURMOUNT-1 (tirzepatide, the trial behind Zepbound's weight approval) required a BMI of 30 or more, or 27 or more with a condition such as high blood pressure, abnormal cholesterol, sleep apnea or heart disease, plus at least one past unsuccessful diet attempt. It excluded people with:

• diabetes
• a weight change of more than 5 kg in the three months before the study
• obesity caused by another hormonal disorder or a genetic syndrome
• a history of pancreatitis
• a personal or family history of medullary thyroid carcinoma or MEN 2
• significant active or unstable major depression, or another severe psychiatric disorder, in the last two years
• any lifetime suicide attempt

STEP 1 (semaglutide 2.4 mg, the Wegovy weight trial) excluded anyone with an HbA1c of 6.5% or higher at screening, and anyone who reported losing or gaining more than 5 kg (11 lb) in the 90 days before screening.

Notice something? The people who most often ask "will this work for me?" (people with diabetes, people already losing weight, people with complicated mental health histories) are exactly the ones these two trials left out.

Who was missing from GLP-1 trials, and where does the evidence come from instead?

Group	Example of how trials handled it	Where other evidence comes from
People with type 2 diabetes	Excluded from STEP 1 and SURMOUNT-1	Separate trials. STEP 2 required type 2 diabetes with an HbA1c of 7 to 10%
Children and teens	Obesity trials above enrolled adults (18+)	Separate pediatric trials. Wegovy injection's label now covers ages 12+; Zepbound's label is adults only
Pregnant people	Excluded; contraception typically required	Pregnancy exposure registries. Zepbound's label lists one (1-844-524-0039)
Recent serious depression or past suicide attempt	Excluded from SURMOUNT-1	Large pooled analyses and health-record studies (see below)
Past pancreatitis, thyroid cancer risk	Excluded	Post-marketing reports and long-term safety follow-up
People with heart disease but no diabetes	Not the focus of weight trials	SELECT enrolled 17,604 adults aged 45+ with established heart disease and no diabetes

That last row is the hopeful part. Gaps get noticed, and then someone designs a trial for them.

Does "who was in the trial" change how the results apply to you?

Sometimes a lot. STEP 1 (no diabetes) and STEP 2 (type 2 diabetes) both tested semaglutide 2.4 mg for 68 weeks. Average weight change was -14.9% in STEP 1 and -9.6% in STEP 2. Same drug, same dose, different people, noticeably different number. If you have diabetes and only ever saw the STEP 1 headline, your expectations were calibrated to a group you were not in. (More on that in why two studies of the same GLP-1 can disagree, and in GLP-1 weight loss with type 2 diabetes.)

Demographics matter too. The two-year STEP 5 trial enrolled 304 people, 77.6% of them female and 93.1% white. The SURMOUNT-4 maintenance trial was 71% women. These trials were not designed to compare men and women, or different ethnic groups, so they cannot tell you much about whether those groups respond differently. Splitting a small group into smaller groups after the fact is risky; see GLP-1 trial subgroups.

A real example: mood, exclusions and the 2026 label change

Here is where eligibility gets interesting. Trials like SURMOUNT-1 excluded people with recent severe psychiatric illness or any past suicide attempt. So when post-marketing reports about suicidal thoughts appeared in 2023, trials alone could not fully answer the question for people with those histories.

In January 2026, FDA asked makers of Wegovy, Zepbound and Saxenda to remove the suicidal behavior and ideation warning from their labels. FDA's review pooled 91 placebo-controlled trials (107,910 participants) and also used a health-record study of more than 2.2 million people with type 2 diabetes, and found no increased risk. The current Wegovy and Zepbound labels list that section as "Removed" in 02/2026.

Two things are true at once: that is strong, reassuring evidence, and FDA still tells clinicians to refer anyone who discloses suicidal thoughts for mental health evaluation. If that history is part of your story, it is worth a direct conversation with your prescriber rather than assuming any trial was about you. Our mental health evidence guide goes deeper. If you are in crisis, call or text 988.

How to check a trial's eligibility yourself in five minutes

1. Find the trial's NCT number in the paper or news story (it looks like NCT04184622).
2. Open that record on ClinicalTrials.gov and scroll to Participation Criteria (inclusion and exclusion criteria).
3. Note the age range, BMI cut-offs, diabetes status and the conditions excluded.
4. In the paper, find "Table 1" (baseline characteristics): average age, sex, race and starting weight.
5. Ask: how many of these exclusions apply to me?

If you track your medicines, conditions and symptoms somewhere like GLPily, bring that list to your appointment. It makes the "was I in the trial?" conversation faster and more specific.

What to ask your prescriber if you were not in the studied group

• "The main trial excluded people with my condition. What evidence do we have for someone like me?"
• "Is there a different trial, or a section of the label, that covers my situation?"
• "What should I watch for, and when should I call?"
• "Does my history change how we'd judge whether it's working?"

Common questions

Why did GLP-1 weight-loss trials exclude people with diabetes?

Diabetes changes how much weight people tend to lose and adds glucose-lowering medicines into the mix, so it was studied separately (STEP 2, SURMOUNT-2). Neither design is "better"; they answer different questions.

If I was excluded from the trials, does that mean the drug is unsafe for me?

No. It means that particular trial does not directly answer the question for you. Your prescriber combines other trials, the label's specific-population sections and your history.

Where can I see who was in a trial?

On the ClinicalTrials.gov record (criteria) and in the published paper's baseline-characteristics table. FDA also publishes Drug Trials Snapshots for many new drugs, summarizing who took part in the trials behind the original approval.

Are pregnant people ever studied?

Usually not in drug trials. Information comes from animal studies, registries and case reports, which is why labels like Zepbound's point to a pregnancy exposure registry.