Once-Weekly Subcutaneous Semaglutide 2.4 mg Injection is Cost-Effective for Weight Management in Spain Authors: Andreu Altés, Óscar Moreno-Pérez, Miquel Sastre-Belío, Ened Rodríguez-Urgellés, Ferran Pérez-Alcántara, Sara Larsen, Alberto Martín-Lorenzo, Miriam Sandín, Josep Vidal Published: 2026 ABSTRACT IntroductionA current major challenge for national health systems (NHS) is the increase of obesity and overweight among populations. Subcutaneous semaglutide 2.4 mg, a glucagon-like peptide 1 analogue, has been approved by the European Medicines Agency as an adjunct to a reduced-calorie diet and increased physical activity (diet and exercise [D&E]) for the treatment of obesity. The aim of this study was to evaluate the cost-effectiveness of semaglutide 2.4 mg in combination with D&E compared with D&E alone in the treatment of patients with obesity in Spain.MethodsThe analyses were performed using version 26 of the Core Obesity Model, a Markov state transition model, to project health outcomes and costs at 40 years, based on the evolution of risk factors associated with obesity. Efficacy and safety data were obtained from the STEP 1 trial (body mass index [BMI], systolic blood pressure and glycaemic status), involving a cohort of adults with an average age of 46 years and obesity (BMI ≥ 30 kg/m2). The analysis was carried out from the perspective of the Spanish NHS. The results were evaluated in terms of cost per quality-adjusted life year (QALY) gained. Only direct healthcare costs were included. Univariate and probabilistic sensitivity analyses were performed to ensure the robustness of the findings.ResultsCompared with D&E, semaglutide 2.4 mg in combination with D&E generated an additional 0.1049 QALYs and a €2685 increase in costs over 40 years, resulting in an incremental cost-effectiveness ratio of €25,589/QALY. Semaglutide 2.4 mg was shown to be cost-effective under a cost-effectiveness threshold of €30,000/QALY. The robustness of this finding was confirmed by the sensitivity analyses and scenarios evaluated.ConclusionsOn the basis of the model-based analysis, semaglutide 2.4 mg, in combination with D&E, was estimated to be a cost-effective therapeutic alternative in Spain for adults with obesity, compared to treatment based on D&E alone. These results may be useful for optimizing decision-making.Supplementary InformationThe online version contains supplementary material available at 10.1007/s12325-026-03644-x. Key Summary Points Why carry out this study?Obesity is a chronic health condition associated with high morbidity/mortality and a significant healthcare cost burden in Spain. Effective and timely management of obesity is key to reducing both clinical and economic burden on patients and the healthcare system.This economic evaluation aimed to assess whether semaglutide 2.4 mg in combination with diet and exercise (D&E) compared to D&E alone is cost-effective for the treatment of adults with obesity (patients with a body mass index [BMI] ≥ 30 kg/m2) in Spain.The analysis provides a Spain-specific assessment, reflecting local clinical practice and including relevant Spanish subpopulations and a representative obesity cohort.What was learned from this study?When compared with D&E alone, semaglutide 2.4 mg showed higher health benefits and total costs with an incremental cost-effectiveness ratio of €25,589/quality-adjusted life year (QALY) gained, demonstrating its cost-effectiveness under the present model-based analysis.Different sensitivity analyses (deterministic and probabilistic) confirmed semaglutide 2.4 mg to be cost-effective in most cases at a willingness-to-pay threshold of €30,000/QALY. Alternative scenario analyses showed semaglutide 2.4 mg to be cost-effective in patients with BMI ≥ 35 kg/m2 and BMI ≥ 30 kg/m2 and at least one cardiovascular event. Why carry out this study?Obesity is a chronic health condition associated with high morbidity/mortality and a significant healthcare cost burden in Spain. Effective and timely management of obesity is key to reducing both clinical and economic burden on patients and the healthcare system.This economic evaluation aimed to assess whether semaglutide 2.4 mg in combination with diet and exercise (D&E) compared to D&E alone is cost-effective for the treatment of adults with obesity (patients with a body mass index [BMI] ≥ 30 kg/m2) in Spain.The analysis provides a Spain-specific assessment, reflecting local clinical practice and including relevant Spanish subpopulations and a representative obesity cohort.What was learned from this study?When compared with D&E alone, semaglutide 2.4 mg showed higher health benefits and total costs with an incremental cost-effectiveness ratio of €25,589/quality-adjusted life year (QALY) gained, demonstrating its cost-effectiveness under the present model-based analysis.Different sensitivity analyses (deterministic and probabilistic) confirmed semaglutide 2.4 mg to be cost-effective in most cases at a willingness-to-pay threshold of €30,000/QALY. Alternative scenario analyses