Vivian Fonseca, MD, is professor of medicine, assistant dean for clinical research, and the Tullis-Tulane Alumni Chair in Diabetes at Tulane University School of Medicine.

Vivian Fonseca, MD

Improving GLP-1 Adherence in Patients With Diabetes, With Vivian Fonseca, MD

Despite a steady flow of more and more efficacious GLP-1 RA treatments for obesity and diabetes, adherence rates remain low across multiple studies.

Prior research has cited a variety of reasons for discontinuation, including high cost and the treatment burden of many injection-based GLP-1 RAs. Additionally, the drugs are famously associated with various gastrointestinal effects including nausea, vomiting, and diarrhea.

Presented at the American Diabetes Association (ADA) Scientific Sessions 2026 in New Orleans, Louisiana, by Leigh Perreault, MD, adjoint professor of medicine at the University of Colorado Anschutz, the present analysis worked to distill clear reasons for significantly lower adherence and persistence.

Vivian Fonseca, MD, professor of medicine, assistant dean for clinical research, the Tullis-Tulane Alumni Chair in Diabetes, and the chief of the Section of Endocrinology at Tulane University Medical Center in New Orleans, was affiliated with the study, and spoke with the 

 editorial team to discuss the study and its broader implications for the groundbreaking weight loss treatment.

“A well-known example of this is statins – when they first came out, they were revolutionary in preventing cardiovascular disease,” Fonseca told 

 in an exclusive interview. “While adherence used to not be particularly good, it’s improved tremendously in the last few years as it’s become more accepted by the medical community and the general public. However, there’s still a certain amount of non-adherence to care.”

The analysis collected a series of observational studies from 2014-2025 reporting persistence and/or adherence to GLP-1s in patients with type 2 diabetes (T2D). Persistence was defined as treatment gap thresholds from 30-90 days, while adherence was established as a proportion of days covered ≥80%. Among the investigated GLP-1s were tirzepatide, semaglutide, dulaglutide, and liraglutide. Additionally, the team utilized meta-analyses using restricted maximum likelihood random effects modeling to estimate overall pooled proportions and 95% confidence intervals.

A total of 28 studies were ultimately included in the trial, totaling 267,542 patients. Of these, 69% were persistent at 6 months (95% CI, 65-73%), and 49% were persistent at 12 months (95% CI, 43-55%). By 12 months, dulaglutide demonstrated the highest persistence rates (56%; 95% CI, 55-56%), followed by semaglutide (53%; 95% CI, 44-62%) and liraglutide (43%; 95% CI, 30-55%). Additionally, 50% of patients were adherent at 6 months (95% CI, 45-55%) and 42% were adherent at 12 months (95% CI, 37-47%). Dulaglutide also showed the highest adherence of included GLP-1 therapies (50%; 95% CI, 43-56%), followed by semaglutide (43%; 95% CI, 41-44%) and liraglutide (34%; 95% CI, 29-39%).

Ultimately, Perreault and colleagues concluded that GLP-1 persistence and adherence outside of clinical trials remained consistently low across the included studies. They note a substantial unmet need for strategies to promote sustained GLP-1 use to capitalize on the drugs’ efficacy in improving patient outcomes. Fonseca also provided advice for clinicians, highlighting the best way to address concerns about lifelong treatment use and potential workarounds for gastrointestinal adverse events.

“I think you need to talk to the patient about the advantages of long-term endurance. Tell them that this is for life,” Fonseca said. “You have obesity, which is a lifelong disease. It’s not going away because you lost the weight for a few months; you will regain the weight if you stop the therapy. But you also need to discuss tolerability issues, warn them about side effects, and say that there are ways to adhere to it.”

Editors’ Note: Fonseca reports disclosures with Novo Nordisk, Regeneron, Abbott, Corcept, Eli Lilly, and others.

Perreault L, Rasouli N, Sanogo F, et al. A Meta-Analysis of Persistence and Adherence to Glucagon-Like Peptide 1-Based Treatments among Patients with Type 2 Diabetes in the United States. Abstract presented at the American Diabetes Association (ADA) Scientific Sessions 2026, New Orleans, LA. June 5-8, 2026.

Heisey HD, Gregg LP, Verrico CD, Villareal DT, Fletcher TL. Rates of, Reasons for, and Reactions to Discontinuation of GLP-1 Receptor Agonists: A Narrative Review. Diabetes Obes Metab. Published online June 5, 2026. 

Videos

Fonseca discusses substantially lower persistence and adherence with GLP-1s in real-world practice, highlighting methods for clinicians to address this.

GLP-1 Discontinuation in Type 2 Diabetes: What Happens When Patients Stop?

A real-world analysis of more than 18,000 patients with 

 found nearly half discontinued glucagon-like peptide-1 receptor agonist (GLP-1 RA) therapy within 1 year, with discontinuation independently associated with substantially lower odds of achieving guideline-recommended weight loss and glycemic targets.¹

American Diabetes Association (ADA) 2026 Scientific Sessions

, quantify what many clinicians observe in practice: GLP-1 RA persistence is a major determinant of whether patients realize the metabolic benefits these agents can provide. Current guidelines recognize ≥ 5% weight reduction as clinically meaningful and an HbA1c < 7% as a standard glycemic target in T2D.

