The GLP-1 ROI Evidence
The published evidence on whether GLP-1 coverage pays for itself. Includes cost-offset studies, cost-effectiveness analyses, real-world claims data, and an interactive calculator.
Last updated: April 21, 2026
Context
Overview
This page compiles the published evidence on GLP-1 return on investment for commercial payors and self-insured employer populations. It covers medical cost offset studies, SELECT cardiovascular data, actuarial models, real-world claims data, and the assumptions that drive conflicting conclusions. The calculator at the bottom of the page uses these inputs.
Other pages in this series cover pricing, persistence and adherence, weight regain after stopping, side effects and safety, and real-world weight loss.
Core question
Medical cost-offset evidence
Does GLP-1 spending reduce downstream medical costs enough to offset drug spend? Independent analyses consistently find that drug spend exceeds medical offsets at current prices. Manufacturer-funded analyses report medical cost reductions but exclude drug costs from the totals.
Independent analyses
| Study | Population | N | Horizon | Key finding | Net direction |
|---|---|---|---|---|---|
| Prime Therapeutics JMCP 2024; AMCP 2025 poster | Commercial, obesity without T2D | 3,346 vs 8,343 controls | 2 years | Total cost of care +$6,994/user in year 1, +$4,206 in year 2. Medical costs +$1,338 (not statistically significant). No reduction in obesity-related events. | No offset |
| Aon Phase 2 Jan 2026 | Commercial, all indications (70% T2D) | 192,000 matched | 18–30 months | Medical cost growth slowed 3–6 ppts vs controls. At ≥80% adherence: 7–9 ppts. Total cost of care still higher with drug spend included. | Slower growth only |
| AssuredPartners Feb 2025 | Commercial, diabetics | 9,236 vs 10,625 | 5 years | Annual medical savings ~$560. Annual drug cost ~$6,540. Cost of care CAGR 13.01% vs 18.25% in non-users. | Net negative |
| UnitedHealthcare Internal, reported in industry decks | Commercial, metabolic conditions | Not disclosed | 12 months | 91% increase in PMPM costs 12 months after initiation. | Cost increase |
| EBRI Oct 2025 | Commercial employer simulation | 49.3M modeled | Annual | Projected premiums +5.3% to +13.8%. Report concludes "no evidence to suggest savings would fully offset GLP-1 prices." | Premium increase |
| CBO Oct 2024 | Medicare, obesity + CV | National | 9 years (2026–2034) | Projected +$35B net federal spending. By 2034: $7.1B drug spend vs $1B savings. | No offset |
| Hwang et al. JAMA Health Forum, Apr 2025 | Medicare simulation, obesity without T2D/CVD | 30M modeled | 10 years (2026–2035) | Drug costs $65.9B, offsets $18.2B. Net increase $47.7B. Moderate scenario (5% uptake, 30% discount) still +$8B. | No offset |
Manufacturer-funded analyses (Novo Nordisk)
| Study | Population | N | Horizon | Key finding (caveat in-line) | Net direction |
|---|---|---|---|---|---|
| OFFSET study Diabetes Obes Metab, 2022 | T2D + CVD hospitalization | Not specified | Not specified | Budget neutral: lower inpatient/outpatient offset higher drug costs. T2D+CVD population only; pre-semaglutide era. | Budget neutral |
| Komodo/Novo multimorbidity ERPOR, Jan 2026 | OW/OB + ≥2 complications | Not specified | 101 days | Medical costs 27% lower ($3,870/yr); drug cost excluded from total. 101-day follow-up. | Projected savings |
| SHINE-HF Clinical Therapeutics, 2025 | OW/OB + heart failure | 408 | Not specified | Medical costs 28% lower (−$8,544/yr); drug cost excluded from total. Small sample. | Projected savings |
| Novo/Value in Health 2025 | Medicare, all indications | National model | 10 years | Projected net savings of $715M across all indications. Includes T2D (largest savings); assumes generic entry year 7. | Projected savings |
Independent and manufacturer-funded analyses reach different conclusions on the same underlying question, largely because they measure different things. Independent analyses measure total cost of care (medical + pharmacy). Manufacturer-funded analyses typically report medical costs only, with drug costs excluded from the total. Both approaches are internally consistent but answer different questions, and readers should check which denominator a study uses before citing its conclusion.
A second source of divergence is population selection. Manufacturer analyses focus on high-risk subgroups (T2D with CVD, OW/OB with multiple complications, heart failure) where medical costs are highest and most potentially offsettable. Independent analyses more often cover broader commercial obesity populations where baseline medical costs are lower and offsets are harder to demonstrate.