showed semaglutide 2.4 mg to be cost-effective in patients with BMI ≥ 35 kg/m2 and BMI ≥ 30 kg/m2 and at least one cardiovascular event. Introduction Obesity is a chronic, complex, progressive and multifactorial disease characterized by abnormal or excessive accumulation of fat (adiposity) that is detrimental to health [1]. It currently represents one of the main challenges for public health worldwide as a result of its impact in terms of early morbidity and mortality and its many associated complications [2–4]. In recent decades, global prevalence of obesity has increased almost threefold [5, 6]. Spain has one of the highest obesity rates in Europe [7], with a total prevalence of around 22%, affecting 1 in 5 adults [8]. Around 42,000 deaths were attributed to excess body weight in Spain in 2019 [9]. Higher body mass index (BMI) levels are linked to an elevated risk of complications, increased morbidity and mortality, and reduced patient quality of life [10–19]. Complications such as cardiovascular disease (CVD), type 2 diabetes mellitus (T2DM), metabolic fatty liver disorder, obstructive sleep apnoea (OSA), osteoarthritis, anxiety, depression and some types of cancer are closely related to obesity and BMI increase [10, 20–26]. Excess body weight represents a high economic and social burden to healthcare systems, mainly due to the high cost of managing chronic disorders and productivity losses [27]. In Spain, it is estimated that a 15% weight reduction in people with obesity could prevent multiple complications and generate savings of €8 billion over 10 years [28]. In addition, a recent report projected that the total cost associated with obesity in 2025 will reach €130,780 million (8% of the gross domestic product), rising to €161,477 million in 2030—an increase of 23% [29]. The prevention and management of overweight and obesity represent major challenges due to their complex and multifactorial nature involving the interaction of genetic, physiological, behavioural and environmental factors [30]. The management of obesity includes behavioural and lifestyle interventions, drug treatment and bariatric surgery [31]. While lifestyle interventions such as diet and exercise (D&E) can result in significant health benefits, they usually lead to modest short-term weight loss. Despite these interventions, a large proportion of patients have difficulties related to metabolic adaptation and/or maintaining D&E, which may contribute to the recovery of lost body weight [32]. In fact, in most cases, weight loss cannot be maintained in the long term, with a proportional recovery of lost weight of between one-third and two-thirds at 1 year, and over 95% at 5 years [33]. The recommendations of the GIRO guide (Spanish Guide for the Comprehensive and Multidisciplinary Management of Obesity in Adults), conducted with the participation of 38 scientific societies and 12 patient associations in Spain, state that drug treatment for obesity should be considered in individuals with a BMI ≥ 27 kg/m2 with at least one complication related to adiposity present, including T2DM, high blood pressure, OSA, CVD, heart failure (HF) and osteoarthritis, among other health problems or in individuals with a BMI ≥ 30 kg/m2 regardless of the presence of adiposity-related complications [31]. Recently, the World Health Organization published its first guide on the use of therapies based on glucagon-like peptide 1 (GLP-1) analogues to treat obesity as a chronic, recurrent disease, recommending the use of GLP-1 therapies in adults, excluding pregnant women, for the long-term management of obesity. In addition, emphasis was placed on the need for global policies to ensure equitable access and the preparation of healthcare systems for their implementation [34]. In this regard, pharmacological treatments can play a key role in improving the health outcomes of patients with obesity, as well as in helping to reduce the economic impact of obesity on the Spanish National Health System (NHS) [28]. Semaglutide 2.4 mg, a GLP-1 analogue authorized by the European Medicines Agency, has demonstrated a favourable benefit–risk balance by promoting weight loss. Semaglutide 2.4 mg reduces energy intake, increases feelings of satiety, fullness and control of eating, reduces feelings of hunger, and frequency and intensity of cravings [35]. Semaglutide 2.4 mg has been evaluated in over 25,000 people with obesity and overweight in the STEP (Semaglutide Treatment Effect in People with Obesity) and SELECT (Semaglutide Effects on Cardiovascular Outcomes in People With Overweight or Obesity) clinical trial programs, across 14 clinical trials [36–51]. In the STEP trials, treatment with semaglutide 2.4 mg was associated with a clinically relevant mean weight loss of 16.7–20.7% in people with obesity or overweight without T2DM [36, 38, 39, 43, 44, 51]. In addition, semaglutide 2.4 mg is the only obesity drug with evidence of reduced risk of cardiovascular (CV) death, all-cause mortality, stroke and myocardial infarction (MI) [48], and has