STEP-3 trial deep-dive: semaglutide 2.4 mg plus intensive behavioral therapy (Wadden 2021 JAMA)
Last verified 2026-05-28 · Phase 3a · Completed (results published February 24, 2021) · NCT03611582 ↗
STEP-3 (NCT03611582) is the Semaglutide Treatment Effect in People with obesity trial that paired weekly semaglutide 2.4 mg with intensive behavioral therapy (IBT) in adults without diabetes. Novo Nordisk randomized 611 participants 2:1 to weekly subcutaneous semaglutide 2.4 mg (n=407) or matching placebo (n=204), with both arms receiving the same IBT package: 30 individual counseling visits with a registered dietitian over 68 weeks, a portion-controlled low-calorie diet of 1000-1200 kcal/day using meal replacements for the first 8 weeks, transition to a 1200-1800 kcal/day conventional-food hypocaloric diet through week 68, and a physical-activity prescription that started at 100 minutes per week and increased by 25 minutes every 4 weeks to a target of 200 minutes per week. Wadden and colleagues published the primary results in JAMA on February 24, 2021. The semaglutide arm lost a mean 16.0% of body weight versus 5.7% on placebo — a 10.3 percentage-point additional benefit on top of one of the most intensive lifestyle programs ever combined with an obesity drug in a registrational trial.
- Enrollment
- 611
- Duration
- 68 weeks of treatment with 30 IBT visits, plus a 7-week off-treatment follow-up (week 75)
- Drug
- Semaglutide 2.4 mg (Wegovy)
- Population
- Adults aged 18 or older with BMI ≥30, or ≥27 with at least one weight-related comorbidity (hypertension, dyslipidemia, obstructive sleep apnea, or cardiovascular disease). Type 2 diabetes was excluded; HbA1c ≥6.5% at screening was an exclusion criterion. Mean baseline body weight 105.8 kg (SD 22.9), mean BMI 38.0 kg/m² (SD 6.7), mean age 46 years (SD 13), 81.0% women (495/611), conducted at 41 sites in the United States from August 2018 to April 2020.
Primary endpoint
Percentage change in body weight from baseline to week 68 (coprimary, treatment-policy estimand)
Treatment arm
−16.0%
Comparator
−5.7%
Treatment difference: Estimated treatment difference −10.3 percentage points (95% CI −12.0 to −8.6; P<0.001)
Both arms received identical 30-visit intensive behavioral therapy with structured calorie and activity targets, so the 10.3-point difference isolates the pharmacologic effect of semaglutide layered on top of maximal lifestyle intervention.
Secondary endpoints
| Endpoint | Treatment | Comparator | Difference |
|---|---|---|---|
| Body-weight reduction ≥5% at week 68 Nearly half of the placebo arm met the conventional 5% clinical threshold on IBT alone, demonstrating that the lifestyle program itself was effective and not a sham comparator. | 86.6% | 47.6% | P<0.001 |
| Body-weight reduction ≥10% at week 68 | 75.3% | 27.0% | P<0.001 |
| Body-weight reduction ≥15% at week 68 More than half of the semaglutide arm crossed the 15% threshold that approaches the lower bound of sleeve-gastrectomy outcomes at one year. | 55.8% | 13.2% | P<0.001 |
| Body-weight reduction ≥20% at week 68 | 35.7% | 3.7% | P<0.001 |
| Absolute body-weight change at week 68 (kg) | −16.8 kg | −6.2 kg | Estimated treatment difference approximately −10.6 kg |
| Waist-circumference change at week 68 (cm) | −14.6 cm | −6.3 cm | Estimated treatment difference −8.3 cm; P<0.001 |
| Systolic blood pressure change at week 68 (mmHg) | −5.6 mmHg | −1.6 mmHg | Estimated treatment difference −3.9 mmHg |
| Diastolic blood pressure change at week 68 (mmHg) | −3.0 mmHg | −0.8 mmHg | Estimated treatment difference −2.2 mmHg |
| HbA1c change at week 68 (percentage points) Participants were non-diabetic at entry; the HbA1c shift reflects movement within normoglycemic and prediabetic ranges and is consistent with the prediabetes-regression signal seen in STEP-1. | −0.51 | −0.27 | Estimated treatment difference approximately −0.24 percentage points |
