PATIENT EXPERIENCE WITH GLP-1 DOSE TITRATION AND ITS ASSOCIATION WITH TREATMENT ADHERENCE; REAL-WORLD INSIGHTS FROM THE ACCESSCMD REGISTRY
Author(s)
Serena Macleod, BSc, CertHE1, Olina Efthymiadou, MPharm, PhD1, Emma-Louise Shelton, MSc1, Mark Toshner, MD, PhD2.
1uMed, London, United Kingdom, 2Department of Medicine, University of Cambridge, Cambridge, UK, Cambridge, United Kingdom.
1uMed, London, United Kingdom, 2Department of Medicine, University of Cambridge, Cambridge, UK, Cambridge, United Kingdom.
OBJECTIVES: Despite the proven efficacy of GLP-1 therapies in clinical trials, evidence suggests that in real clinical practice discontinuation rates exceed 40% within the first year. Little evidence exists on GLP-1 users’ experience with dose titration and how it affects treatment adherence. We examined the association between patient-reported GLP-1 dose titration experience and treatment persistence.
METHODS: The AccessCMD registry, linking longitudinal electronic health record (EHR) data with patient-reported experience measures (PREMs) was used to identify and study adult GLP-1 users in the UK. Logistic regression explored the likelihood of treatment persistence (defined as EHR and/or self-reported current GLP-1 use) based on PREMs around titration experience (very easy to very hard), instructions clarity, side-effects and patient-reported support needs.
RESULTS: 1,099 AccessCMD participants completed the GLP-1 titration experience survey and 1,012 responses were eligible for analysis. At survey completion, 77.4% (n=783) participants remained on GLP-1 therapy and 22.6% (n=229) had discontinued. Most GLP-1 types impacted persistence (p<0.05), although diabetes-indicated tirzepatide demonstrated the strongest positive association with persistence (OR=7.91, p<0.001). Very easy titration and very clear titration instructions were associated with 67% (OR=1.67, p<0.05) and 73% (OR=1.73, p<0.05) higher odds of persistence respectively. Reporting instruction-related unmet support needs was also associated with higher odds of persistence (OR=2.35, p<0.05). Finally, private (vs. prescription/subsidised) GLP-1 access was associated with 78% lower odds of persistence (OR=0.22, p<0.001), followed by side-effects, with vomiting specifically associated with 57% lower odds of persistence (OR=0.43, p=0.006).
CONCLUSIONS: The strongest predictors of persistence were GLP-1 type and instructions clarity, whereas private access route and vomiting were the strongest discontinuation predictors. Instruction-related unmet support needs were positively associated with persistence, potentially reflecting treatment engagement among current users, and reinforcing the importance of instructions clarity. Linking EHR with PREMs allowed for identification and detailed characterisation of the GLP-1 users’ treatment journey and respective support needs.
METHODS: The AccessCMD registry, linking longitudinal electronic health record (EHR) data with patient-reported experience measures (PREMs) was used to identify and study adult GLP-1 users in the UK. Logistic regression explored the likelihood of treatment persistence (defined as EHR and/or self-reported current GLP-1 use) based on PREMs around titration experience (very easy to very hard), instructions clarity, side-effects and patient-reported support needs.
RESULTS: 1,099 AccessCMD participants completed the GLP-1 titration experience survey and 1,012 responses were eligible for analysis. At survey completion, 77.4% (n=783) participants remained on GLP-1 therapy and 22.6% (n=229) had discontinued. Most GLP-1 types impacted persistence (p<0.05), although diabetes-indicated tirzepatide demonstrated the strongest positive association with persistence (OR=7.91, p<0.001). Very easy titration and very clear titration instructions were associated with 67% (OR=1.67, p<0.05) and 73% (OR=1.73, p<0.05) higher odds of persistence respectively. Reporting instruction-related unmet support needs was also associated with higher odds of persistence (OR=2.35, p<0.05). Finally, private (vs. prescription/subsidised) GLP-1 access was associated with 78% lower odds of persistence (OR=0.22, p<0.001), followed by side-effects, with vomiting specifically associated with 57% lower odds of persistence (OR=0.43, p=0.006).
CONCLUSIONS: The strongest predictors of persistence were GLP-1 type and instructions clarity, whereas private access route and vomiting were the strongest discontinuation predictors. Instruction-related unmet support needs were positively associated with persistence, potentially reflecting treatment engagement among current users, and reinforcing the importance of instructions clarity. Linking EHR with PREMs allowed for identification and detailed characterisation of the GLP-1 users’ treatment journey and respective support needs.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
PCR164
Topic
Clinical Outcomes, Patient-Centered Research, Real World Data & Information Systems
Topic Subcategory
Adherence, Persistence, & Compliance
Disease
Cardiovascular Disorders (including MI, Stroke, Circulatory), Diabetes/Endocrine/Metabolic Disorders (including obesity), Nutrition