Cardiovascular evidence
SELECT and cost-effectiveness
SELECT (Lincoff et al., NEJM 2023) is an often cited trial for the cost-offset argument: a 20% reduction in major adverse cardiovascular events over 3.3 years in a population with established atherosclerotic disease. The cost-effectiveness of that benefit depends heavily on how closely an employer population resembles the trial cohort.
SELECT enrolled older adults (mean age 61.6, 72% male) with established CVD and no diabetes. Typical employer populations are working-age, largely without established CVD, and have real-world persistence of 32–63% at 12 months. The table below compares the cohorts on eight characteristics.
| Characteristic | SELECT trial | Typical employer population |
|---|---|---|
| Mean age | 61.6 years | 25–55 (working age) |
| Sex | 72% male | ~50/50 |
| CVD status | 100% established ASCVD (prior MI, stroke, or PAD) | Majority without established CVD |
| Diabetes | Excluded (HbA1c <6.5%) | 30–40% of CVD-risk subgroup has T2D |
| Background therapy | 90% on statins, 86% on antiplatelets | Substantially lower |
| Persistence | 82.5% of treatment time | 32–63% at 12 months; 8% at 3 years |
| MACE reduction | 20% (HR 0.80, P<0.001) | Not directly measured in employer populations |
| Number needed to treat | 67 over 3.3 years | Likely higher given lower baseline CV risk |
Sources: Lincoff et al., NEJM 2023; FDA CV indication approved label, March 2024.
Actuarial guidance
Actuarial and consulting models
EBRI, Mercer, Milliman, and Aon are the organizations employers most commonly rely on for actuarial guidance. None projects positive ROI within an employer-relevant time horizon at current prices. Some show favorable directional signals under specific conditions (high adherence, managed programs).
EBRI and AssuredPartners project premium increases and net negative returns. Mercer and Milliman decline to project positive ROI and recommend plan-specific claims analysis. Aon reports favorable medical cost trends in the adherent subgroup at 18–30 months, though total cost of care remains higher than controls.
| Organization | Date | Type | Key assumptions | Projected impact | Net direction |
|---|---|---|---|---|---|
| EBRI | Oct 2025 | Premium simulation | $617–766 per 30-day supply; 42% persist 12+ weeks; $90 copay | +5.3% to +13.8% premium increase | Negative |
| Mercer | Nov 2025 | Framework | ~$1,000/mo list; 2–5% utilization; 1-in-12 persist at 3 years | "Immediate ROI remains uncertain." Recommends plan-specific claims analysis. | Uncertain |
| Milliman | Aug 2023; Sep 2025 | Markov forecast | >68% do not maintain 12-month therapy; models 26% drug spend waste | Utilization steady-state 7.9–12.7% of commercial population | Waste modeled |
| Aon | Jan 2026 | Claims analysis | 192,000 users; digital twin matching; 35% flat rebate | Medical cost trend slowed 7% by year 2 (adherent); MACE hospitalizations down 44% | Favorable trend |
| AssuredPartners | Feb 2025 | 5-year claims | 9,236 diabetic users; medical + Rx costs | Medical savings $560/yr; drug cost $6,540/yr | Negative |
Cost-effectiveness
Cost-effectiveness analyses (ICER and QALY)
Cost-effectiveness analyses answer a different question than employer ROI studies. They ask whether the additional health benefit justifies the additional spending at a population level, not whether an employer recoups its spending in a budget window.
The standard metric is the incremental cost-effectiveness ratio (ICER) — the additional cost per additional quality-adjusted life year (QALY) gained. Standard US willingness-to-pay thresholds are $100,000–$150,000 per QALY. A ratio below the threshold means the intervention is considered cost-effective; it does not mean an employer sees net savings.
Three major 2025 analyses reached different conclusions. Hwang et al. (JAMA Health Forum) modeled weight-mediated benefits only and found $467,676/QALY (semaglutide) and $197,023/QALY (tirzepatide), with 0% probability of cost-effectiveness at $100K/QALY. ICER's 2025 final report incorporated SELECT cardiovascular outcomes and found $61,400/QALY and $53,400/QALY, cost-effective at standard thresholds. Novo Nordisk's SELECT-based analysis (McEwan et al., 2025) estimated $136,271/QALY at list price and $32,219/QALY with a 48% rebate. The main driver of divergence is whether SELECT cardiovascular outcomes are included.
