Predicting GLP-1 Discontinuation From Pharmacy Claims

ABSTRACT

Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) have rapidly become high-visibility therapies for type 2 diabetes and obesity, with growing evidence of cardiovascular benefit. Yet multiple real-world studies show that adherence and persistence are modest: In obesity cohorts without diabetes, approximately one-third of members remain on therapy at 1 year, and among those with type 2 diabetes, nearly half discontinue by 12 months and approximately 70% by 24 months. This gap between trial efficacy and real-world use raises critical questions for payers, pharmacy benefit managers, and employers funding GLP-1 RA coverage.

This commentary proposes a pragmatic, claims-based framework for improving GLP-1 RA persistence. Routine pharmacy claims can be used to define new-start cohorts, calculate standard measures such as the proportion of days covered, and identify early warning signs, including delayed refills, prolonged use of starter doses, high out-of-pocket costs, and patterns of gastrointestinal adverse effects. These signals can drive simple rule-based alerts and, where appropriate, low-complexity predictive models to prioritize outreach.

Clinical pharmacists, nurse case managers, and digital tools can then translate these risk flags into concrete support for members—such as cost navigation, adverse effect management, expectation setting, and coordination with prescribers—while maintaining a “do no harm” stance. Finally, the commentary highlights the importance of monitoring GLP-1 RA persistence through an equity lens, ensuring that supportive interventions reduce, rather than widen, disparities in access and outcomes.

Takeaway Points

Managing glucagon-like peptide-1 receptor agonist (GLP-1 RA) therapies is now a core payer challenge. This commentary outlines a practical framework for using claims data to protect both outcomes and pharmacy spending.

• GLP-1 RA persistence in real-world practice is far lower than in trials, especially for patients with obesity without diabetes.
• Pharmacy and medical claims can be repurposed into a near–real-time “persistence radar” focused on the first 3 to 12 months of therapy.
• Simple rule-based alerts and basic predictive models are sufficient to target pharmacist outreach and digital nudges to members at the highest risk of dropout.
• Equity must be built in from the start so that supportive outreach reaches members facing the greatest structural and financial barriers.

Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) have moved from niche diabetes drugs to cornerstone therapies at the center of public attention. Findings of a large, randomized trial (NCT03574597) demonstrate that semaglutide (Wegovy) 2.4 mg reduces major adverse cardiovascular events in people with obesity and established cardiovascular disease, even in the absence of diabetes.¹

However, real-world data paint a more fragile picture of persistence. In a commercial cohort of adults using GLP-1 RAs for weight management without diabetes, approximately one-third remained on therapy at 1 year.² Among patients with type 2 diabetes, a US claims analysis found that 47.7% discontinued GLP-1 RA therapy by 12 months and 70.1% by 24 months; only 50.9% were adherent at 1 year.³

Payer reports echo these findings: Many members stop GLP-1 RAs before sustained weight loss and cardiometabolic benefits emerge.⁴ With 5-figure annual drug costs, early dropout undermines outcomes and the value case for coverage.⁵

The same pharmacy claims used for standard quality reporting can be leveraged to detect emerging GLP-1 RA nonadherence in close to real time and trigger targeted support. Pharmacy claims are scalable, but gaps do not always mean discontinuation. Apparent nonpersistence may reflect switching, supply disruptions, cash-pay fills, or clinically guided titration pauses; these should be treated as signals for supportive outreach rather than definitive discontinuation.

Why Members Stop: Drivers of GLP-1 RA Discontinuation

The reasons for discontinuation of GLP-1 RAs are multidimensional and vary by indication. Survey-based research involving injectable GLP-1 RAs in type 2 diabetes highlights gastrointestinal (GI) adverse effects, inadequate perceived glycemic response, and injection burden as key drivers of stopping therapy from both the patient and physician perspectives.⁶ These issues often surface early in treatment, particularly during dose titration.

In obesity cohorts, common reasons for stopping or switching the GLP-1 RAs semaglutide or tirzepatide (Mounjaro) include high out-of-pocket costs, coverage changes, adverse effects, perceived weight loss plateau, and shortages.⁷ Higher co-payments are associated with lower 1-year adherence to GLP-1 RA and sodium-glucose cotransporter 2 inhibitor therapies.⁸

Analyses also suggest that persistence is lower among individuals using GLP-1 RAs for obesity without diabetes than among those with type 2 diabetes, and that reinitiation after discontinuation is uncommon.^3,9 Together, these data imply that many members never experience the full cardiometabolic benefits that motivated coverage decisions.

