Ozempic, Mounjaro, and other GLP 1 drugs: how effective are they for weight loss
A data-backed, research-driven article about GLP 1 drugs, Ozempic and Mounjaro. A crisp documentation of GLP 1 studies by our data scientists: its benefits, side effects, varied by patient cohorts and dosages, complications, and how it varies by patient cohorts and dosages. All based on real-world trials and findings.
Soumyajoy and Piyasa
Dec 26, 2025 |
17 mins
Introduction
GLP-1 drugs moved from endocrinology to everyday conversation unusually fast. In the last two years, they’ve been discussed on late-night shows, podcasts, and celebrity interviews, and it goes with the tag, “miracle” weight-loss drugs.
Andy Cohen, for instance, has publicly described using a GLP-1drug for weight loss under medical supervision.
There’re huge noise and hype around these drugs, given the celebrity endorsements + media coverage + recent FDI approvals. However, for us, data >>> opinions and hype - especially when human lives are at stake.
This article is for you if you are looking for the truth and evidence layer behind the GLP 1 drugs: Ozempic (semaglutide) and Mounjaro (tirzepatide).
Disclaimer: This is not a guide for choosing a medication. It’s a data-backed reading of the current evidence against GLP 1 drugs, its acceptance across countries, and scepticisms around it, with detailed notes and data inferences - structured for healthcare and pharma leaders.
Where metabolism, signals, and digestion become unpredictable
For healthy individuals, digestion is mostly signal driven, where glucose rises with food intake, leading to insulin release. Satiety signals take charge, reducing appetite, normalizing glucose levels.
With obesity and type 2 diabetes, this loop becomes distorted. Insulin release becomes abnormal, and a weakened satiety signal brings back hunger sooner; intake becomes decoupled from actual energy needs.
Two individuals could consume similar diets and follow similar routines; but it becomes difficult to track glucose and weight trajectories.
For healthcare and research teams, this biological noise complicates measurement: short-term studies may look promising, while long-term outcomes remain inconsistent and averages mask individual variance.
How does GLP 1 and GIP fit into the picture?
When we eat, the gut does more than digest food. It sends signals.
Among the most important are incretin hormones—chemical messengers released from the intestine that prepare the body to handle incoming glucose. The two primary incretins are GLP-1 (Glucagon-Like Peptide-1) and GIP (Glucose-Dependent Insulinotropic Polypeptide).
Together, they amplify what is known as the incretin effect: insulin secretion that is proportional to food intake rather than fixed or delayed. In metabolically healthy individuals, the incretin effect regulates blood sugar, appetite, and energy storage without conscious control.
This incretin-based signalling is skewed and broken in obesity and type 2 diabetes.
We have already seen how insulin resistance + blunted satiety signalling + dysregulated gastric emptying exacerbates everything – making it difficult to predict even with consistent diets.
Where other medical treatments fail
For decades, the Incretin-based limitation has been a murky spot in obesity treatments and observations.
Lifestyle programs typically deliver 5–10% weight loss, much of which is regained within 2–5 years (Look AHEAD, DiRECT).
Older diabetes drugs improved glucose metrics but had minimal impact on weight or long-term cardio-metabolic risk.
Bariatric surgery achieved 20–30% weight loss and diabetes remission—but at the cost of invasiveness, expense, and limited accessibility.
As a result, a gap persisted: there was no scalable medical therapy that could reliably improve glycemic control, induce substantial weight loss, and reduce downstream risk.
This is the context in which incretin-based therapies re-emerged—not as cosmetic solutions, but as attempts to restore a broken feedback loop.
How GLP-1 leads to weight loss and HbA1c reduction
GLP-1, the naturally occurring hormone released from the small intestine after eating, performs the following actions:
Stimulates insulin secretion only when glucose is elevated
Suppresses glucagon release, limiting excess glucose production
Slows gastric emptying, reducing post-meal spikes
Acts on central appetite pathways to increase satiety
The limitation was duration. Native GLP-1 degrades within minutes, making it clinically impractical.
GLP-1 receptor agonists (GLP 1 combined with a receptor)—such as semaglutide—were designed to mimic this hormone while remaining active for days. From a systems perspective, they don’t introduce a new pathway; they stabilize and extend an existing one.
Why dual GIP + GLP-1 signaling matters
GIP, the second incretin hormone, was long overlooked because of inconsistent early results. More recent preclinical and human data reframed its role.