The analysis drew from the Market Clarity claims-EHR linked database, identifying 18,187 adults with T2D, BMI of ≥ 25 kg/m², and baseline HbA1c of ≥ 7% who initiated a GLP-1 RA between January 2021, and June 2024.¹ All patients had BMI and HbA1c measurements within 90 days before initiation and a follow-up measure at 365 ± 90 days after initiation.

Discontinuation was defined as a gap of ≥ 60 days without any GLP-1 supply. Multivariable regression adjusted for clinical and demographic characteristics estimated the association between discontinuation and achievement of BMI and HbA1c goals over 1 year.

Over the follow-up period, 8551 patients (47.0%) discontinued GLP-1 RA therapy.¹ Among those who discontinued versus those who persisted, 33.0% versus 55.0% achieved ≥ 5% BMI reduction, 12.1% versus 26.9% achieved ≥ 10% BMI reduction, and 31.2% versus 60.9% reached an HbA1c < 7%, respectively.

After multivariable adjustment, patients who discontinued were 41% less likely to achieve ≥ 5% BMI reduction (risk ratio [RR], 0.59; 95% CI, 0.57-0.62), 56% less likely to achieve ≥ 10% BMI reduction (RR, 0.44; 95% CI, 0.41-0.47), and 47% less likely to achieve an HbA1c < 7% (RR, 0.53; 95% CI, 0.51-0.55).¹

The analysis also identified a gradient effect among those who did not discontinue. Approximately 10% of nominally persistent patients had prescription fill rates at or below 80%, reflecting suboptimal adherence without meeting the 60-day gap threshold for discontinuation. Even within the persistence group, less consistent use was associated with attenuated outcomes compared with patients filling prescriptions regularly, reinforcing dose consistency as a meaningful variable alongside persistence itself.

Vivian Fonseca, MD, Tullis Tulane Alumni Chair in Diabetes at Tulane University School of Medicine and an investigator on the analysis, told 

 the discontinuation rates observed are consistent with what he encounters in clinical practice. He cited tolerability, cost, injection burden, and patient expectations as primary drivers. He noted the pattern differs from other chronic disease contexts: patients initiating GLP-1 RAs for weight loss often do not anticipate lifelong therapy, unlike those starting antihypertensives or statins.

"GLP-1 is very good, not just for weight, but also for treating diabetes," Fonseca said. "If they stop, [glucose control] will deteriorate unless they're moving over to some other medication."

Fonseca emphasized expectation-setting as a clinical tool for reducing early discontinuation. He noted a subset of patients achieve little to no weight loss on GLP-1 RAs and may require alternative approaches, making early identification of non-responders a practical priority. He also highlighted the need for studies on weight maintenance strategies, suggesting aggressive titration to a maximum tolerated dose may not be necessary indefinitely and more nuanced long-term regimens could support sustained adherence.

Editors’ note: Fonseca reports relevant disclosures with Novo Nordisk, Regeneron, Abbott, Corcept, Eli Lilly, and others.

Fonseca V, Girguis M, Donsmark M, et al. Impact of GLP-1–based treatment discontinuation on weight loss and glycemic goals among patients with type 2 diabetes: quantifying an opportunity to improve treatment benefits. Presented at: American Diabetes Association 2026 Scientific Sessions; June 2026.

American Diabetes Association. Standards of care in diabetes. 

https://professional.diabetes.org/standards-of-care

Vivian Fonseca, MD, explains data showing nearly half of patients with T2D discontinued GLP-1 therapy within 1 year, reducing their likelihood of achieving weight loss and HbA1c goals.

GLP-1 Discontinuation Cuts Chances of Meeting Weight Loss, Glycemic Goals in T2D

CAPTAIN-T2D: New Trial Investigates Treating Elevated Cortisol in T2D, With Vivian Fonseca, MD

The phase 2 Clofutriben And Placebo Against INtractable Type 2 Diabetes (CAPTAIN-T2D) trial, investigating clofutriben to improve glycemic control and metabolic health in patients with difficult-to-control 

 and elevated cortisol, has officially enrolled its first patient.

CAPTAIN-T2D directly follows positive results from the CATALYST trial in 2025, which saw 136 patients with T2D and hypercortisolism randomly assigned in a 2:1 ratio to either mifepristone 300-900 mg once daily (n = 91) or placebo (n = 45) for 24 weeks. The primary endpoint was the change in HbA1c, which worked out to a -1.32% difference compared to placebo. Additionally, patients receiving mifepristone saw reductions in body weight and waist circumference.

“So, what we want to do now in CAPTAIN-T2D is to use another approach at treating this elevated cortisol,” Vivian Fonseca, MD, professor of medicine, assistant dean for clinical research, and the Tullis-Tulane Alumni Chair in Diabetes at Tulane University School of Medicine, told 

 in an exclusive interview. “In this instance, we’re using an 11-beta hydroxysteroid dehydrogenase inhibitor, which uses a completely different approach from mifepristone. You’re addressing intracellular cortisol by blocking this enzyme that leads to conversion of the inactive cortisone to cortisol, and so you’re decreasing cortisol activity in the tissues.”

