Use and Spending Trends of GLP-1 and Dual GLP-1/GIP Medicines in Croatia from 2017 to 2024

Glucagon-like peptide-1 (GLP-1) receptor agonists and the dual GLP-1/GIP agonist have reshaped the therapeutic management of type 2 diabetes by influencing international treatment guidelines and prescribing patterns, rather than influencing patient-level outcomes directly within the scope of this study. GLP-1 receptor agonists and, more recently, the dual GLP-1/GIP agonist have demonstrated robust glucose-lowering efficacy and cardiovascular benefits, leading to rapid global uptake [1,2]. Parallel growth in sodium–glucose cotransporter 2 (SGLT-2) inhibitors has further diversified the therapeutic options [3,4]. Although these innovations improve clinical outcomes, their high acquisition costs and expanding indications have created a new axis of pharmaceutical expenditure for antihyperglycemic care [5].

Systematic pharmacoepidemiologic evidence from smaller European Union health systems with constrained pharmaceutical budgets remains limited. National drug-utilization studies are essential to understand how innovation translates into population-level prescribing and to anticipate the budgetary implications for public health insurers. Croatia provides an informative setting: it maintains universal health coverage and centralized drug procurement, while the Croatian Agency for Medicinal Products and Medical Devices (HALMED) publish annual national consumption data based on the World Health Organization's Anatomical Therapeutic Chemical and Defined Daily Dose (ATC/DDD) methodology. These reports enable a reproducible measurement of drug utilization in DDD/1000 inhabitants/day and expenditure in euros, supporting surveillance of therapeutic trends and stewardship of health system resources [6,7].

For policymakers, HALMED's national ATC/DDD data are directly relevant to health system stewardship because they provide the only complete, routinely collected source of standardized information on the diffusion, intensity and budget impact of antihyperglycemic therapies across all care settings. In a system with centralized procurement and a single dominant public payer, these data enable monitoring of how rapidly high-cost agents enter clinical practice, whether prescribing patterns align with guideline recommendations, and how therapeutic shifts translate into fiscal pressure on the pharmaceutical budget. Such surveillance is essential for anticipating reimbursement needs, informing price negotiations and ensuring that resource allocation keeps pace with evolving clinical evidence and population demand.

Comparative European data demonstrate marked heterogeneity in the uptake of novel antihyperglycemic therapies. Analyses of 11 EU countries show that high-income systems such as Austria, Germany, Sweden and the United Kingdom achieved substantially higher volume market shares and annual therapy costs for GLP-1 receptor agonists and other new antidiabetic medicines than Central and Eastern European (CEE) countries, including Croatia, over the first decade after market entry [8]. More recent multinational and global drug-utilization studies confirm that the use of GLP-1 RAs and SGLT-2 inhibitors remains concentrated in high-income health systems, whereas most CEE countries report lower per capita consumption, despite rising trends [5,9,10]. These differences in uptake and budget impact underscore the value of country-level analyses, such as the present study, for positioning Croatia within broader European diffusion patterns.

Previous HALMED bulletins have documented overall growth in antihyperglycemic drug consumption but without analytical interpretation by therapeutic subclass [6]. No publication has quantified the diffusion of GLP-1 receptor agonists or the dual GLP-1/GIP agonist in Croatia or assessed their contribution to total antihyperglycemic drug spending. Such evidence is increasingly relevant for health system sustainability and for comparative assessments across European Union member states.

In Croatia, the dual GLP-1/GIP agonist (ATC A10BX16) was not marketed before 2024; therefore, utilization and expenditure trends for 2017–2023 reflect GLP-1 receptor agonists only, with dual agonist entry captured only in the final study year.

To our knowledge, this is the first peer-reviewed analysis to quantify the uptake and budget impact of GLP-1 receptor agonists and the dual GLP-1/GIP agonist in Croatia. It also provides some of the earliest national-level data on the real-world entry of the dual GLP-1/GIP agonist in Central and Eastern Europe.

Within the broader antihyperglycemic armamentarium, GLP-1 receptor agonists (A10BJ) and the dual GLP-1/GIP agonist (A10BX16) are of particular interest for health system stewardship. They combine high acquisition costs with rapidly expanding indications across type 2 diabetes, obesity and metabolic dysfunction-associated steatotic liver disease, and they have become central to contemporary cardiovascular and renal risk-reduction strategies. These agents also account for a disproportionate share of recent growth in diabetes-drug expenditure in Croatia, whereas older incretin-related agents such as DPP-4 inhibitors (A10BH) are largely generic, associated with more modest incremental clinical benefit and did not materially drive the observed expenditure increase. For these reasons, we focused our detailed analyses on GLP-1 receptor agonists and the dual GLP-1/GIP agonist, while retaining DPP-4 inhibitors within the residual "other oral A10B" category.

