Study analysis · British journal of clinical pharmacology · 2026
Semaglutide beats liraglutide at preventing diabetes—but only after 6 months, and most people quit before it even kicks in.
After six months, semaglutide lowers diabetes risk more than liraglutide, but nearly 8 out of 10 people stop taking liraglutide before they can see any benefit.
Overview
What the study found
The study in plain English — the bottom line, every takeaway we extracted, and what to do with them.
In simple terms
This study looked at people who were already taking one of two weight-loss drugs and saw who got diabetes later. It found that people on semaglutide were less likely to get diabetes than those on liraglutide, but we don’t know if the drug itself caused that — maybe the people who got semaglutide were just healthier to begin with.
What’s the bottom line?
Two similar weight-loss drugs, semaglutide and liraglutide, were compared in real-world use to see which one better prevents type 2 diabetes.
How strong is this study?
The researchers did a good job matching people who took each drug so they were similar in age and health, and they used a lot of data. But since people weren’t randomly assigned to the drugs, we can’t be 100% sure the drug made the difference — it’s like noticing that kids who eat apples are healthier, but maybe they also play outside more.
40 / 100
- COI disclosure+40/40
- Data availabilitydata not shared
- Code availabilitycode not shared
56 / 100
- Randomizationnot randomized
- Blindingblinding unclear
- Control group+15/15
- Sample size (n=57456)+20/20
- Follow-up+10/10
100 / 100
77 / 100
- P-values+15/15
- Effect size+20/20
- Confidence intervals+15/15
- Pre-registrationnot pre-registered
Each component is scored out of 100 and then capped by the study design — a case series cannot reach the ceiling a randomised trial can, however well it is reported.
Where it sits
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 567 / 100
Probability of being correct
Groups of people are followed over time to see who develops an outcome. Strong for identifying risk factors and associations, but cannot prove causation as firmly as RCTs.
This design cannot establish causation — the findings describe an association, not a cause. This is an observational study using real-world data with no randomization. Although the researchers used propensity score matching and regression adjustment to control for confounders, unmeasured factors (like baseline weight, HbA1c, or lifestyle) could still influence outcomes. Without random assignment, we cannot rule out that differences in outcomes are due to factors other than the drug itself.
No Conflicts
No conflicts of interest identified
No conflicts of interest identified; study was funded by public institutions with no involvement in study design, conduct, or reporting.
Funders
Conflict Details
American Heart Association: Received data science award (25GLP1450320)
National Institutes of Health (NIH): Supported by NIH/NHLBI T32 HL007779
National Institutes of Health (NIH): Supported by NIH/NHLBI T32 HL007779
National Institutes of Health (NIH): Supported by NIH/NHLBI K24 HL159246
National Institutes of Health (NIH): Supported by NIH Clinical and Translational Science award UM1TR004405
Independent Analysis Safeguards
- Funding organizations had no role in the design, conduct or reporting of the study.
All authors are affiliated with academic institutions; no industry employment, equity, or consulting relationships disclosed. Funding is exclusively from public, non-industry sources with explicit independence safeguards.
Key takeaways
- 01
After 6 months, semaglutide users had 20% fewer new diabetes cases than liraglutide users; before 6 months, both were equally effective.
- 02
Neither drug showed a clear difference in preventing heart attacks or strokes.
- 03
The diabetes benefit of semaglutide takes time to show up — it’s not immediate, but once it kicks in, it’s stronger than liraglutide’s.
- 04
This matters because many people stop taking these drugs within a year.
Surprising findings
- Semaglutide’s diabetes benefit only emerged after 6 months, despite faster weight loss in early trials.People assume stronger drugs work faster—but here, the more potent drug (semaglutide) took longer to show metabolic benefits, contradicting the 'faster = better' myth.
- Liraglutide users discontinued at nearly double the rate of semaglutide users (79% vs. 53%).Liraglutide is cheaper and has been on the market longer—so you’d expect better adherence. The fact that people quit it so often suggests tolerability or dosing issues.