shown improvement in complications such as HF with preserved ejection fraction, prediabetes, osteoarthritis and metabolic steatohepatitis [40, 47, 50, 52]. To address the long-term economic challenges associated with therapeutic innovations and the uncertainty regarding their relative value, the present study assessed the incremental cost-effectiveness ratio (ICER) of semaglutide 2.4 mg in combination with D&E versus D&E alone for the treatment of adults with obesity (BMI ≥ 30 kg/m2) in Spain. Methods A cost-effectiveness analysis was conducted to assess the use of semaglutide 2.4 mg, in combination with D&E, as treatment for adults with obesity (BMI ≥ 30 kg/m2), versus treatment based on D&E alone. The analysis estimated the ICER to express the incremental difference in health outcomes and costs, measured in quality-adjusted life years (QALYs) and life years (LYs) gained, between semaglutide 2.4 mg, in combination with D&E, and D&E alone. Since there is not an official efficiency threshold for decisions related to the reimbursement of health technologies in Spain, a willingness-to-pay (WTP) threshold of €30,000 per additional QALY gained was assumed. This threshold is aligned with the recent study published by Vallejo-Torres, which places the incremental cost per additional QALY between €27,000 and €34,000 [53], and has been commonly used in other cost-effectiveness publications in Spain. Thus, in the present analysis, ICER was compared with a WTP threshold of €30,000 per additional QALY gained. The analysis was performed from the perspective of the Spanish NHS, including only direct healthcare costs (€ 2025), and considering a time horizon of 40 years (patient lifetime). An annual discount rate of 3% was applied for both costs and future effects. Patients The population included in the analysis was a subsample of patients (n = 1844) from the total population of the STEP 1 trial, comprising adults with an average age of 46.1 years and BMI ≥ 30 kg/m2, eligible to receive treatment with semaglutide 2.4 mg. The baseline characteristics of the baseline cohort are shown in Table S1. Description of the Model The analysis was performed by adapting version 26 of the Core Obesity Model (COM), a Markov state transition model developed to estimate the cost-effectiveness relationship of semaglutide 2.4 mg in combination with D&E. A detailed description of the COM framework and its external validation have been published previously [54, 55]. In brief, the COM is designed to project long-term health and cost consequences through a selected set of clinically and economically important obesity-related complications for which robust evidence and suitable risk-equation inputs are available, rather than to represent the full spectrum of obesity-associated conditions. The selection of modelled outcomes is consistent with the published COM development paper, which describes inclusion based on strength/moderate evidence of association with obesity, impact on healthcare costs, quality of life and/or survival, and sensitivity to weight-management interventions [55]. External validation work has shown that the COM predicts obesity-related event rates with generally good concordance for key endpoints, while also noting areas for continued refinement [54, 55]. To strengthen the validity of the model and ensure that it faithfully reflects clinical practice in Spain, a group of clinical experts, all co-authors of the present manuscript, reviewed and validated the structure, main assumptions and data used. The model was designed to assess the health outcomes and costs associated with the development of obesity-related complications based on surrogate variables such as BMI, lipids, systolic blood pressure (SBP) and glycaemic status. The relationship between these variables and health outcomes was incorporated into the model using risk equations, reflecting the association between these surrogate variables and the incidence of different disease conditions. The obesity-related complications included in the model were T2DM, acute coronary syndrome (ACS) (defined as MI and unstable angina), stroke (including transient ischaemic attack [TIA]), OSA, and knee replacement surgery secondary to osteoarthritis. The conceptual design of COM is shown in Fig. 1.Fig. 1Core obesity model structure. ACS acute coronary syndrome, BMI body mass index, HDL high-density lipoprotein, HF heart failure, OSA obstructive sleep apnoea, SBP systolic blood pressure, T2DM type 2 diabetes mellitus Core obesity model structure. ACS acute coronary syndrome, BMI body mass index, HDL high-density lipoprotein, HF heart failure, OSA obstructive sleep apnoea, SBP systolic blood pressure, T2DM type 2 diabetes mellitus Each health state in the model is associated with a cost and quality of life (utility) while the cohort remains in that specific health state. Events such as knee replacement surgery, stroke, TIA, MI, unstable angina, and bariatric surgery generate a unique cost and loss of quality of life (disutility) during the cycle