The present study therefore aimed to (i) describe national utilization and expenditure trends for antihyperglycemic drugs (ATC A10) from 2017 through to 2024; (ii) quantify growth in GLP-1 receptor agonists (A10BJ) and the dual GLP-1/GIP agonist (A10BX16), relative to insulins and their analogs (A10A) and SGLT-2 inhibitors (A10BK); and (iii) estimate their cumulative budget impact within total antihyperglycemic drug expenditure.

This national pharmacoepidemiologic analysis shows a marked transformation in the utilization and expenditure of antihyperglycemic drugs in Croatia from 2017 through to 2024. Over these eight years, total antihyperglycemic drug spending increased by nearly 78%, with most of the expansion being attributable to GLP-1 receptor agonists and dual GLP-1/GIP agonist. These classes accounted for roughly one-third of total A10 expenditure in 2024, despite representing a modest share of DDD consumption. The findings reflect a structural shift in Croatia's antihyperglycemic pharmacotherapy market, paralleling international patterns of rapid uptake of high-cost, high-impact therapies. In this context, a "structural shift" refers to a measurable re-composition of the national A10 market from long-established insulins and low-cost oral agents towards newer, high-cost GLP-1 RA, the dual GLP-1/GIP agonist and SGLT-2 therapies. Between 2017 and 2024, the combined expenditure share of GLP-1 receptor agonists and the dual GLP-1/GIP agonist increased from 0% to 29.8%, while SGLT-2 inhibitors rose to 15.6% of total A10 expenditure; over the same period, the residual oral A10B group declined from more than half of total spending to 31.7%, and insulin expenditure remained broadly stable at around one quarter of the total (Table 1, Figure 3). These changes indicate that recent growth in Croatia's diabetes-drug budget is driven predominantly by the rapid diffusion of high-cost incretin-based and SGLT-2 agents, rather than by proportional increases in older oral therapies or insulins. Our decision to highlight GLP-1 receptor agonists and the dual GLP-1/GIP agonist reflects their role as the principal recent drivers of A10 expenditure and policy concern, in contrast to DPP-4 inhibitors, which remain embedded in the background of older oral A10B therapies with comparatively limited marginal budget impact.

The A10BX16 subclass first appears in the Croatian national utilization data in 2024, corresponding to the initial market entry of the dual GLP-1/GIP agonist. Although the agent had received European regulatory approval earlier, it only became available in Croatia late in the study window and without reimbursement, requiring out-of-pocket payment. Consequently, the recorded utilization in 2024 was minimal (<0.05 DDD/1000/day) and its contribution to overall expenditure was negligible. These patterns are consistent with the expected early-adoption dynamics of a high-cost therapy introduced outside the public insurance system.

Croatia's rapid uptake of GLP-1 receptor agonists and the dual GLP-1/GIP agonist mirrors Western European trends, despite its smaller population, budget constraints and later market entry, with GLP-1 RAs and SGLT-2 inhibitors expanding sharply in both volume and expenditure. This growth is driven by broadening clinical indications, following cardiovascular and renal-outcome evidence [12,13,14,15]. The observed pattern is consistent with predominantly additive growth in antihyperglycemic drug spending, rather than wholesale substitution of older agents. Total A10 expenditure increased from EUR 54.2 million to EUR 96.5 million between 2017 and 2024, while insulin utilization remained broadly stable, the residual oral A10B group retained a substantial volume share, and GLP-1 receptor agonists and SGLT-2 inhibitors expanded rapidly in both DDDs and budget share (Table 1 and Table 2, Figure 2 and Figure 3). These aggregate trends suggest that novel agents have largely been layered onto, rather than fully replacing existing therapies, although patient-level data would be required to distinguish add-on use from switching with certainty. If recent trajectories persist, the budget share of GLP-1 receptor agonists and dual GLP-1/GIP agonist is likely to increase further, potentially approaching one half of total A10 spending over the medium term; however, this depends on uptake, indication expansion and pricing dynamics. Recent national surveys of Croatian general practitioners confirm that GLP-1 receptor agonists and the dual GLP-1/GIP agonist remain underutilized in primary care: GLP-1 receptor agonists are prescribed to only about 14% of patients and SGLT-2 inhibitors to 21%, indicating that many eligible individuals still do not receive these cardioprotective treatments [16,17].