Practical takeaways
If you're on liraglutide and haven't seen results in 6 months, consider switching to semaglutide—but only if you can stick with it.
This study didn't measure weight loss, HbA1c, or side effects—so switching based on this alone isn't foolproof.
medium confidenceDon't quit your GLP-1 drug before 6 months—you might be giving up right before the real benefit starts.
If you're experiencing severe side effects, consult your doctor—don't just stop cold turkey.
high confidenceIf you're choosing between semaglutide and liraglutide, prioritize adherence—your chance of sticking with it matters more than small differences in efficacy.
Cost, insurance coverage, and side effects vary by person—this study doesn't tell you which is cheaper or easier to tolerate.
medium confidenceWhy this study matters
The 6-Month Delay Mystery
Semaglutide showed no diabetes risk reduction in the first 6 months (HR: 0.99), but after that, it cut risk by 20% (HR: 0.80) compared to liraglutide. This delay matches clinical trials showing semaglutide’s slower, stronger titration schedule.
People expect weight-loss drugs to work fast—but this shows the real benefit of semaglutide takes months to appear, meaning many users give up before seeing results.
The Great Drug Quitting Race
79% of liraglutide users stopped treatment within a year, compared to 53% of semaglutide users. That’s nearly 3 out of 4 people quitting liraglutide—likely due to side effects, cost, or dosing frequency.
Even if semaglutide is better, its advantage might be inflated because more people quit liraglutide—making it look worse by default.
No Heart Protection Advantage Found
Despite both drugs being marketed for heart health, the study found zero significant difference in heart attacks, strokes, or heart failure between semaglutide and liraglutide (HR: 1.25, 95% CI: 0.74–2.11).
Big pharma pushes GLP-1 drugs as heart-saving miracles—but this real-world data says: not yet. The sample was too young and too few events occurred to prove anything.
Want the whole report?
Detailed mode opens the full scientific breakdown — every score component, the methodology, conflicts of interest, the evidence analysis behind each claim, and the raw study data.
Overview
What the study found
The study in plain English — the bottom line, every takeaway we extracted, and what to do with them.
Not medical advice. For informational purposes only. Always consult a healthcare professional. Terms
Two similar weight-loss drugs, semaglutide and liraglutide, were compared in real-world use to see which one better prevents type 2 diabetes.
Research results
After 6 months, semaglutide users had 20% fewer new diabetes cases than liraglutide users; before 6 months, both were equally effective. Neither drug showed a clear difference in preventing heart attacks or strokes.
What this means - more context
The diabetes benefit of semaglutide takes time to show up — it’s not immediate, but once it kicks in, it’s stronger than liraglutide’s. This matters because many people stop taking these drugs within a year.
This study compares the real-world effectiveness of semaglutide versus liraglutide in reducing incident diabetes and cardiovascular disease among adults without diabetes or CVD prescribed for weight management.
Among 57,456 matched users, semaglutide was associated with a 12% lower risk of incident diabetes over one year compared to liraglutide, but only after six months of use; no significant difference was found for myocardial infarction, stroke, or heart failure. High discontinuation rates (53% for semaglutide, 79% for liraglutide) were observed.
Methods Used
Observational cohort study using MarketScan claims data (2021–2023); 1:1 propensity score matching of new users of semaglutide and liraglutide; Cox regression adjusted for comorbidities and medication use; stratified analysis by follow-up time due to violation of proportional hazards assumption.
Main Finding
Semaglutide use was associated with a 12% lower risk of incident diabetes (HR: 0.88; 95% CI: 0.78–0.99) compared to liraglutide, with significant benefit emerging only after six months (HR: 0.80; 95% CI: 0.68–0.94); no significant difference in CVD risk (HR: 1.25; 95% CI: 0.74–2.11).
Confidence Level
Moderate; findings are robust to sensitivity analyses for diabetes, but low event counts for CVD limit precision and causal inference.