in which they occur. During the first year, a cycle length of 3 months was established to more accurately represent treatment effects and consider treatment discontinuation due to lack of response. Subsequently, annual cycles were considered, to which a half-cycle correction was applied to represent that the transition between states can occur at any time within the cycle. Treatment Effect and Progression Parameters Clinical efficacy estimates for semaglutide 2.4 mg in the treatment of obesity, along with D&E, were integrated into the model using data from the STEP 1 trial. This included changes in cardiometabolic factors such as SBP, total cholesterol, high density lipoprotein (HDL) cholesterol, glycaemic status and the proportion of patients achieving weight loss ≥ 5%, regarded as responders (Table S2) [36]. In the base case analysis, the maximum treatment duration was assumed to be 2 years, and the efficacy data used were specific for the subgroup with BMI ≥ 30 kg/m2 (treatment policy estimand). Initially, the effect of treatment was applied in cycle 2 of the model (3–6 months). From cycle 3 (6–9 months) onwards, the effect of treatment was applied to the responders. Changes observed at week 68 during STEP 1 were maintained in the second year of treatment for the variables body weight, SBP, total cholesterol and change in HDL cholesterol. A stopping rule was applied where treatment was discontinued for patients who did not achieve ≥ 5% weight loss; this threshold was used to define responders, reflecting a commonly accepted benchmark for clinically meaningful weight reduction and as a pragmatic criterion for treatment continuation in weight management, consistent with clinical practice guidelines and regulatory perspectives that consider ≥ 5% weight loss as clinically relevant [56–59]. Non-responders (17.2%) were assigned an efficacy estimate corresponding to the group treated with D&E, assuming that all the patients will continue participating in a D&E program regardless of their response to treatment. In the analysis, temporary reversal of prediabetes was established as an effect of the interventions, differentiating between patients with and without CVD, based on data from the STEP 1 study [36] and the SELECT study (SELECT study data on-file), respectively (Table S3). This effect was only applied in the second cycle of the model, as it was directly linked to the weight loss caused by the treatment. After the discontinuation of treatment, the effects of treatment were assumed to gradually decrease. Accordingly, in the base case we assumed a gradual return of risk factor values towards baseline over a defined catch-up period; this assumption is intended as a conservative representation of total regain. Therefore, an annual recovery rate was applied to gradually restore the risk factor values to their initial levels. For body weight, SBP, total cholesterol and HDL cholesterol, a recovery rate over 4 years was assumed based on data from the extension part of study STEP 1, where an exponential model was fit to the observed data to derive inputs for the model beyond 1 year (Table S4) [60]. For the glycaemic status of prediabetes, the post-treatment catch-up rate over 3 years was informed by looking at the change in glycaemic status observed after 52 weeks off-treatment (week 68 vs. week 120) in STEP 1 extension (Table S5) [60]. These time-varying risk factors are updated each cycle and directly determine event risks and health-related quality of life (HRQoL) through the model’s risk equations and BMI-linked utility function. From this point onwards, a constant annual natural progression of weight (up to a maximum age of 68 years), SBP, total cholesterol and HDL cholesterol was assumed (Table S6). Bariatric surgery was modelled as a potential subsequent treatment option for patients not meeting therapeutic goals under non-surgical management, and could occur in either arm, independent of initial assignment (semaglutide 2.4 mg plus D&E or D&E alone), applying Spain-specific eligibility criteria (minimum BMI threshold and maximum age) and a low annual national incidence rate reflecting limited capacity and waiting lists (Table S7); consequently, only a small proportion of the cohort underwent surgery. All the criteria for bariatric surgery eligibility were validated by clinical experts. The effectiveness of bariatric surgery in the outcomes of the model was associated with reductions in BMI, SBP and lipid levels (Table S7). Clinical safety data were obtained from STEP 1 [36]. Serious gastrointestinal adverse events were included (Table S8). Complications To estimate transition probabilities between health states and the incidence of multiple obesity-related complications, the model uses a combination of risk equations derived from different sources. This approach is standard in disease simulation models when no single risk equation captures all relevant endpoints and allows each outcome to be modelled using the best available validated evidence. Risk factors such as