When GIP signaling is combined with GLP-1:
Insulin secretion increases further
Appetite suppression strengthens
Weight loss effects appear additive rather than redundant
This hypothesis led to the development of dual GIP/GLP-1 receptor agonists, commonly called tirzepatide.
Early human studies and later randomized trials demonstrated that dual agonism produced larger reductions in HbA1c and body weight than GLP-1 alone, validating the underlying biology.
What Clinicians expect from an effective medication plan for
In practice, medical professionals evaluate treatments against three benchmarks:
Glycemic control: HbA1c reduction of ≥1–2 percentage points
Weight loss: ≥5–10% body weight reduction, ideally 15–20% in severe obesity
Risk reduction: Lower cardiovascular and renal events without unacceptable safety trade-offs
For the first time, these benchmarks are being met simultaneously.
In randomized trials:
Semaglutide (STEP program) consistently produced ~12–15% weight loss and significant HbA1c reductions
Tirzepatide (SURPASS and SURMOUNT programs) achieved ~15–22% weight loss with HbA1c reductions exceeding 2 percentage points
These outcomes were previously associated only with surgical intervention, which explains the reason for acceptance and reference to GLP 1 drugs as therapy for weight loss and diabetes.
Evidence for GLP-1 drugs: trials, effectiveness proof, and statistical inferences
Before looking at outcomes, it’s worth understanding how the evidence for GLP-1 drugs was generated.
The pivotal trials behind Ozempic and Mounjaro were not limited to a single geography, diet pattern, or healthcare system. They were conducted across multiple countries, ethnicities, and lifestyle contexts, with participants spanning different baseline weights, diabetes durations, and treatment histories.
Here are three major trials that changed the narrative for GLP 1 drugs.
Adjusted Indirect Treatment Comparison (AITC)
The 2022 AITC played a meaningful role before direct head-to-head evidence was widely available. It helped:
Inform early regulatory and reimbursement discussions
Shape expectations around comparative efficacy
Bridge evidence gaps during early market adoption
However, the results must be considered contextual, not decisive. Once SURPASS-2 and later meta-analyses emerged, AITC results became supportive rather than central.
SURPASS-2 (NEJM, 2021)
SURPASS-2 was not designed as an exploratory study. It was designed as a stress test.
The trial compared tirzepatide (5, 10, 15 mg) directly against semaglutide 1 mg, both on a background of metformin— the strongest widely used GLP-1 receptor agonist dose for type 2 diabetes.
Study design highlights
A heterogenous cohort was used to conduct SURPASS 2 where patients with entrenched disease were present.
40-week, randomized, open-label, phase-3 trial
128 sites across 8 countries
1,879 participants with long-standing type 2 diabetes
Mean age: 56.6 years
Mean baseline weight: 93.7 kg
Mean baseline HbA1c: 8.28%
Mean diabetes duration: 8.6 years
Results
Noninferiority to semaglutide was established
Statistical superiority was achieved for HbA1c reduction
Weight loss followed a clear dose–response pattern, with higher doses producing materially larger reductions than semaglutide
Side effects were largely gastrointestinal and comparable across groups:
Nausea: 17–22% (tirzepatide) vs 18% (semaglutide)
Diarrhea: 13–16% vs 12%
Vomiting: 6–10% vs 8%
Severe hypoglycemia was rare (<2%)
What does SURPASS 2 results highlight
Tirzepatide did more than clear the non-inferiority bar.
Effect sizes were not only large. They were precise. Confidence intervals for HbA1c outcomes were narrow, which matters more than point estimates when judging reliability.
When we look at the side effects from a data perspective, there was no signal that superior efficacy came at the cost of disproportionate safety risk.
This combination—larger effect sizes without new safety trade-offs—is what elevated SURPASS-2 from “another positive trial” to a reference point.
Diabetologia (2024): network meta-analysis across 28 Trials
By 2024, only two direct head-to-head trials existed. To answer broader questions, researchers turned to network meta-analysis, integrating both direct and indirect comparisons.
This analysis pooled 28 randomized controlled trials involving 23,622 participants, spanning:
Multiple countries
Trial durations from 24 to 104 weeks
Background therapies centered around metformin
Including Japanese-only cohorts, adding ethnic diversity
Mean baseline characteristics across studies were remarkably consistent:
HbA1c ~8.3%
Weight ~88.8 kg
Age ~58 years
Methodologically, this was a mature analysis:
PRISMA-NMA standards
Random-effects modeling
Bias assessed using ROB-2
Confidence graded via CINeMA
Treatment ranking using P-scores
Results
Across the treatment network:
Tirzepatide showed greater reductions in HbA1c and body weight than semaglutide at comparable doses
The advantage was dose-dependent
Tirzepatide 15 mg ranked highest for both outcomes
Heterogeneity was generally low, and network consistency tests showed minimal incoherence—important signals that the comparisons were statistically sound.