Study Flags
Red Flags
- •No baseline weight or HbA1c measurements
- •High discontinuation rates may bias results
- •Low CVD event count limits statistical power
No biological mechanisms were identified in this study. This may be an epidemiological, observational, or survey-based study that reports associations rather than proposing causal biological pathways.
Surprising Findings
Semaglutide’s diabetes benefit only emerged after 6 months, despite faster weight loss in early trials.
People assume stronger drugs work faster—but here, the more potent drug (semaglutide) took longer to show metabolic benefits, contradicting the 'faster = better' myth.
Practical Takeaways
If you're on liraglutide and haven't seen results in 6 months, consider switching to semaglutide—but only if you can stick with it.
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 567 / 100
Probability of being correct
Groups of people are followed over time to see who develops an outcome. Strong for identifying risk factors and associations, but cannot prove causation as firmly as RCTs.
Human Cohort Study
Subject
Moderate probability
on the GRADE evidence scale
This study looked at people who were already taking one of two weight-loss drugs and saw who got diabetes later. It found that people on semaglutide were less likely to get diabetes than those on liraglutide, but we don’t know if the drug itself caused that — maybe the people who got semaglutide were just healthier to begin with.
No conflicts of interest were detected in this study. No score impact.
Strengths
- Large sample size (n=57,456) increases statistical power for detecting moderate effects.
- Used a target trial emulation framework to mimic a randomized trial design.
- Employed 1:1 matching on key variables (age, sex, index date) to reduce selection bias.
Weaknesses
- No randomization — treatment assignment was observational, leaving risk of unmeasured confounding.
- Blinding status unknown — could not assess potential detection or reporting bias.
- Lacked baseline measurements for key confounders like BMI and HbA1c, which are strongly related to outcomes.
Methodology
Evidence Keywords
Statistical Reporting
Not medical advice. For informational purposes only. Always consult a healthcare professional. Terms
Two similar weight-loss drugs, semaglutide and liraglutide, were compared in real-world use to see which one better prevents type 2 diabetes.
Research results
After 6 months, semaglutide users had 20% fewer new diabetes cases than liraglutide users; before 6 months, both were equally effective. Neither drug showed a clear difference in preventing heart attacks or strokes.
What this means - more context
The diabetes benefit of semaglutide takes time to show up — it’s not immediate, but once it kicks in, it’s stronger than liraglutide’s. This matters because many people stop taking these drugs within a year.
This study compares the real-world effectiveness of semaglutide versus liraglutide in reducing incident diabetes and cardiovascular disease among adults without diabetes or CVD prescribed for weight management.
Among 57,456 matched users, semaglutide was associated with a 12% lower risk of incident diabetes over one year compared to liraglutide, but only after six months of use; no significant difference was found for myocardial infarction, stroke, or heart failure. High discontinuation rates (53% for semaglutide, 79% for liraglutide) were observed.
Methods Used
Observational cohort study using MarketScan claims data (2021–2023); 1:1 propensity score matching of new users of semaglutide and liraglutide; Cox regression adjusted for comorbidities and medication use; stratified analysis by follow-up time due to violation of proportional hazards assumption.
Main Finding
Semaglutide use was associated with a 12% lower risk of incident diabetes (HR: 0.88; 95% CI: 0.78–0.99) compared to liraglutide, with significant benefit emerging only after six months (HR: 0.80; 95% CI: 0.68–0.94); no significant difference in CVD risk (HR: 1.25; 95% CI: 0.74–2.11).
Confidence Level
Moderate; findings are robust to sensitivity analyses for diabetes, but low event counts for CVD limit precision and causal inference.
Study Flags
Red Flags
- •No baseline weight or HbA1c measurements
- •High discontinuation rates may bias results
- •Low CVD event count limits statistical power
No biological mechanisms were identified in this study. This may be an epidemiological, observational, or survey-based study that reports associations rather than proposing causal biological pathways.