BMI, SBP, total cholesterol, and HDL cholesterol were used as surrogate variables. First CV events were predicted using the QRisk3 model [61], while recurrent CV events were modelled using data from the SELECT study (data on-file) for normoglycaemia (Tables S9 and S10) and the UKPDS 82 model for T2DM [62]. The incidence of T2DM was estimated using the QDiabetes algorithm [63], the prevalence of OSA was estimated from the Sleep Heart Health Study [64], and the incidence of knee replacement surgery was obtained from the literature [65] and Spanish population data [66] (Table S11). Utility Utility was derived from a multivariate statistical model that included as determinant factors the patient BMI, demographic and socioeconomic characteristics, lifestyle behaviours, and complications associated with obesity [67]. The regression analysis was implemented in the model to provide an initial complication-free utility, dependent on the BMI of the group in each cycle, age and sex (using the corresponding regression coefficients). Because the utility function is derived from a 5-level EQ-5D (EQ-5D-5L) multivariable model [63] and is BMI-dependent, changes in BMI translate into changes in overall HRQoL over time. The EQ-5D-5L captures patient-centred impacts relevant to obesity, including limitations in mobility and usual activities, and pain/discomfort, as reflected in the underlying data used to estimate the regression. Disutilities associated with events and health states were then applied, adopting an additive approach as the cohort progressed through health states with complications or experienced events. Disutilities applied to complications and events were obtained from the literature, thus representing marginal disutility specific to the complication, thereby ensuring the additive approach (Table S12). When health states combined multiple complications, the disutilities associated with each were added and subtracted from the baseline utility, following an approach that compensates for the lack of data on the combined impact of these complications. Mortality In the analyses performed, mortality was calculated using the annual probabilities of mortality by age and sex of the general population, obtained from the Spanish mortality tables (Table S13) [68]. These probabilities were adjusted to exclude deaths related to complications of obesity, using disease-specific mortality data (Tables S14 and S15) [69], and subtracting them from the general population mortality. Subsequently, non-disease-specific mortality was adjusted by hazard ratio (HR) per unit increase in BMI, based on a large study by the UK Clinical Practice Research Datalink [70], which allowed disease-specific mortality rates to be obtained according to BMI. This was done to include the additional mortality associated with obesity, due to causes that were not considered in the model and taking into account a possible underestimation of mortality during the validation. The BMI-dependent HRs were multiplied by the mortality rates of the modelled events (MI, unstable angina, stroke, knee replacement surgery associated with osteoarthritis) and the HRs corresponding to post-ACS, post-stroke and T2DM [71–73] (Table S16). Healthcare Resource Utilization and Costs The Spanish NHS perspective was assumed for costs. Semaglutide acquisition costs were based on Novo Nordisk prices per 28-day pack (prices considered as March 2026): semaglutide 0.25 mg (€127.06), 0.5 mg (€127.06), 1.7 mg (€146.58) and 2.4 mg (€169.13). The unit costs applied are shown in Table S17. Sensitivity Analyses Deterministic sensitivity analyses (DSA) were performed to assess how the changes in model parameters impact the outcomes. For this purpose, different parameters were set within a predefined range including values both below and above the baseline value. The parameters evaluated included aspects related to clinical efficacy, disutilities and costs, among others. The results of the DSA were presented using a tornado diagram, highlighting the 10 variables with the greatest impact on the ICER in decreasing order. A probabilistic sensitivity analysis (PSA) was performed to assess uncertainty in the results of the model. One thousand Monte Carlo simulations were performed, where instead of using point values for the parameters, random values determined by parametric distributions were employed. A gamma distribution was considered for costs and the parametric distributions for the rest of the parameters are shown in the supplementary material. The results of the PSA are presented in the cost-effectiveness plane, which shows the incremental cost and QALY results of semaglutide 2.4 mg in combination with D&E versus D&E for each simulation performed; and in the acceptability curve, which shows, from the simulations performed, the probability that a treatment is cost-effective for different WTP thresholds. In addition, various scenario analyses were performed to evaluate the robustness of the results of the cost-effectiveness model and to explore the