CAPTAIN-T2D is a 2-part trial – the first part will see patients with poorly controlled T2D despite being on multiple medications screened for elevated cortisol using an overnight dexamethasone suppression test. This period will last between 5 and 9 weeks for most participants.

Eligible patients will then be randomly assigned to either clofutriben 0.2 mg, 2 mg, 6 mg, or 12 mg, or a placebo control, all of which will be administered daily. This second part will last for 24 weeks, with a follow-up phone call 4 weeks after the final dose.

To be eligible for inclusion, patients must be ≥18 years old with an HbA1c ≥7.5% documented within 3 months prior to screening. Additionally, patients require ≥1 of the following:

Diabetes complications – retinopathy, nephropathy, neuropathy, 

Hypertension needing ≥2 adequately dosed AHMs

Adequately dosed basal or basal plus prandial insulin on top of ≥1 other ADM

Adequately dosed incretin agonists on top of ≥1 other ADM

Evidence or history of osteoporosis or non-traumatic fracture

Established diagnosis of a neoplastic source of hypercortisolism and have failed, are ineligible for, or declined surgery.

CAPTAIN-T2D has 2 primary endpoints: the first is the percentage of patients with both morning serum cortisol >1.8 mg/dL and morning plasma dexamethasone >-140 ng/dL after a single dexamethasone 1 mg dose taken the night prior. The second endpoint involves glycated hemoglobin A1c percent change from baseline to week 24 by treatment. Secondary endpoints include fasting plasma glucose change from baseline and the frequency of treatment-emergent adverse events.

The trial is expected to enroll a total of 1500 patients and is planned to end in June of 2028.

“We’ve begun to use this method a lot in other conditions, such as cancer: you find a particular defect, and you treat it with therapy that’s specifically targeted to that defect,” Fonseca said. “Unfortunately, we don’t have much of that in diabetes. We’ve been using a one-glove-fits-all approach, giving everybody every new medication that comes along, and the success has been mixed. Overall, control is better than it was 20 years ago, but it’s still not right, and I think identifying abnormalities could help with better outcomes.”

Editor’s Note: Fonseca reports disclosures with Abbott, AstraZeneca, Novo Nordisk, Sanofi-Aventis, Eli Lilly, Bayer, and others.

Sparrow Pharmaceuticals. Sparrow Pharmaceuticals Announces First Patient Enrolled in Phase 2b Clinical Trial Evaluation Clofutriben for Difficult-to-Control Type 2 Diabetes. 

. January 6, 2026. Accessed January 23, 2026. 

https://www.biospace.com/press-releases/sparrow-pharmaceuticals-announces-first-patient-enrolled-in-phase-2b-clinical-trial-evaluating-clofutriben-for-difficult-to-control-type-2-diabetes

DeFronzo RA, Fonseca V, Aroda VR, et al. Inadequately controlled type 2 diabetes and hypercortisolism: Improved glycemia with mifepristone treatment. 

Sparrow Pharmaceuticals. Trial Against INtractable Type 2 Diabetes (CAPTAIN-T2D) (CAPTAIN-T2D). ClinicalTrials.gov Identifier: NCT07296484. Updated January 22, 2026. Accessed January 23, 2026. 

https://clinicaltrials.gov/study/NCT07296484?term=NCT07296484&rank=1

Fonseca explains the structure and goals of the ongoing trial, which has enrolled its first patient to clofutriben treatment against hard-to-control T2D.

Panelists discuss how improving adherence requires both individual strategies like smartphone reminders and smart medication monitoring devices, as well as population-level interventions including reducing sodium content in processed foods and harmonizing hypertension guidelines to address the epidemic of poorly controlled blood pressure.

Optimizing Patient Adherence and Targeting the Root Causes of Treatment Resistance 

Video content above is prompted by the following: 

Modern adherence improvement strategies leverage smartphone technology and digital health tools to address the fundamental challenge of medication-taking persistence. Smart medication caps, phone alarms, and remote monitoring systems provide objective data about patient medication-taking behavior while enabling more targeted interventions. These technological solutions must be implemented thoughtfully, recognizing that opening a pill bottle doesn’t guarantee medication consumption, but they provide valuable conversation starters between patients and providers.

Addressing resistant hypertension requires population health approaches targeting root causes, particularly dietary sodium reduction through food industry regulation. International examples from the UK demonstrate that reducing sodium content in processed foods can achieve meaningful population-level blood pressure improvements. Public health initiatives must balance individual choice with protective policies, similar to approaches used for clean water and food safety, recognizing that personal responsibility alone cannot address the epidemic of poor blood pressure control.

The future of hypertension management lies in addressing underlying causes rather than simply adding medications indefinitely. This includes governmental leadership in food policy, health care system redesign for chronic disease management, and individual education about lifestyle modifications. The goal is creating environments where patients who want to maintain good health can do so without being undermined by unhealthy food systems, while preserving individual choice and autonomy in health care decisions.

Vivian Fonseca, MD

Articles