| Triglycerides change at week 68 (% from baseline) | −22.5% | −6.5% | — |
Adverse events
| Event | Treatment rate | Comparator rate |
|---|---|---|
| Any gastrointestinal disorder Highest pooled GI rate of any STEP trial, attributable to the intensive caloric restriction layered on top of the GLP-1 effect. | 82.8% | 63.2% |
| Nausea Most common AE; clustered during the 16-week dose-escalation phase and largely resolved by maintenance. | 58.2% | 22.1% |
| Constipation | 36.9% | 24.5% |
| Diarrhea | 36.1% | 22.1% |
| Vomiting | 27.3% | 10.8% |
| Gallbladder-related events Consistent with the cholelithiasis signal observed across GLP-1 obesity trials and tied to the rate of weight loss. | 4.9% | 1.5% |
| Serious adverse events (any) | 9.1% | 2.9% |
| Discontinuation due to gastrointestinal events Overall study-drug discontinuation was 5.9% on semaglutide vs 2.9% on placebo; no deaths and no cases of acute pancreatitis were reported. | 3.4% | 0% |
Clinical significance
STEP-3 is the trial insurance medical directors cite when prior-authorization criteria require concurrent enrollment in a structured lifestyle program. The placebo arm — which received 30 dietitian visits, a meal-replacement low-calorie diet, and 200 minutes per week of programmed activity — lost a mean 5.7% of body weight, demonstrating that the IBT component alone is clinically meaningful and is not a sham comparator. Semaglutide added another 10.3 percentage points on top, reaching a mean 16.0% total loss with 55.8% of participants crossing the 15% threshold. The result reframes GLP-1 therapy and behavioral therapy as additive interventions rather than competing strategies, undermines the older clinical argument that drugs are only for patients who have failed lifestyle, and supports payer policies that require both. STEP-3 also produced the largest mean weight loss of any STEP trial without a head-to-head comparator, almost certainly because of the IBT scaffolding.
Frequently asked questions
Does semaglutide work without intensive behavioral therapy?
Yes, but STEP-3 demonstrates that adding IBT raises the ceiling. STEP-1, which paired semaglutide 2.4 mg with only monthly lifestyle counseling, produced a mean 14.9% weight loss. STEP-3, which paired the same drug at the same dose with 30 dietitian visits, a 1000-1200 kcal/day meal-replacement low-calorie diet for the first 8 weeks, and a graded physical-activity prescription reaching 200 minutes per week, produced a mean 16.0% weight loss. The 1.1 percentage-point difference between trials is roughly the marginal contribution of high-intensity IBT on top of the drug. The placebo arms tell the same story in reverse: STEP-1 placebo lost 2.4% on monthly counseling, STEP-3 placebo lost 5.7% on the full IBT package.
What counts as intensive behavioral therapy in STEP-3?
The STEP-3 IBT package was modeled on the CMS-covered IBT for Obesity benefit and on the Diabetes Prevention Program. It included 30 individual in-person counseling visits with a registered dietitian over 68 weeks (weekly during the first 8 weeks, then tapering), a portion-controlled low-calorie diet of 1000-1200 kcal/day using meal replacements for the first 8 weeks, transition to a 1200-1800 kcal/day conventional-food hypocaloric diet through week 68, and a physical-activity prescription that started at 100 minutes per week and increased by 25 minutes every 4 weeks to a target of 200 minutes per week of moderate-intensity activity. Both arms received the identical IBT protocol; only the injection differed.
Does insurance require IBT alongside semaglutide?