Safety findings echoed earlier trials:
Gastrointestinal side effects increased with dose for both drugs
No increase in severe hypoglycemia or serious adverse events
This is the point where all the isolated evidence converged.
2025 Direct comparative meta-analysis
The most recent layer came from a direct comparative meta-analysis that excluded indirect comparisons altogether.
This study aggregated four studies—two randomized trials and two large retrospective cohorts—comprising 28,827 patients, making it the largest direct comparison dataset to date.
Across diverse populations:
Mean follow-up ~36 weeks
Baseline BMI ranged from 30.8 to 39.1
Included both T2DM and select non-T2DM cohorts
Results
Mean weight loss:
Tirzepatide: −11.4%
Semaglutide: −7.3%
Treatment difference: −4.84 kg in favor of tirzepatide (95% CI: −6.21 to −3.47)
Evidence vs unknown: what’s proved vs what still needs answers
For both Mounjaro and Ozempic, there is strong data to prove HbA1c reduction, dose-response weight loss, GI AE profile, consistent direction across RCT + real-world. But there are some areas which are still work in progress: durability, discontinuation effects, rare AE detection, subgroup response prediction, long-horizon outcomes.
How GLP 1 drugs work beyond trial period: obesity and diabetes aren’t short span conditions, but the data is limited on weight loss patterns for 3–5+ years on continuous therapy.
What happens if the consumption is interrupted: how weight gain happens once stopped, how much time it takes to regain the same amount of weight lost, and does the appetite signals revert to old conditions.
How response varies in heterogenous group: average is given more importance, but variance is also real. The super responder vs nonresponder groups and how much is explained by baseline BMI, HbA1c, insulin resistance, genetics, behavior, co-meds.
Generalization across under-represented groups: elderly patients, extreme obesity categories, other non-represented ethnic groups, etc.
Effects beyond weight loss and HbA1c: long run answers on other conditions like kidney outcomes across broader cohorts, heart failure, microvascular conditions, etc.
Are Ozempic and Mounjaro approved, referenced, and used all over the world?
Positive results don’t mean Semaglutide (Ozempic/Wegovy) and tirzepatide (Mounjaro/Zepbound) are approved all over the world. But they are accepted in countries like the United States, EU, UK, Japan, Australia, and parts of Asia and the Middle East. Their approvals are grounded in large, multi-country randomized trials that enrolled diverse populations.
Countries that strongly accepted evidence against Ozempic and Mounjaro
United States: After rigorous testing through three trial programs: STEP, SURPASS, SURMOUNT and acceptable short to medium term efficacy, FDA has approved both both semaglutide and tirzepatide. Usage is widespread, particularly in endocrinology and obesity clinics. As of late 2025, overall GLP-1 prescriptions accounted for about 6.5% of all prescriptions written in the US as of September 2025.
European Union & UK: Regulators like EMA and NICE has approved the usage of semaglutide and tirzepatide, but with restrictions on who must receive treatment. In UK, for example, semaglutide for obesity is approved but restricted to specific populations and durations, reflecting concern about sustainability at scale.
Japan & East Asia: Japan has accepted the usage of GLP 1 drugs to some extent. But they still question its efficacy across different population groups.
India: Both Ozempic and Mounjaro are now available in India through major hospitals and pharmacies. Tirzepatide is marketed locally (e.g., via Cipla partnerships), and uptake is growing among urban populations.
Pricing, availability, and market impact
Region | Ozempic (Semaglutide) | Mounjaro (Tirzepatide) |
|---|---|---|
United States | ~$350–$500 (with coverage/negotiation) | ~$975–$1,300 |
Europe | €59–€270 (country & dose dependent) | €206–€330 |
Japan | ~$169 | ~$319 |
India | ₹8,800–₹11,175 | ₹13,125–₹25,781 |
*Prices vary by dose, insurance coverage, and national pricing controls.
With GLP 1 drugs, the concern is more on the affordability than availability. In many countries where it’s sold through prescriptions,
Treatment is largely out-of-pocket
Insurance coverage is minimal
Monthly costs remain prohibitive for most patients
This has limited real-world adoption despite regulatory approval.