Lower persistence in obesity cohorts may reflect less consistent coverage, greater out-of-pocket exposure, greater prior authorization friction, and interruptions driven by affordability or supply constraints. Differences in perceived clinical urgency, expectations for speed of benefit, and tolerance of adverse effects may also contribute to earlier stopping than seen in patients with type 2 diabetes.

A Claims-Based Early-Warning Framework

Most health plans already compute medication adherence metrics for chronic medications as part of quality reporting. Extending this infrastructure to GLP-1 RAs involves several steps.

Define the GLP-1 cohort. New starts can be identified using a 6- to 12-month washout in pharmacy claims. Indications can be approximated using International Statistical Classification of Diseases, Tenth Revision codes (eg, E11.x and E66.x) and benefits data.^1-3 Because drivers differ by indication, thresholds and outreach content should be tailored.

Track core metrics. For each new start, plans can calculate the following:

Proportion of days covered at 90 and 365 days, using 0.8 as a standard adherence cut point³

Persistence (claims-indicated), defined as time from initiation to a gap of 90 or more days without on-hand supply

Early refill patterns, including first-refill delay and any gaps greater than 7 to 14 days in the first 90 days

These basic measures provide an initial “persistence radar” focused on the first 3 to 12 months of therapy, when dropout risk is highest.²

Add clinical and cost context. Claims can be used to add clinically meaningful context, as follows:

Prolonged use of starter doses beyond label-recommended titration windows may indicate tolerability issues or inadequate follow-up.

New fills for antiemetics or GI agents soon after initiation can serve as markers of GI adverse effects.⁶

Out-of-pocket cost per fill or abrupt changes in co-pay or coinsurance may signal financial barriers that could trigger early discontinuation.^4,8

Linking members’ zip codes to neighborhood deprivation or social vulnerability indices can provide additional context on structural barriers and social risk.^2,4

Build simple flags and risk tiers. Simple rules (eg, missed first refill by ≥ 7 days, starter dose beyond 12 weeks, co-pay above a threshold) can identify members who may benefit from outreach. Plans with analytics capacity can add lightweight models using baseline adherence, care fragmentation proxies (prescribers/pharmacies), denial events, and socioeconomic indicators. The goal is to direct limited pharmacist and nurse time toward members at the highest risk of early discontinuation.

Translating Risk Flags Into Support

Risk scores matter only if they trigger action. For GLP-1 RAs, these 3 intervention types are most practical; outreach scripts and escalation should be tailored by indication (eg, affordability navigation for obesity vs diabetes care coordination):

Proactive pharmacy outreach. Clinical pharmacists or medication therapy management teams can contact flagged members and do the following:

Explore barriers to adherence and understand the member’s experience with adverse effects, expectations, and daily routines.^6,7

Offer practical strategies for managing GI symptoms, such as slower titration, dietary adjustments, and injection timing.⁶

Set realistic expectations for glycemic and weight trajectories, drawing on trial and real-world data to avoid disappointment-driven dropout.^1,2,7

Help members navigate prior authorization renewals, step-therapy requirements, and patient assistance programs when cost is the primary barrier.^4,8

Digital nudges. For digitally engaged members, app or portal notifications timed to refill windows and titration milestones may do the following:

Reinforce adherence behaviors with tailored
reminders.

Deliver brief, evidence-based education on cardiometabolic benefits and treatment timelines.^1,5

Prompt members to report adverse effects or concerns, triggering pharmacist follow-up when claims patterns suggest emerging problems.

Provider feedback loops. Deidentified dashboards of persistence and reinitiation by practice, drug, and indication can highlight where dropout is highest and where dosing or counseling may need refinement.^3,9 Framed as quality-improvement feedback, they can strengthen payer-clinician collaboration.

Keeping an Equity Lens

Emerging data suggest that GLP-1 RA use and persistence vary by race, ethnicity, and socioeconomic context, with members in more disadvantaged neighborhoods and those facing higher cost sharing more likely to discontinue.^2,4,8 These differences reflect structural barriers, not individual failings. An equity-aware claims strategy should therefore do the following:

Include neighborhood deprivation or social vulnerability indices as prioritization variables for supportive outreach, not as reasons to limit access.

Offer multilingual outreach and interpreter support during pharmacist or nurse calls.

Ensure that nondigital channels (phone, mail, community partners) are available to members who do not use apps or portals.