Note: in the table below, "ICER" in the column header refers to the Institute for Clinical and Economic Review (the nonprofit that published the 2025 report), not the metric defined above.
| Parameter | Hwang et al. (JAMA Health Forum, Mar 2025) | ICER final report (Dec 2025) |
|---|---|---|
| Semaglutide ICER | $467,676/QALY | $61,400/QALY |
| Tirzepatide ICER | $197,023/QALY | $53,400/QALY |
| Cost-effective at $100K/QALY? | No (0% probability) | Yes (well below threshold) |
| SELECT CV data used? | No (weight-mediated benefits only) | Yes (full CV outcomes) |
| Net price used | ~$8,412/yr (sema); ~$6,236/yr (tirz) | ~$6,829/yr (sema); ~$7,973/yr (tirz) |
| Population modeled | 126M US adults meeting trial criteria | US adults with obesity, lifetime horizon |
Even ICER's more favorable 2025 model shows cost-effectiveness, not cost-savings: the drug still costs more than it saves over a lifetime, but produces enough quality-adjusted life years to justify the excess cost at standard willingness-to-pay thresholds. Employer ROI analyses ask whether the plan recoups its spending in medical offsets over a 1–5 year budget window. QALY-based cost-effectiveness asks whether population health gains justify the spending over a lifetime. Both questions are valid; they have different answers because they measure different things.
Using SSR Health manufacturer discount data, Hwang et al. modeled a population of 126 million eligible US adults. To reach $100,000/QALY, tirzepatide would need to drop to about $4,334/year ($361/month), an additional 30.5% discount beyond current net prices. Semaglutide would need to drop to about $1,522/year ($127/month), an additional 81.9% discount. At Novo Nordisk's announced $675/month WAC effective January 2027, applying the ~41% obesity rebate yields an estimated employer net of roughly $398/month ($4,781/year), still above tirzepatide's threshold and far above semaglutide's.
Employer experience
Employer claims data
The KFF/Peterson Health System Tracker ran five employer focus groups in October 2025 covering 100+ companies employing 250,000+ people. No employer in the sample reported positive measured ROI on GLP-1 coverage for weight loss.
Employers recognize clinical effectiveness but remain unconvinced about cost offset. 59% of firms with 5,000+ workers reported higher-than-expected utilization, and BCBS Massachusetts spent $300M+ on GLP-1s before posting a $400M operating loss in 2024 and dropping weight-loss coverage for 2026. 34% of large employers now require lifestyle programs before GLP-1 approval, up from 10% in 2024.
No individual employer has publicly reported measured positive ROI on GLP-1 coverage for weight loss. The findings below are from PBMs, health plans, and consulting firms that measured across many employer clients.
| Organization | Type | Measurement approach | Finding | Net direction |
|---|---|---|---|---|
| Aon | Consulting firm (own workforce + client data) | Claims analysis, 192,000 users | Favorable medical cost trend by 18–30 months at ≥80% adherence; total cost of care still higher | Favorable trend |
| AssuredPartners | Broker (client data) | 5-year claims analysis | $560 annual medical savings vs $6,540 annual drug cost | Negative |
| Prime Therapeutics | PBM (16.5M lives) | Matched cohort | +$6,994 per user in year 1, +$4,206 in year 2 | Negative |
| UnitedHealthcare | Health plan | Internal analysis | 91% PMPM cost increase 12 months post-initiation | Negative |
| BCBS Massachusetts | Health plan | Internal | $300M+ GLP-1 spend; $400M operating loss in 2024; dropped weight-loss coverage for 2026 | Negative |
Sources: KFF/Peterson Health System Tracker, October 2025; KFF 2025 Employer Health Benefits Survey; WTW 2025 Best Practices Survey; Business Group on Health.
Employer survey data: 43% of firms with 5,000+ workers now cover GLP-1s for weight loss, up from 28% in 2024. 77% of large employers rate managing GLP-1 costs as "extremely or very important." 96% express concern about long-term cost implications.
Other side
Counterarguments
The case for GLP-1 coverage extends beyond direct medical cost offsets. These arguments are real, though harder to quantify.
Indirect costs and retention value
The total economic burden of obesity is estimated at $6,472 per employee with obesity per year ($2,427 presenteeism, $1,755 absenteeism, $1,514 medical costs, $664 disability, $112 workers' comp) in an Eli Lilly-funded analysis (GlobalData/PMC 2024). Novo Nordisk funded causal estimates of obesity-attributable productivity loss are substantially lower, at $271–$542 per employee per year (Cawley et al., JOEM 2021). No study has directly measured GLP-1-specific absenteeism or productivity outcomes.
