STEP Trial Results Decoded: What 15% Weight Loss Means

The number that put semaglutide in every health headline: 14.9% mean body weight loss at 68 weeks. That figure came from STEP-1, the flagship trial in Novo Nordisk's Semaglutide Treatment Effect in People with Obesity (STEP) program, and it changed how obesity medicine thinks about pharmacological treatment.

But "14.9% mean weight loss" is a summary statistic, and summary statistics conceal things. The STEP trial results tell a more complicated and more useful story when you look past the headline — particularly around what "mean" means, who drives the number, and what happens after the trial ends.

The STEP Program: What Trials Were Run

The STEP program was a series of Phase 3 trials designed to evaluate semaglutide 2.4 mg weekly for weight management across different populations. The main trials:

STEP-1 (Wilding et al., 2021, NEJM) is the trial most people cite. It enrolled 1,961 adults with BMI ≥30, or BMI ≥27 with a weight-related comorbidity, who did not have type 2 diabetes. All participants received lifestyle counseling. Two-thirds received semaglutide 2.4 mg; one-third received placebo.

Mean vs. Median: Why One Number Isn't Enough

The 14.9% figure is a mean — the average across all semaglutide-arm participants. Weight loss distributions in GLP-1 trials are not symmetric, which is why looking only at the mean can mislead.

In STEP-1:

• About 86% of semaglutide participants lost at least 5% of body weight
• About 69% lost at least 10%
• About 50% lost at least 15%
• About 32% lost at least 20%

That 50% losing 15%+ tells you the mean is not far from the median, which is actually encouraging — the average isn't being pulled up by a small number of extreme responders. But the responder/non-responder split is real: roughly 14% of semaglutide participants lost less than 5% of body weight.

Who falls in the non-responder tail? The data don't give a clean answer, but candidates include people with insulin resistance severe enough to blunt GLP-1 response, people who struggled with adherence or side effects that pushed them to discontinue, and people with atypical gut pharmacology. The trial design can't fully separate "didn't respond biochemically" from "didn't take the drug as prescribed."

The Diabetes Penalty: STEP-2 Results

STEP-2, which enrolled people with type 2 diabetes, found 9.6% weight loss at 68 weeks vs. 3.4% for placebo. That's a meaningful treatment effect — but it's substantially less than STEP-1.

Why? The biology isn't completely settled, but leading explanations:

• Insulin resistance dampens the weight-loss pathway even when GLP-1 signaling is activated
• Baseline medications: many T2D participants were on insulin or sulfonylureas, which promote weight gain and counteract the GLP-1 effect
• Different starting physiology: people with established T2D have different metabolic responses to appetite reduction

For patients with diabetes considering semaglutide for weight management, STEP-2 is the relevant benchmark, not STEP-1.

The Behavioral Therapy Modifier: STEP-3

STEP-3 added intensive behavioral therapy (IBT) on top of semaglutide and found 16.0% weight loss — modestly better than STEP-1's 14.9%. The combined intervention was superior to IBT alone (5.7%) and superior to semaglutide without structured behavioral support.

The practical implication: lifestyle support amplifies drug response. The incremental gain from IBT isn't dramatic in percentage terms, but it's real. If you have access to structured behavioral support — a dietitian, a formal program, a bariatric clinic — using it while on semaglutide is better than not using it.

The Regain Problem: STEP-4

STEP-4 is the trial many people find uncomfortable. Participants completed a 20-week semaglutide run-in phase (losing about 10.6% of body weight), then were randomized to continue semaglutide or switch to placebo.

Over the following 48 weeks:

• Placebo group: regained about two-thirds of their lost weight (returned toward baseline)
• Semaglutide continuation group: continued to lose approximately 7.9% more weight

The message from STEP-4 is direct: semaglutide's effects are contingent on continued use. Weight regain after stopping is the norm, not the exception. This isn't a character flaw — it reflects the drug's mechanism. You're pharmacologically suppressing appetite. Stop the drug, appetite returns.

This has real implications for how patients and providers should think about duration of treatment. Semaglutide for chronic weight management is increasingly framed as a long-term therapy, not a 12-month course.

STEP-5: Two-Year Data

STEP-5 ran a 104-week version of the STEP-1 protocol and found 15.2% mean weight loss — essentially unchanged from the 68-week number. This matters because it suggests the weight loss is durable while on the drug, without tolerance attenuating the effect over time.

There was modest regain between weeks 60–104 compared to the peak effect around week 60 in some analyses, but the plateau was stable rather than progressive.

Why Your Number May Look Different

Even if you're perfectly adherent and have no diabetes, several factors shape where you land on the responder distribution:

• Dose attained: participants who had to stay at sub-maximal doses (1.0 or 1.7 mg) due to tolerability tend to see proportionally less weight loss. The 2.4 mg dose is meaningfully more effective than lower levels for weight, even if the side effect difference is modest.
• Protein intake and resistance training: people preserving more lean mass tend to look better on body composition measures, though this doesn't necessarily change the scale number.
• Sleep and cortisol: high chronic stress and poor sleep blunt GLP-1 effectiveness through mechanisms that aren't fully understood but are clinically observed.
• Gut microbiome variability: emerging research suggests individual variation in gut microbiome composition affects GLP-1 response, though this isn't actionable yet.
• Starting BMI: people with higher starting BMI typically lose more weight in absolute terms (kg) but similar percentages; the relationship isn't perfectly linear.

The 14.9% figure is a real, rigorous finding. It's also a population average, not a promise.

Related reading

• Semaglutide: the complete guide
• Why your GLP-1 weight loss plateaued
• Tirzepatide vs semaglutide
• GLP-1 weight loss expectations