These patterns underscore the need for real-time monitoring to guide pricing and reimbursement policy, especially considering these drugs are emerging as a promising option for the obesity and metabolic dysfunction-associated steatotic liver disease (MASLD) with fibrosis, which might substantially consume the healthcare budget [18,19,20,21]. Building on this expanding therapeutic scope, growing evidence demonstrates that dual GIP/GLP-1 receptor agonists achieve substantial and sustained weight loss in non-diabetic obese individuals, suggesting that clinical demand for this class will continue to rise and further intensify budgetary pressures [22]. While the study benefits from complete national data and standardized metrics, its aggregate design limits inference because HALMED data lack patient-level detail, differentiate neither care settings nor true reimbursed costs and rely on DDD units that may not reflect real-world dosing. As a result, the analytical depth and clinical interpretability of our findings are necessarily restricted to system-level patterns in drug utilization and expenditure, rather than individual treatment effects or outcomes.

The acceleration in the use of GLP-1 receptor agonists and the dual GLP-1/GIP agonist use observed after 2021 likely reflects the impact of international guideline updates and expanded indications. In particular, the 2018 ADA–EASD consensus report and subsequent updates recommended GLP-1 receptor agonists and SGLT-2 inhibitors for patients with established cardiovascular disease or high-risk profiles, regardless of HbA1c levels. The European Medicines Agency (EMA) approvals for new cardiovascular and renal indications between 2019 and 2022 also expanded their eligibility. These developments likely influenced national prescribing behavior, even in the absence of formal changes to Croatian reimbursement criteria. Reimbursement of once-weekly GLP-1 RAs and broader formulary access may also have facilitated uptake in later years. Despite these gains, real-world implementation in primary care remains incomplete.

Although international consensus statements and cardiovascular-outcome evidence clearly shaped therapeutic expectations in Croatia, national professional guidance also played a role. The Croatian Society for Diabetes and Metabolic Disorders of the Croatian Medical Association has, since 2016, regularly issued expert recommendations on the use of oral and injectable antihyperglycemic agents, which were disseminated at national diabetes congresses and communicated to the Croatian Health Insurance Fund (HZZO) for consideration in reimbursement decisions. These recommendations have guided clinical decision-making in practice, but they were not published as publicly available formal documents and therefore cannot be cited directly. A new consolidated national guideline is currently in final preparation and is expected to be published in early 2026. Consequently, the observed utilization trends during 2017–2024 reflect the combined influence of evolving international evidence and locally disseminated expert society recommendations, even though no formally updated national guideline was issued during the study window.

Comparable trends have been documented in neighboring and regional health systems, with Western and Northern European countries generally reporting an earlier and steeper uptake of GLP-1 RAs and SGLT-2 inhibitors than CEE countries [5,8,10]. Within CEE, Hungary and others show dynamic growth of novel antidiabetic drugs but persistently substantial use of older oral agents, reflecting the structural and budgetary constraints [5,10]. Against this backdrop, Croatia's 2017–2024 trajectory—where GLP-1 receptor agonists and the dual GLP-1/GIP agonist already account for nearly one-third of A10 expenditure—suggests convergence towards higher-uptake EU systems, despite a later market entry. Croatia's trajectory therefore aligns closely with regional patterns despite the later market entry and a more constrained pharmaceutical budget. This convergence suggests that evidence from cardiovascular and renal outcome trials exerted similar prescribing pressure across Central Europe, overriding structural differences in the procurement and reimbursement systems.

In a universal coverage system with a single dominant public payer, rapid uptake of high-cost therapies has direct implications for pharmaceutical budget sustainability. The shift toward GLP-1 RAs and SGLT-2 inhibitors—combined with expanding cardiovascular-, renal- and obesity-related indications—creates upward expenditure pressure that may require future adjustments in pricing negotiations, reimbursement criteria or therapeutic guidelines. The Croatian system's constrained fiscal envelope means that continued diffusion of GLP-1 receptor agonists and the dual GLP-1/GIP agonist must be balanced against competing health system priorities.

This study provides timely, actionable evidence for national payers and policymakers navigating the budgetary implications of high-cost therapies with expanding indications. As such, it offers a case study for similar smaller European health systems with limited fiscal space for pharmaceuticals that are managing therapeutic innovation under resource constraints.

Several limitations must be acknowledged. HALMED reports provide aggregate national data that do not distinguish patient characteristics, diabetes duration, clinical indications, or treatment lines. Expenditure figures reflect the published prices and may not capture confidential rebates or negotiated discounts. DDD values do not necessarily correspond to real-world dosing regimens for agents such as GLP-1 receptor agonists, potentially over- or underestimating true consumption. Hospital and outpatient channels are aggregated, preventing sector-specific analysis. Finally, this study does not adjust for changes in diabetes prevalence or therapeutic guidelines across the observation period. Despite these constraints, the complete national coverage and the standardized ATC/DDD methodology ensure strong internal consistency and comparability across years.