Monitor persistence stratified by race, ethnicity, language, and geography where these data are available and collected appropriately.

The aim is to prevent GLP-1 RAs from becoming another innovation that works best for members who already face the fewest barriers.

Governance and “Do No Harm”

Given the sensitivity around GLP-1 RA costs, adherence initiatives require guardrails: Risk flags should trigger support only, not coverage restriction or automatic step-downs. Plans should track persistence, total cost of care, and member experience/complaints. Safeguards include separating outreach flags from coverage workflows, prohibiting adverse actions based solely on flags, defaulting to assistance, auditing equity impact, and enforcing governance accountability.

Conclusions

GLP-1 RAs represent a convergence of strong clinical evidence, intense patient interest, and substantial payer investment.^1,5 Yet without deliberate attention to persistence, much of this potential will be lost to early discontinuation. Pharmacy and medical claims offer payers a ready-made platform to detect emerging nonadherence, target supportive outreach, and monitor equity in real time.

Plans that can demonstrate measurable improvements in GLP-1 RA persistence over the next few years will be better positioned with employers, regulators, and members who are rightly asking whether today’s spending is translating into tomorrow’s outcomes.

Acknowledgments

The author used OpenAI’s ChatGPT for language refinement and editorial support during manuscript preparation. The author reviewed and revised all content and takes full responsibility for the final manuscript.

Author Affiliation: Independent researcher and analyst.

Source of Funding: None.

Author Disclosures: The author reports no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design; analysis and interpretation of data; drafting of the manuscript; critical revision of the manuscript for important intellectual content; and administrative, technical, or logistic support.

Address Correspondence to: Aayush Sisodia, MSHI, BDS, independent researcher and analyst. Email: aayushsisodia19@gmail.com.

REFERENCES

1. Lincoff AM, Brown-Frandsen K, Colhoun HM, et al; SELECT Trial Investigators. Semaglutide and cardiovascular outcomes in obesity without diabetes. N Engl J Med. 2023;389(24):2221-2232. doi:10.1056/NEJMoa2307563

2. Gleason PP, Urick BY, Marshall LZ, Friedlander N, Qiu Y, Leslie RS. Real-world persistence and adherence to glucagon-like peptide-1 receptor agonists among obese commercially insured adults without diabetes. J Manag Care Spec Pharm. 2024;30(8):860-867. doi:10.18553/jmcp.2024.23332

3. Weiss T, Carr RD, Pal S, et al. Real-world adherence and discontinuation of glucagon-like peptide-1 receptor agonists therapy in type 2 diabetes mellitus patients in the United States. Patient Prefer Adherence. 2020;14:2337-2345. doi:10.2147/PPA.S277676

4. Real-world trends in GLP-1 treatment persistence and prescribing for weight management. Blue Health Intelligence. May 2024. Accessed June 24, 2026. https://www.bcbs.com/media/pdf/BHI_Issue_Brief_GLP1_Trends.pdf

5. Pearson SD, Whaley CM, Emond SK. Affordable access to GLP-1 obesity medications: strategies to guide market action and policy solutions in the US. J Comp Eff Res. 2025;14(9):e250083. doi:10.57264/cer-2025-0083

6. Sikirica MV, Martin AA, Wood R, Leith A, Piercy J, Higgins V. Reasons for discontinuation of GLP1 receptor agonists: data from a real-world cross-sectional survey of physicians and their patients with type 2 diabetes. Diabetes Metab Syndr Obes. 2017;10:403-412. doi:10.2147/DMSO.S141235

7. Gasoyan H, Butsch WS, Casacchia NJ, et al. Reasons for discontinuation of obesity pharmacotherapy with semaglutide or tirzepatide in clinical practice. Obesity (Silver Spring). 2025;33(12):2296-2303. doi:10.1002/oby.70058

8. Essien UR, Singh B, Swabe G, et al. Association of prescription co-payment with adherence to glucagon-like peptide-1 receptor agonist and sodium-glucose cotransporter-2 inhibitor therapies in patients with heart failure and diabetes. JAMA Netw Open. 2023;6(6):e2316290. doi:10.1001/jamanetworkopen.2023.16290

9. Rodriguez PJ, Zhang V, Gratzl S, et al. Discontinuation and reinitiation of dual-labeled GLP-1 receptor agonists among US adults with overweight or obesity. JAMA Netw Open. 2025;8(1):e2457349. doi:10.1001/jamanetworkopen.2024.57349