impact of different assumptions on the estimates, including alternative time horizons, longer maximum treatment durations, analysis of a cohort of patients without T2DM, an alternative initial proportion of women, gradual restoration of the values of surrogate variables after treatment according to natural progression, changes in mortality considered, exclusion of treatment discontinuation, exclusion of bariatric surgery as a potential subsequent treatment option, and use of the efficacy of the trial product estimand (Tables S18 and S19). In addition, the inclusion of health states of colorectal, breast and endometrial cancers as additional complications associated to obesity was considered, together with the health states of the complications already included in the base case, whose incidences were estimated from the risk equations obtained from the National Institute of Health of the American Association of Retired Persons (NIH-AARP) Diet and Health Study [74], the Million Women Study [75] and from a systematic review and meta-analysis [76], respectively (Tables S20 and S21). Furthermore, an analysis was performed using a patient cohort demographic profile reflecting the characteristics representative of patients with obesity in Spain, validated by clinical experts (Table S22). Scenario analyses were also carried out in two additional subpopulations: (i) patients with BMI ≥ 35 kg/m2 and (ii) patients with BMI ≥ 30 kg/m2 who presented at least one CV event (MI, stroke or peripheral arterial disease). The demographic and clinical characteristics of these subpopulations (Tables S23 and S24), as well as the corresponding efficacy data (Tables S25 and S26, respectively) are presented in the supplementary material. Ethics Approval This article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors. Results Base Case Over the tested time horizon, treatment with semaglutide 2.4 mg in combination with D&E yielded an additional 0.1049 QALY and 0.0472 LY gained versus D&E, resulting from the lesser occurrence of obesity-related events (Table 1), with an additional cost of €2685 versus D&E. This additional cost was mainly due to the difference in the cost of obesity management, which includes the pharmacological cost that was partially offset by the costs associated with the occurrence of events (Table 2). The resulting ICER for the model-based analysis was €25,589/QALY (Table 2), and since this is below the assumed WTP threshold of €30,000/QALY, semaglutide 2.4 mg can be considered a cost-effective treatment. Table 1Breakdown of clinical results for semaglutide 2.4 mg vs. D&ESemaglutide 2.4 mg + D&ED&EIncrementalCumulative incidence per 100 patients CVD*42.5443.21− 0.66 Stroke or TIA17.0417.24− 0.21 Bariatric surgery0.010.29− 0.29 Knee replacement90.8291.19− 0.37 Hospitalization for HF1.661.67− 0.02Patient years in health state No complications14.9414.370.57 OSA13.7413.98− 0.24 Prediabetes7.667.86− 0.2 T2DM6.396.71− 0.32 Prediabetes (history of CVD)0.030.010.02 Post-ACS2.922.96− 0.03 Post-stroke1.421.44− 0.02ACS acute coronary syndrome, CVD cardiovascular disease, D&E diet and exercise, HF heart failure, MI myocardial infarction, OSA obstructive sleep apnoea, T2DM type 2 diabetes mellitus, TIA transient ischaemic attack*Includes non-fatal and fatal MI and non-fatal and fatal anginaTable 2Base case cost-effectiveness results for semaglutide 2.4 mg vs. D&ESemaglutide 2.4 mgD&EIncremental vs. D&ETotal costs€47,962€45,277€2685Obesity treatment€12,717€9036€3681Blood pressure treatment€267€266€1Treatment for T2DM€925€985− €59Complications: health states€25,750€26,603− €852Complications: events€8303€8388− €86Total QALY16.0815.970.1049Total LY19.9619.910.0472ICER (€/QALY)25,589ICER (€/LY)56,933D&E diet and exercise, ICER incremental cost-effectiveness ratio, LY life years, QALY quality-adjusted life year, T2DM type 2 diabetes mellitus Table 1Breakdown of clinical results for semaglutide 2.4 mg vs. D&ESemaglutide 2.4 mg + D&ED&EIncrementalCumulative incidence per 100 patients CVD*42.5443.21− 0.66 Stroke or TIA17.0417.24− 0.21 Bariatric surgery0.010.29− 0.29 Knee replacement90.8291.19− 0.37 Hospitalization for HF1.661.67− 0.02Patient years in health state No complications14.9414.370.57 OSA13.7413.98− 0.24 Prediabetes7.667.86− 0.2 T2DM6.396.71− 0.32 Prediabetes (history of CVD)0.030.010.02 Post-ACS2.922.96− 0.03 Post-stroke1.421.44− 0.02ACS acute coronary syndrome, CVD cardiovascular disease, D&E diet and exercise, HF heart failure, MI myocardial infarction, OSA obstructive sleep apnoea, T2DM type 2 diabetes mellitus, TIA transient ischaemic attack*Includes non-fatal and fatal MI and non-fatal and fatal angina Breakdown of clinical results for semaglutide 2.4 mg vs. D&E ACS acute coronary syndrome, CVD cardiovascular disease, D&E diet and exercise, HF heart failure, MI myocardial infarction, OSA obstructive sleep apnoea, T2DM type 2 diabetes mellitus, TIA transient ischaemic attack *Includes non-fatal and fatal MI and non-fatal and fatal angina Table 2Base case cost-effectiveness results