Temporal changes in pharmaceutical policy may have influenced the observed trends but could not be disentangled in this analysis. HALMED reports capture the final aggregate effect of prescribing behavior, reimbursement criteria, reference pricing and tender outcomes on national consumption and list-price expenditure; however, they do not include explicit information on when such policy changes occurred or how strongly they affected specific subclasses. As a result, we cannot determine whether annual fluctuations or inflection points reflect shifts in clinical demand, alterations in reimbursement or prescribing restrictions, or tender-driven price changes. The trends should therefore be interpreted as descriptive system-level patterns, rather than causal reflections of specific policy or pricing interventions.

To avoid denominator-driven bias arising from mid-period DDD revisions, we harmonized the dataset by recalculating A10BJ02 and all affected aggregates (A10BJ, A10B and total A10), using the updated DDD.

We did not adjust utilization rates for temporal changes in diabetes prevalence. Data from the International Diabetes Federation (IDF) indicate that the absolute number of adults with diabetes in Croatia increased from roughly 0.22 million in 2011 to approximately 0.41 million in 2024 [23]. This suggests that part of the observed increase in DDD/1000 inhabitants/day reflects a growing patient population. However, even accounting for this rise in case numbers, the relative expansion of GLP-1 receptor agonists and SGLT-2 inhibitors and their disproportionate contribution to total A10 expenditure indicates a genuine intensification and shift in pharmacotherapy, rather than a purely prevalence-driven effect.

This descriptive national analysis also highlights several research priorities. First, linking the HALMED ATC/DDD data with the Croatian Health Insurance Fund's (HZZO) claims and primary-care electronic health records would enable patient-level analyses of indication, treatment lines and appropriateness of GLP-1 and SGLT-2 use, including equity in access across age, sex and region. Second, longitudinal cohort studies using such linked datasets are needed to assess whether the increased utilization of novel antihyperglycemic agents translates into measurable improvements in cardiovascular, renal and liver outcomes at the population level. Third, formal budget-impact and cost-effectiveness models should quantify the medium-term fiscal implications of expanding GLP-1-based and SGLT-2 indications, including obesity and MASLD, under alternative reimbursement and pricing scenarios. Finally, cross-country comparative studies employing harmonized ATC/DDD methodologies could position Croatia's trajectory within a broader European context and identify structural features of health systems that facilitate or constrain a clinically appropriate, fiscally sustainable uptake of high-cost antihyperglycemic drugs.

Croatia's antihyperglycemic drug landscape underwent a marked transformation between 2017 and 2024, characterized by the rapid expansion of GLP-1 receptor agonists and the dual GLP-1/GIP agonist and sustained growth of SGLT-2 inhibitors. These therapies now represent the principal drivers of expenditure growth and have significantly reshaped the composition of national diabetes pharmacotherapy. Inflation-adjusted results confirm that market expansion—not price inflation—was the dominant contributor to rising costs.

Given their escalating budget impact, Croatia requires structured, class-specific monitoring of antihyperglycemic drugs. At a minimum, this should include (i) annual national ATC/DDD analyses from HALMED that report utilization and expenditure separately for insulins, older oral agents, SGLT-2 inhibitors and GLP-1-based therapies; (ii) internal HZZO dashboards that track quarterly expenditure for high-cost classes against budget projections; and (iii) periodic linkage of HALMED aggregates with reimbursement and primary-care data to identify shifts in prescribing patterns and potential over- or under-use in high-risk populations. Such targeted monitoring would enable Croatian decision-makers to detect rapid changes in uptake, anticipate budget pressures and align pricing and reimbursement decisions with evolving clinical evidence.

Routine analysis of HALMED ATC/DDD data is therefore a necessary foundation for pharmaceutical stewardship, but it is not sufficient on its own for fully informed policy decisions. For reimbursement design, assessment of appropriate use and evaluation of clinical outcomes, HALMED aggregates need to be complemented by HZZO claims data and, where feasible, primary-care electronic health records. Only the combination of these sources can show not just how much of each class is used and at what cost, but which patients receive these therapies, under which indications and with what downstream consequences for health outcomes and resource use.

In practical terms, these findings support several policy responses for Croatia and comparable health systems. First, HZZO could periodically review reimbursement criteria for GLP-1-based and SGLT-2 therapies to ensure that coverage prioritizes patients at the highest cardiovascular-, renal- or obesity-related risk, thereby maximizing clinical value per euro spent. Second, national budget impact analyses based on HALMED and HZZO data should be used in price negotiations and, where appropriate, in designing risk-sharing or outcome-based agreements for high-cost incretin-based agents. Third, formulary committees and professional societies should use these utilization patterns to align local prescribing recommendations with international guidelines, while discouraging low-value use in lower-risk populations. Finally, tracking regional and provider-level variation in the uptake of GLP-1 and SGLT-2 therapies could help identify unwarranted underuse or overuse and inform targeted educational or regulatory interventions.