Separately, vendor-sponsored surveys report employee interest in GLP-1 coverage: 29–30% of employees said they would switch jobs for access, and 78% of non-covering employers said they would cover if costs were lower. These surveys come from organizations with commercial interests in the answer (9amHealth, NFP), so the specific numbers warrant caution, though the directional signal is consistent.
Sources: GlobalData / PMC. "Assessing the Economic Impact of Obesity and Overweight on Employers." Nutrition & Diabetes. 2024. Eli Lilly-funded. Cawley et al., JOEM 2021. Novo Nordisk-funded. NFP 2026; 9amHealth 2026; WTW 2025.
Coverage parity and equity
The American Medical Association (November 2023) passed a resolution urging payers to ensure coverage parity for evidence-based obesity treatment. NEJM published arguments that excluding obesity medications undermines efforts to address inequities, given higher obesity prevalence among Black (49.9%) and Hispanic (45.6%) adults. Statins, insulin, and antihypertensives are covered without ROI proof; applying a different standard to obesity treatment is, in this framing, a matter of values rather than economics.
Pricing changes
Pricing developments
Recent announcements (Novo Nordisk's $675/month WAC for semaglutide starting January 2027, Eli Lilly's Employer Connect at $449/month, NovoCare's $349/month cash-pay, and the administration's $245/month Medicare price) represent 50–75% reductions from list prices. They move toward, but do not yet reach, break-even thresholds for most employer analyses. See the costs page for the full pricing table, effective dates, and net-price derivations.
Model it
ROI calculator
This calculator uses the evidence above to model pharmacy spend against medical cost offsets for your population. All inputs are pre-populated with published values and are fully adjustable. Presets draw from the cost-offset studies in this page (AssuredPartners, Prime Therapeutics, Novo Nordisk's post-2027 pricing). The calculator does not recommend a specific answer; it shows what the math looks like under the assumptions you choose.
GLP-1 ROI estimator
All outputs update in real time. Medical offsets apply only to members who persist on treatment; persistence decays across years based on published real-world data (Prime Therapeutics 2021 and 2024 cohorts).
Simplified model for illustration. Does not incorporate PBM administrative fees, member cost-sharing, indirect cost savings, or weight regain after discontinuation. Year 2–3 persistence rates for 2024 cohorts are extrapolated. Consult Milliman, Mercer, or your PBM's analytics team for plan-specific modeling.
Transparency
How this page was made
Selection criteria. We included actuarial reports, peer-reviewed cost-effectiveness analyses, claims analyses from brokers and PBMs, government reports (CBO, EBRI), and manufacturer-sponsored studies where flagged as such. Independent and manufacturer-funded analyses are shown in separate tables so readers can compare them directly. Marketing-style single-vendor projections were excluded.
Where the numbers come from. Every number in the tables is from the cited study or report. Where a study provided multiple figures (e.g., year 1 vs year 2 costs), we show what the study emphasized and link to the source. Pricing estimates for January 2027 post-WAC net costs (~$4,781/year Wegovy) are derived from Novo Nordisk's announced $675/month WAC times 12, times the ~41% obesity rebate observed in 2024–2025 claims data; these are estimates, not confirmed net prices.
What's not included. Productivity and retention ROI from GLP-1 coverage (no peer-reviewed direct measurement exists). Oral GLP-1 cost-offset modeling (Wegovy pill approved October 2025; no ROI data yet). Compounded GLP-1 telehealth economics (the compounding market contracted sharply in 2025 following FDA enforcement). Medicare Part D GLP-1 spend forecasts beyond 2035 (CBO and Hwang horizons end there).
Known limitations. First, most large employer-facing ROI analyses are done by organizations with commercial positions: Aon and AssuredPartners are consultants selling benefits services; Prime is a PBM; EBRI is funded by insurance and HR industry members. None are peer-reviewed in the traditional academic sense. Second, the 2024 cohort persistence improvement may reflect resolution of supply shortages rather than a durable change; 2–3 year data for this cohort does not yet exist. Third, pricing is moving quickly in 2026–2027 and any ROI estimate has a short shelf life at current prices.
Author and update cadence. Compiled by Ray Wu, MD/MBA — physician-founder working on metabolic health technology. Last updated April 21, 2026. Corrections welcome via the contact page.
References
Sources
Questions about this data? Corrections or updates?
Get in touch