Surprising Findings
Semaglutide’s diabetes benefit only emerged after 6 months, despite faster weight loss in early trials.
People assume stronger drugs work faster—but here, the more potent drug (semaglutide) took longer to show metabolic benefits, contradicting the 'faster = better' myth.
Practical Takeaways
If you're on liraglutide and haven't seen results in 6 months, consider switching to semaglutide—but only if you can stick with it.
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 567 / 100
Probability of being correct
Groups of people are followed over time to see who develops an outcome. Strong for identifying risk factors and associations, but cannot prove causation as firmly as RCTs.
Human Cohort Study
Subject
Moderate probability
on the GRADE evidence scale
This study looked at people who were already taking one of two weight-loss drugs and saw who got diabetes later. It found that people on semaglutide were less likely to get diabetes than those on liraglutide, but we don’t know if the drug itself caused that — maybe the people who got semaglutide were just healthier to begin with.
No conflicts of interest were detected in this study. No score impact.
Strengths
- Large sample size (n=57,456) increases statistical power for detecting moderate effects.
- Used a target trial emulation framework to mimic a randomized trial design.
- Employed 1:1 matching on key variables (age, sex, index date) to reduce selection bias.
Weaknesses
- No randomization — treatment assignment was observational, leaving risk of unmeasured confounding.
- Blinding status unknown — could not assess potential detection or reporting bias.
- Lacked baseline measurements for key confounders like BMI and HbA1c, which are strongly related to outcomes.
Methodology
Evidence Keywords
Statistical Reporting
Scoring
How strong is this study?
The researchers did a good job matching people who took each drug so they were similar in age and health, and they used a lot of data. But since people weren’t randomly assigned to the drugs, we can’t be 100% sure the drug made the difference — it’s like noticing that kids who eat apples are healthier, but maybe they also play outside more.
40 / 100
- COI disclosure+40/40
- Data availabilitydata not shared
- Code availabilitycode not shared
56 / 100
- Randomizationnot randomized
- Blindingblinding unclear
- Control group+15/15
- Sample size (n=57456)+20/20
- Follow-up+10/10
100 / 100
77 / 100
- P-values+15/15
- Effect size+20/20
- Confidence intervals+15/15
- Pre-registrationnot pre-registered
Each component is scored out of 100 and then capped by the study design — a case series cannot reach the ceiling a randomised trial can, however well it is reported.
Where it sits
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 567 / 100
Probability of being correct
Groups of people are followed over time to see who develops an outcome. Strong for identifying risk factors and associations, but cannot prove causation as firmly as RCTs.
This design cannot establish causation — the findings describe an association, not a cause. This is an observational study using real-world data with no randomization. Although the researchers used propensity score matching and regression adjustment to control for confounders, unmeasured factors (like baseline weight, HbA1c, or lifestyle) could still influence outcomes. Without random assignment, we cannot rule out that differences in outcomes are due to factors other than the drug itself.
No Conflicts
No conflicts of interest identified
No conflicts of interest identified; study was funded by public institutions with no involvement in study design, conduct, or reporting.
Funders
Conflict Details
American Heart Association: Received data science award (25GLP1450320)
National Institutes of Health (NIH): Supported by NIH/NHLBI T32 HL007779
National Institutes of Health (NIH): Supported by NIH/NHLBI T32 HL007779
National Institutes of Health (NIH): Supported by NIH/NHLBI K24 HL159246
National Institutes of Health (NIH): Supported by NIH Clinical and Translational Science award UM1TR004405
Independent Analysis Safeguards
- Funding organizations had no role in the design, conduct or reporting of the study.
All authors are affiliated with academic institutions; no industry employment, equity, or consulting relationships disclosed. Funding is exclusively from public, non-industry sources with explicit independence safeguards.
Standing
The people behind it
The researchers who wrote the study this analysis is built on.
Authored by
8 researchersIf this is your work, this is how we attribute it on Fit Body Science. Ethan Cannon is listed as the lead author.