Background
A shortage of medicines containing Glucagon-Like Peptide-1 receptor agonists (GLP-1 RA) has been affecting European Union (EU) Member States since 2022. The medicines belonging to the class of GLP-1 RA are either authorised for the treatment of diabetes or authorised for weight management in patients diagnosed with obesity, with the exception of Mounjaro (tirzepatide), a glucose-dependent insulinotropic polypeptide (GIP) and GLP-1 RA that is authorised for both indications.
Increased demand for these medicines has contributed to this shortage in addition to other causes, e.g., capacity constraints. Excessive off-label use for cosmetic weight loss of some of these medicines has raised concerns
This study aims to provide an overview of the characteristics of patients prescribed with GLP-1 RA and to describe the pattern of use, including switching between GLP-1 RA substances and to other antidiabetics, as well as switching between selected brands.
Objectives
This study is a routinely repeated study of a previous DARWIN EU® study (P3-C1-008, DARWIN EU® – Drug Utilisation Study on GLP-1 Receptor Agonists), with substantial adaptations described in section 5. Amendments and updates to answer the research question “What are the patterns of use of GLP-1 RA in 2015 – 2025?”.
The study will comprise the following objectives:
Methods
Study design
Cohort study
Population
The study population will include all individuals present in the database during the study period: 01/01/2015 up until the latest data available for each data partner.
Variables
Exposure: Overall use of GLP-1 RA, by ingredient (exenatide, liraglutide, lixisenatide, dulaglutide, semaglutide, and tirzepatide), and by pre-specified brands (Wegovy, Ozempic, Rybelsus, Trulicity, Victoza, Saxenda, and Mounjaro).
Note: Data partners without brand information will be only perform analyses at the ingredient level.
Relevant covariates: For stratification: overall, age, sex, indication cohorts (type 2 diabetes mellitus /obesity), prescriber speciality (where available), calendar month [incidence/prevalence], calendar year.
Statistical analysis
Population-level drug utilisation: Monthly incidence rates and monthly point prevalence estimates, both with 95% confidence interval.
Patient-level drug utilisation: New users of GLP-1 RA will be characterised by the relevant covariates listed above for the overall study period and by calendar year. The characterisation will be calculated for overall GLP-1 RA use, by ingredient, and by pre-specified brand.
Patient-level characterisation of GLP-1 RA switching of substances and of GLP-1 RA within-substance switching of strength [i.e., Objectives 3-6]: Switching will be reported using Sankey diagram. Time-to-switch will be based on survival analysis. We will report median time-to-switch,
(based on median survival: time where 50% of the patients in the group had switch to another treatment/strength), average time-on-treatment (based on restricted mean survival: average time patients are alive and still had not changed the treatment/strength over the whole follow-up period [10 years]), and Kaplan-Meier plots. Analysis will be reported for the overall study period [Objectives 3 and 4]; and between 2015–2020, and annually between 2021 and 2025 [Objectives 5 and 6]. Objectives 5 and 6 will be carried out subject to feasibility of execution.
A minimum count of 5 will be applied when reporting results, with any smaller values reported as ““<5”.