Other diabetic and weight loss drugs are cheaper in the short term, approved for limited duration, and often less effective. But with GLP 1 drugs, the scenario flipped. Higher cost, much higher efficacy, and chronic use model. So, the healthcare systems to treat obesity not as a lifestyle issue, but as a budget-impacting chronic disease.
Historically, chronic metabolic therapies have rarely been cheap, be it insulin analogs, DPP-4 inhibitors, or combination diabetes regimes.
Many of these already cost patients hundreds of dollars per month in developed markets and were largely inaccessible in developing ones without subsidies.
GLP-1 therapies did not introduce high pricing. But the fact that they are long-term or lifelong therapies has split the audience into categories: those who can afford long-term therapies and those whose medication bills were solely reliant on insurance.
Let’s consider this scenario for a well-developed economy:
30 to 40% of adults qualify for obesity treatment
Therapy costs $5,000–$15,000 per year
Treatment is long-term
Sustainability is still a question, even with well-funded healthcare systems. That’s why developed countries deal with GLP 1 drugs in a systematic way. There will be tight eligibility criteria, prescriptions are time-limited, and highest-risk patients are prioritized.
With developing countries, where insurance hasn’t scaled beyond life-threatening diseases, the equation is totally different. Despite the widespread availability of Ozempic and its counterparts in countries in India, it’s consumption is patterned around:
Urban, higher-income populations
Self-funded treatments
No or limited long-term adherence
Data analytics has become foundational for healthcare. Now more than ever.
Drug development, clinical trials, post-marketing surveillance, and population-level outcomes now generate volumes of data that were simply not manageable through traditional analysis alone. As therapies move faster from lab to market, the ability to interpret this data correctly becomes just as critical as the science behind the therapy itself.
This is especially visible in areas like metabolic diseases, where outcomes are influenced by a combination of biology, behavior, geography, and long-term adherence. Here, isolated signals rarely tell the full story. Patterns matter. Variability matters. And so does understanding where results hold up—and where they don’t.
That is where data analytics plays a decisive role.
In case of GLP 1 therapy, data was required to prove what biology knew for years, all of which required rigorous data design, statistical modeling, and inferences.
These therapies did not gain acceptance simply because early trials showed promising averages. Confidence built over time as data from multiple studies, geographies, and methodologies converged on the same conclusions. Meta-analyses, comparative studies, and observational datasets reinforced what individual trials suggested—while also highlighting limitations and open questions.
This layered evidence is what turns results into trust.
So, in a space where, endpoints are clinically meaningful, even a minute dose can matter, and heterogeneity is expected, more than technical proficiency is required.
When analytics is applied with domain awareness, it accelerates learning instead of distorting it. It helps teams identify what is statistically significant and clinically relevant. It clarifies where evidence is strong, where it is emerging, and where it is still insufficient.
Explaining this GLP 1 studies
For therapies like GLP-1 drugs, data-driven analysis has made it possible to:
Compare treatments meaningfully across trials
Understand dose–response relationships
Separate short-term efficacy from noise and long-term trends
Identify where results generalize—and where they don’t
It is about proving how, for whom, and under what conditions it works. That distinction matters in a field where outcomes affect both lives and healthcare systems.
Win with a multidisciplinary team
The work behind this article—and the accompanying whitepaper—follows the same principle.
We did not look for conclusions. We looked for patterns, consistency, and limits.
By analyzing randomized trials, meta-analyses, and real-world data together, we aimed to show not just what the data says about GLP-1 therapies—but why it says it, and where uncertainty still remains.
This is the role of data science in healthcare today: to move conversations from opinion to evidence, from belief to measurement.
When done well, it helps the right therapy reach the right patient at the right time—and helps life-saving innovation scale responsibly.
Contact us and we would be happy to explain our findings or discuss other upcoming healthcare data trends.
Finally,
As therapies become more complex and adoption scales globally, the role of data analytics is no longer limited to validation and interpretation. It is part of how modern healthcare learns, adapts, and improves.
The GLP-1 case is a reminder of what becomes possible when rigorous science is paired with equally rigorous analysis—allowing results to be tested, challenged, and trusted before they are widely acted upon.
What this means for different leaders
If you run clinical development, you need analytics that go beyond primary endpoints—capable of explaining heterogeneity, durability, and real-world performance early enough to inform trial design and downstream strategy.
If you manage formularies, you need better charts that connect clinical efficacy to population-level value—linking outcomes, adherence, and cost over time, not just sheets pointing trial averages.
If you lead health population health programs, you need data and insights that show where interventions scale, where they don’t, and how impact varies across populations.
References:
by Soumyajoy and Piyasa
Working at datakulture