for semaglutide 2.4 mg vs. D&ESemaglutide 2.4 mgD&EIncremental vs. D&ETotal costs€47,962€45,277€2685Obesity treatment€12,717€9036€3681Blood pressure treatment€267€266€1Treatment for T2DM€925€985− €59Complications: health states€25,750€26,603− €852Complications: events€8303€8388− €86Total QALY16.0815.970.1049Total LY19.9619.910.0472ICER (€/QALY)25,589ICER (€/LY)56,933D&E diet and exercise, ICER incremental cost-effectiveness ratio, LY life years, QALY quality-adjusted life year, T2DM type 2 diabetes mellitus Base case cost-effectiveness results for semaglutide 2.4 mg vs. D&E D&E diet and exercise, ICER incremental cost-effectiveness ratio, LY life years, QALY quality-adjusted life year, T2DM type 2 diabetes mellitus Sensitivity Analysis and Scenarios The DSA identified the discount rate as the parameter with the greatest influence on the results (Fig. 2), while the rest of the evaluated variables showed a more limited effect. In any case, the considered variations did not modify the conclusion of the base case, since in none of the univariate scenarios did the ICER exceed the threshold of €30,000/QALY, thus supporting the robustness of the results of the model.Fig. 2Tornado diagram depicting key drivers for cost-effectiveness of semaglutide 2.4 mg compared to D&E (€2026). CVD cardiovascular disease, ICER incremental cost-effectiveness ratio, OSA obstructive sleep apnoea, QALY quality-adjusted life year, T2DM type 2 diabetes mellitus Tornado diagram depicting key drivers for cost-effectiveness of semaglutide 2.4 mg compared to D&E (€2026). CVD cardiovascular disease, ICER incremental cost-effectiveness ratio, OSA obstructive sleep apnoea, QALY quality-adjusted life year, T2DM type 2 diabetes mellitus The PSA showed all simulations to be located in the northeast quadrant of the cost-effectiveness plane (greater cost, higher gain in QALYs) (Fig. 3). In 100% of the simulations, the ICER was below the WTP threshold of €30,000/QALY (Fig. 4), 97% when the lower range of €27,000/QALY was considered [53].Fig. 3Cost-effectiveness plane for semaglutide 2.4 mg vs. D&E. CI confidence interval, QALYs quality-adjusted life yearsFig. 4Cost-effectiveness acceptability curve for semaglutide 2.4 mg vs. D&E. QALYs quality-adjusted life years, WTP willingness-to-pay Cost-effectiveness plane for semaglutide 2.4 mg vs. D&E. CI confidence interval, QALYs quality-adjusted life years Cost-effectiveness acceptability curve for semaglutide 2.4 mg vs. D&E. QALYs quality-adjusted life years, WTP willingness-to-pay The results of the scenario analyses are shown in Table 3. All these analyses evidenced the robustness of the results, with an ICER below the threshold of €30,000/QALY. Specifically, in the scenario analyses performed for the three subpopulations (cohort of patients with characteristics representative of the patients with obesity in Spain, BMI ≥ 35 kg/m2 and BMI ≥ 30 kg/m2 with at least one CV event), the treatment maintained (and even improved) the cost-effective profile observed in the base case, with lower ICERs (€20,647, €23,887 and €17,071/QALY gained, respectively) all below the established threshold of €30,000/QALY gained. The only exception was when considering a time horizon of 20 years (€32,175/QALY), although this result is within the WTP range reported in the literature [53]. This demonstrates the importance of using a sufficient time horizon to reflect the clinical benefits throughout the lifetime of the patient. Table 3Results of alternative scenario analysisParametersAlternative variableICER (€/QALY)Base case result25,589Proportion of female patients50%25,202Time horizon20 years32,17530 years27,713Treatment duration – both comparators5 years29,54915 years29,487EstimandTrial product23,925Stopping rule for semaglutide 2.4 mgNot included26,193Bariatric surgery as a potential subsequent treatmentNot included25,590Gradual restoration of post-treatment risk factors valuesGradually restore the values of risk factors according to natural progression28,475MortalityNot adjusted by HR28,196% T2DMCohort without T2DM25,576Complications associated with obesityEndometrial, breast and colorectal cancers included26,853Analysis cohortBMI ≥ 35 kg/m223,887BMI ≥ 30 kg/m2 and at least one CV event17,071Cohort adjusted to the Spanish population with obesity20,647BMI body mass index, CV cardiovascular, T2DM type 2 diabetes mellitus, HR hazard ratio, ICER incremental cost-effectiveness ratio, QALY quality-adjusted life year Table 3Results of alternative scenario analysisParametersAlternative variableICER (€/QALY)Base case result25,589Proportion of female patients50%25,202Time horizon20 years32,17530 years27,713Treatment duration – both comparators5 years29,54915 years29,487EstimandTrial product23,925Stopping rule for semaglutide 2.4 mgNot included26,193Bariatric surgery as a potential subsequent treatmentNot included25,590Gradual restoration of post-treatment risk factors valuesGradually restore the values of risk factors according to natural progression28,475MortalityNot adjusted by HR28,196% T2DMCohort