SOURCE-LINKED STUDY SUMMARY
Liraglutide & Prediabetes: What the 3-Year Trial Found
PED EVIDENCE INTERPRETATION
Did liraglutide delay diabetes in people with prediabetes?
During treatment, diabetes was diagnosed in 2% of liraglutide participants versus 6% on placebo by 160 weeks. Only half of randomized participants completed follow-up, and those who withdrew were not followed after discontinuation. [1]
- Design
- Double-blind randomized trial; 2,254 participants
- Population
- Adults with prediabetes and obesity, or overweight with comorbidities
- Comparison
- Daily subcutaneous liraglutide 3.0 mg vs placebo; both with diet and activity support
- Duration
- 160 weeks
The evidence at a glance
| Outcome | Liraglutide | Placebo | Interpretation |
|---|---|---|---|
| Diabetes diagnosed on treatment [1] | 26/1,472 (2%) | 46/738 (6%) | Hazard ratio 0.21; 95% CI 0.13–0.34 |
| Body-weight change [1] | −6.1% | −1.9% | Estimated difference −4.3 percentage points; 95% CI −4.9 to −3.7 |
| Serious adverse events [1] | 15% | 13% | Descriptive incidence; does not establish equivalence |
| Withdrew before week 160 [1] | 47% | 55% | Missing follow-up is central to interpretation |
OUR READING OF THE EVIDENCE
What this changes
The randomized comparison supports delayed diabetes diagnosis during the studied treatment period in this population. [1]
What it does not establish
The result does not establish permanent prevention after stopping treatment. Missing outcomes after withdrawal leave uncertainty about the long-term trajectory. [1]
Put it in context
Sources and transparency
Funded by Novo Nordisk. Abstract-based interpretation; not an independent assessment of all adverse events or methodological quality. [1]
Questions this study helps answer
Does this prove diabetes was permanently prevented?
No. The reported benefit concerns diagnoses during treatment. It does not establish permanent prevention after stopping liraglutide. [1]
Why is the dropout rate important?
Missing outcomes can change how well a result represents everyone originally randomized. People who withdrew were not followed after discontinuation, which limits conclusions beyond those observed. [1]
Original compact summary and study details
- What was studied
- A randomized, double-blind human trial compared liraglutide with placebo. The primary outcome was time to type 2 diabetes by 160 weeks.
- Who / model studied
- 2254 adults with prediabetes and obesity, or overweight with additional conditions, were randomized.
- Studied dose & schedule
- Participants received subcutaneous liraglutide 3.0 mg or matched placebo daily, alongside reduced-calorie eating and increased activity.
- What the study found
- Diabetes was diagnosed during treatment in 2% versus 6%. Weight fell 6.1 versus 1.9 percent at 160 weeks. Serious adverse events occurred in 15% versus 13%.
- Limitations & context
- Participants who withdrew were not followed after stopping treatment, limiting interpretation of outcomes beyond continued treatment.
Read summary source ↗ · Prepared 9/23/2026 · AI-assisted, source-based summary
Review record & corrections
Summary prepared: . Source scope: abstract. This is an AI-assisted source summary, not an independent clinician sign-off.
No public field-by-field revision history is available for this card yet. The date above is the summary date, not proof that every field was updated then.
Inspect the source · Correction policyResearch findings apply to the population and conditions studied. This educational summary is not a treatment recommendation or a formal assessment of study quality.
Original publication ↗ · DOI: 10.1016/s0140-6736(17)30069-7