Many commercial plans and most state Medicaid programs that cover GLP-1 therapy for chronic weight management cite STEP-3 (or its summary in the Wegovy FDA label) as the rationale for requiring concurrent enrollment in a structured lifestyle program. Typical prior-authorization criteria require documentation of a reduced-calorie diet, increased physical activity, and at least monthly visits with a clinician, dietitian, or behavioral coach. The level of intensity varies — few plans require the full 30-visit STEP-3 protocol — but the principle that semaglutide is approved as an adjunct to lifestyle, not as a substitute, traces directly back to this trial.
Was IBT alone effective in STEP-3?
Yes. Participants randomized to placebo plus the full IBT package lost a mean 5.7% of body weight at week 68, with 47.6% achieving at least 5% loss, 27.0% achieving at least 10% loss, and 13.2% achieving at least 15% loss. Those are clinically meaningful outcomes by any modern obesity-medicine standard and are substantially better than the 2.4% mean placebo loss seen in STEP-1, where the placebo arm received only monthly counseling. STEP-3 confirms that high-intensity, structured behavioral therapy produces real weight loss on its own; the trial's contribution is showing that semaglutide adds to that result rather than replacing it.
How does STEP-3 versus STEP-1 inform real-world expectations?
STEP-1 and STEP-3 bracket the lifestyle-intensity continuum a real-world patient might experience. STEP-1, with monthly counseling only, reported a mean 14.9% weight loss on semaglutide; STEP-3, with weekly-then-monthly dietitian visits and meal replacements, reported 16.0%. Real-world cohort studies of community semaglutide use (Gasoyan 2024 and others) typically report mean losses in the 5-10% range, well below either trial. The most plausible explanation is that real-world patients receive substantially less behavioral support and less consistent medication adherence than either STEP arm. STEP-3 is the upper-bound benchmark; STEP-1 is the realistic ceiling for routine clinical care; observational data is the floor.
Why were both arms given IBT instead of a no-treatment control?
Designing a trial with a no-IBT control arm would have been ethically and methodologically problematic. The FDA requires that anti-obesity medications be tested as adjuncts to a reduced-calorie diet and increased physical activity, the standard-of-care baseline for obesity treatment. Withholding lifestyle counseling from a control arm in a 68-week trial of adults with obesity would have failed equipoise and produced results not applicable to real prescribing. By giving both arms identical high-intensity IBT, STEP-3 isolates the pharmacologic effect of semaglutide above and beyond maximal behavioral therapy, which is the question both regulators and payers actually need answered.
References
- 1.Wadden TA, Bailey TS, Billings LK, Davies M, Frias JP, Koroleva A, Lingvay I, O'Neil PM, Rubino DM, Skovgaard D, Wallenstein SOR, Garvey WT; STEP 3 Investigators. Effect of Subcutaneous Semaglutide vs Placebo as an Adjunct to Intensive Behavioral Therapy on Body Weight in Adults With Overweight or Obesity: The STEP 3 Randomized Clinical Trial. JAMA. 2021. PMID: 33625476.
- 2.U.S. National Library of Medicine. Research Study to Investigate How Well Semaglutide Works in People Suffering From Overweight or Obesity (STEP 3) — Study Results. ClinicalTrials.gov, NCT03611582. 2021. https://clinicaltrials.gov/study/NCT03611582
- 3.Wilding JPH, Batterham RL, Calanna S, et al.; STEP 1 Study Group. Once-Weekly Semaglutide in Adults with Overweight or Obesity. N Engl J Med. 2021. PMID: 33567185.
- 4.U.S. Food and Drug Administration. FDA Approves New Drug Treatment for Chronic Weight Management, First Since 2014 (Wegovy approval announcement). FDA News Release, June 4, 2021. 2021. https://www.fda.gov/news-events/press-announcements/fda-approves-new-drug-treatment-chronic-weight-management-first-2014