Ozempic Evidence Base Graded by GRADE: What the Clinical Data Actually Show

What GRADE Means and Why It Matters for Ozempic Prescribing

GRADE (Grading of Recommendations Assessment, Development and Evaluation) rates the certainty of evidence on a four-level scale: High, Moderate, Low, and Very Low. A High-certainty rating means further research is very unlikely to change the estimated effect. A Moderate rating means confidence is reasonable but new data could alter the conclusion. Understanding where semaglutide sits on that scale changes how confidently a clinician can counsel a patient.

The ADA Standards of Care in Diabetes 2024 state: "GLP-1 receptor agonists with proven cardiovascular benefit are recommended for patients with type 2 diabetes and established cardiovascular disease, independent of HbA1c or baseline glucose-lowering therapy." That recommendation carries a Level A (highest) evidence grade in the ADA framework, which maps directly to GRADE High certainty. [1]

Why randomized data dominate here

Semaglutide 0.5 to 2.0 mg has been studied in more than 10 phase 3 randomized controlled trials enrolling tens of thousands of participants. The risk of bias across the SUSTAIN program is generally low: trials were double-blind or active-controlled, used centralized randomization, and pre-specified primary endpoints. Observational confounding is not a major concern for the glycemic and cardiovascular outcomes.

How GRADE downgrades apply

Even within the SUSTAIN program, some endpoints lose a GRADE point for inconsistency (heterogeneous populations) or imprecision (narrow confidence intervals crossing a clinical threshold). The sections below flag each downgrade explicitly so the reader can judge.

GRADE High: Glycemic Efficacy (HbA1c and Fasting Glucose)

Semaglutide 0.5 to 2.0 mg produces clinically significant HbA1c reductions across a wide range of comparators and backgrounds. The evidence is large in volume, consistent in direction, and low in risk of bias, earning a GRADE High rating for this outcome.

SUSTAIN-1 through SUSTAIN-5: Placebo- and Active-Controlled Data

SUSTAIN-1 (N=388, 30 weeks, monotherapy) showed semaglutide 0.5 mg reduced HbA1c by 1.45 percentage points and semaglutide 1.0 mg by 1.55 percentage points versus 0.02 percentage points for placebo (both P<0.0001). [2] SUSTAIN-2 (N=1,231, 56 weeks) compared semaglutide against sitagliptin 100 mg; semaglutide 1.0 mg outperformed sitagliptin by 0.77 percentage points on HbA1c reduction. [3]

SUSTAIN-3 (N=813, 56 weeks) pitted semaglutide 1.0 mg against exenatide extended-release 2 mg. Semaglutide produced a 1.5 percentage point HbA1c reduction versus 0.9 percentage points for exenatide ER (difference: 0.62 percentage points, P<0.0001). [4] SUSTAIN-4 (N=1,089, 30 weeks) showed semaglutide 1.0 mg reduced HbA1c by 1.21 percentage points versus 0.83 percentage points for insulin glargine, despite lower rates of hypoglycemia. [5]

SUSTAIN-7: Head-to-Head vs. Dulaglutide

SUSTAIN-7 (N=1,201, 40 weeks) is one of the most clinically important direct comparisons in the GLP-1 class. Semaglutide 1.0 mg reduced HbA1c by 1.9 percentage points versus 1.4 percentage points for dulaglutide 1.5 mg (difference: 0.5 percentage points, P<0.001). Body weight fell by 6.5 kg with semaglutide 1.0 mg versus 3.0 kg with dulaglutide 1.5 mg. At 0.5 mg vs. 0.75 mg doses, HbA1c reduction was 1.5 versus 1.1 percentage points and weight loss was 4.6 vs. 2.3 kg, respectively. [6]

These data are internally consistent, replicated across comparators, and show a clear dose-response. No GRADE downgrade applies.

SUSTAIN-10 and SUSTAIN CHINA: Broader Population Validity

SUSTAIN-10 (N=577, 30 weeks) confirmed superiority over liraglutide 1.2 mg: semaglutide 1.0 mg reduced HbA1c by 1.7 versus 1.0 percentage points (P<0.0001). [7] Consistency across Asian populations (SUSTAIN CHINA) and elderly subgroups did not reveal meaningful heterogeneity, supporting external validity without GRADE downgrade for indirectness.

GRADE High: Cardiovascular Outcomes

Ozempic's cardiovascular evidence is the strongest pillar of its clinical profile and the primary reason guidelines recommend it preferentially in patients with or at high risk for atherosclerotic cardiovascular disease.

SUSTAIN-6: Cardiovascular Safety and Superiority

SUSTAIN-6 (N=3,297, 104 weeks) was designed as a non-inferiority cardiovascular outcomes trial (CVOT) but showed superiority. The primary three-point MACE endpoint (cardiovascular death, non-fatal myocardial infarction, non-fatal stroke) occurred in 6.6% of semaglutide-treated patients versus 8.9% of placebo-treated patients (HR 0.74, 95% CI 0.58 to 0.95, P<0.001 for non-inferiority; P=0.02 for superiority). [8] The number needed to treat to prevent one MACE event over two years was approximately 45.

A GRADE downgrade for imprecision might be considered given the relatively wide confidence interval and the trial's primary non-inferiority design. Most guideline panels retain a High-certainty rating by combining SUSTAIN-6 with the SELECT data below.

SELECT: Cardiovascular Outcomes in Non-Diabetic Obesity

SELECT (N=17,604, mean follow-up 34.2 months) enrolled adults with established cardiovascular disease and overweight or obesity but without diabetes. Semaglutide 2.4 mg (Wegovy) reduced three-point MACE by 20% (HR 0.80, 95% CI 0.72 to 0.89, P<0.001). [9] Although SELECT used the 2.4 mg dose rather than the 0.5 to 2.0 mg Ozempic dose range, the mechanism is identical and the biological plausibility for extrapolation is high. Guidelines such as the 2024 ADA Standards cite both trials together. [1]

The combined SUSTAIN-6 and SELECT evidence base earns GRADE High for the claim that semaglutide reduces MACE in adults with type 2 diabetes and established or high-risk ASCVD.

GRADE Moderate: Body Weight Reduction in Type 2 Diabetes

Body weight falls consistently with semaglutide in the SUSTAIN program. The rating drops to Moderate rather than High because weight was a secondary endpoint in most trials and the confidence intervals around between-group differences are wider than for HbA1c.

Pooled weight loss across SUSTAIN

Across SUSTAIN-1 through SUSTAIN-8, semaglutide 1.0 mg produced mean weight reductions of 3.5 to 6.5 kg versus comparators ranging from 0.9 kg (placebo) to 3.0 kg (dulaglutide 1.5 mg). [6] The between-group advantage is real but the absolute magnitude varies with baseline BMI, diet co-intervention, and trial duration.

Why only Moderate certainty?

Three factors apply a partial GRADE downgrade here. First, weight was not the primary endpoint in any SUSTAIN trial, creating potential outcome-reporting bias. Second, the follow-up period rarely exceeded 56 weeks, raising questions about durability. Third, patients who discontinue semaglutide regain weight rapidly, a point confirmed by the STEP-4 trial (Wegovy arm, 68 weeks) where withdrawal after 20 weeks led to near-complete weight regain by week 68. [10] This does not reduce the certainty of the short-term effect, but it does affect how the clinical benefit is framed for patients.

Practical weight expectations in T2D

A useful clinical anchor: in SUSTAIN-7, the single best comparator-controlled weight trial in T2D, semaglutide 1.0 mg produced 6.5 kg loss versus 3.0 kg with dulaglutide 1.5 mg over 40 weeks. [6] For a 100 kg patient, that is a 6.5% body weight reduction, sufficient to produce clinically meaningful improvements in blood pressure, liver fat, and sleep apnea by most established thresholds.

GRADE Moderate: Renal Outcomes

Semaglutide's effects on kidney function are a growing area of investigation. The FLOW trial (N=3,533) was a dedicated renal outcomes trial for semaglutide 1.0 mg in patients with type 2 diabetes and chronic kidney disease. Published in 2024, it showed semaglutide reduced the primary composite kidney outcome (40% sustained decline in eGFR, kidney failure, or renal/cardiovascular death) by 24% (HR 0.76, 95% CI 0.66 to 0.88, P<0.001). [11]

The GRADE certainty for renal endpoints rates as Moderate rather than High because FLOW is a single trial and the lower bound of the confidence interval, while statistically significant, allows for a smaller true effect. The 2024 KDIGO guidelines now include GLP-1 RA consideration for CKD with T2D, consistent with this Moderate-certainty evidence. [12]

GRADE Low: Effects on Non-Alcoholic Fatty Liver Disease (NAFLD/MASLD)

Semaglutide shows consistent histological improvements in NASH/MASLD in phase 2 data. The NASH phase 2 trial (N=320, 72 weeks) showed 59% of semaglutide 0.4 mg daily patients achieved NASH resolution versus 17% placebo (P<0.001). [13] However, 0.4 mg daily is not the approved subcutaneous weekly formulation, and phase 3 NASH data for the weekly SC formulation were still maturing as of early 2025.

The GRADE certainty for NAFLD/MASLD resolution using semaglutide 0.5 to 2.0 mg SC weekly is Low: one phase 2 trial, different dose form, no hard clinical endpoints (no trial of cirrhosis or liver-related mortality as primary outcome). This should be communicated clearly when patients ask whether Ozempic will "fix" their fatty liver.

GRADE Low to Very Low: Cognitive Function and Addiction-Related Outcomes

Observational and mechanistic data suggest GLP-1 receptors in the brain may modulate reward circuitry, appetite signaling, and possibly neurodegeneration. A 2023 analysis using Danish registry data (N=23,731) found semaglutide associated with lower rates of alcohol use disorder and smoking relapse. [14] These are hypothesis-generating findings. No RCT with cognitive or addiction primary endpoints has been completed for semaglutide 0.5 to 2.0 mg SC weekly as of early 2025.

GRADE certainty: Very Low. The evidence is observational, susceptible to healthy-user bias, and not yet replicated in controlled settings. Clinicians should not recommend Ozempic for cognitive or addiction indications outside a research context.

Safety Evidence: Characterization and GRADE Applicability

Safety evidence in GRADE is rated by the same framework but applies primarily to serious harms. For semaglutide, the key signals are:

Gastrointestinal adverse events

Nausea is reported in 15 to 20% of patients in the SUSTAIN program (vs. 5 to 8% placebo), vomiting in 5 to 10%, and diarrhea in 8 to 12%. [8] These are GRADE High-certainty adverse events: large, consistent, dose-dependent, and replicated across all trials. The dose-escalation schedule (starting at 0.25 mg for 4 weeks) reduces but does not eliminate GI burden.

Medullary thyroid carcinoma and pancreatitis

The boxed warning for medullary thyroid carcinoma (MTC) is based on rodent data showing dose-dependent C-cell hyperplasia. No human RCT has demonstrated a statistically significant MTC signal. The GRADE certainty for MTC risk in humans is Very Low (animal data only), but the FDA maintains the warning given the severity of potential harm. Pancreatitis is listed as a potential risk; the meta-analysis by Bethel et al. (2018) found no statistically significant increase in confirmed pancreatitis in cardiovascular outcomes trials for GLP-1 RAs as a class. [15]

Diabetic retinopathy complications

SUSTAIN-6 reported a higher rate of diabetic retinopathy complications in the semaglutide group (3.0% vs. 1.8%, HR 1.76, P<0.001). [8] The mechanism is believed to be rapid HbA1c lowering rather than a direct semaglutide effect on the retina, analogous to worsening retinopathy with intensive insulin therapy. GRADE certainty for this signal: Moderate. Clinicians should screen for advanced retinopathy before initiating semaglutide in patients with poorly controlled diabetes.

How the GRADE Profile Informs Real Prescribing Decisions

A complete GRADE evidence profile organizes outcomes by their certainty level and direction, allowing clinicians to weigh benefits against harms rationally.

Patients with T2D and established ASCVD

For this group, High-certainty evidence supports prescribing semaglutide as a cardiovascular risk-reduction agent that also lowers HbA1c. The 2024 ADA Standards of Care place GLP-1 RAs with proven CV benefit (including semaglutide) at the top of the glucose-lowering algorithm for this population regardless of HbA1c level, an approach also endorsed by the European Association for the Study of Diabetes (EASD). [1]

Patients with T2D and CKD stages G3, G4

Moderate-certainty renal outcome evidence from FLOW supports semaglutide use with dose adjustment guidance. The 24% reduction in the composite renal endpoint is clinically meaningful, and the drug is generally well-tolerated in moderate CKD. Prescribers should verify eGFR trajectory and ensure adequate hydration counseling given the risk of dehydration from GI side effects. [11]

Patients seeking weight loss without T2D

The approved agent for obesity is semaglutide 2.4 mg (Wegovy), not Ozempic. The GRADE evidence for weight loss at 0.5 to 2.0 mg in non-diabetic patients is Moderate at best and drawn from trials where T2D was required for enrollment. Off-label prescribing of Ozempic for obesity is not supported by the same evidence base as Wegovy, and the FDA explicitly approved distinct products for distinct indications.

Summary GRADE Table for Semaglutide 0.5 to 2.0 mg SC Weekly

| Outcome | Best Trial(s) | Certainty | Effect Size | |---|---|---|---| | HbA1c reduction (T2D) | SUSTAIN-1 through SUSTAIN-10 | High | 1.1 to 1.8 pp vs. Placebo | | 3-point MACE (T2D + ASCVD) | SUSTAIN-6, SELECT | High | HR 0.74 to 0.80 | | Body weight (T2D) | SUSTAIN-7 | Moderate | 3.5 to 6.5 kg vs. Comparators | | Composite renal endpoint | FLOW | Moderate | HR 0.76 | | NASH resolution | Phase 2 NASH trial | Low | 59% vs. 17% placebo (different dose form) | | Cognitive / addiction outcomes | Registry data only | Very Low | Hypothesis-generating only | | MTC risk in humans | Rodent data only | Very Low | No signal in RCTs | | Diabetic retinopathy worsening | SUSTAIN-6 | Moderate | HR 1.76 (with rapid HbA1c lowering) |