without T2DM25,576Complications associated with obesityEndometrial, breast and colorectal cancers included26,853Analysis cohortBMI ≥ 35 kg/m223,887BMI ≥ 30 kg/m2 and at least one CV event17,071Cohort adjusted to the Spanish population with obesity20,647BMI body mass index, CV cardiovascular, T2DM type 2 diabetes mellitus, HR hazard ratio, ICER incremental cost-effectiveness ratio, QALY quality-adjusted life year Results of alternative scenario analysis BMI body mass index, CV cardiovascular, T2DM type 2 diabetes mellitus, HR hazard ratio, ICER incremental cost-effectiveness ratio, QALY quality-adjusted life year Discussion The present model-based analysis shows that treatment with semaglutide 2.4 mg administered weekly, together with D&E, is cost-effective versus treatment based on D&E alone, in adults with obesity (BMI ≥ 30 kg/m2), from the perspective of the Spanish NHS. Treatment with semaglutide 2.4 mg resulted in a 0.1049 QALY increase, at an additional cost of €2685. The ICER was €25,589/QALY gained, which is below the WTP threshold range described by Vallejo-Torres of €27,000 to €34,000/QALY [53]. The different scenario analyses and the DSA confirmed the robustness of the findings. The PSA showed a 100% probability that semaglutide 2.4 mg is cost-effective, considering an efficiency threshold of €30,000 per QALY. The efficacy and safety data from the STEP 1 trial used in the cost-effectiveness analyses provide further support to the effectiveness of semaglutide 2.4 mg in reducing body weight, SBP, lipid levels and glycaemic control versus the benefits of D&E. Other studies have estimated the cost-effectiveness of semaglutide 2.4 mg versus D&E in Europe, but our study is the first one to consider patients with BMI ≥ 30 kg/m2 for the base case of the analysis. In Portugal, the ICER for patients with BMI ≥ 30 kg/m2 and at least one weight-related comorbidity was €13,459/QALY (€18,459/QALY for patients with BMI ≥ 30 kg/m2) [77]. In turn, in the UK, the ICER for patients with BMI ≥ 30 kg/m2 and at least one weight-related comorbidity was £14,827/QALY [78]. These results are consistent with those of the present study, and show semaglutide 2.4 mg to be an efficient treatment in patients with obesity across different healthcare systems. While obesity is associated with a broad range of clinical consequences, the COM represents a selected set of obesity-related complications for which there is relevant evidence, validated risk equations, and robust cost and utility inputs (e.g. T2DM, ACS and stroke/TIA, OSA, knee replacement and selected cancers) [54, 55]. As such, other conditions not explicitly modelled (e.g. liver disease, mental health outcomes and functional impairment) are not directly captured beyond what is reflected through the model’s health-state/event utilities and disutilities, and the results should be interpreted accordingly. However, while most of the effect is likely to have been captured, the exclusion of other complications could mean that the benefits and potential savings associated with weight loss are partially underestimated. The duration of the STEP 1 study was 68 weeks, which is a short period for observing differences in definitive health outcomes, such as the risk of complications and mortality. Thus, 40-year results rely on extrapolation using validated risk equations to translate trial-observed changes in BMI and cardiometabolic risk factors into projected long-term events and QALYs [54, 55]. These assumptions offer a pragmatic approach when long-term data are limited, as is typical in chronic disease modelling (including obesity). Still, this remains a source of uncertainty and is reflected in the higher ICER in the 20-year scenario; accordingly, cost-effectiveness should be interpreted as dependent on assumptions about persistence of effects and long-term risk trajectories. In practice, treatment pathways are more complex than a single fixed course (e.g. interruptions, re-initiation, variable adherence). While the model considers discontinuation and post-discontinuation catch-up, it does not explicitly capture intermittent use or re-initiation; future work using longer-term follow-up and real-world persistence data could help reduce this uncertainty. Assumptions regarding post-discontinuation weight regain were implemented as a conservative structural mechanism to avoid attributing indefinite benefits to treatment once it is stopped, while still reflecting that reversal is typically progressive rather than instantaneous. Accordingly, the model assumes a gradual convergence of body weight toward baseline after discontinuation. In this regard, recent evidence syntheses support this type of pattern [79, 80]. This evidence also notes that available trial follow-up after cessation is limited, so longer-term projections rely on extrapolation. Because post-cessation trajectories remain uncertain and may vary across patients and care pathways, long-term results should be interpreted as conditional on these structural modelling choices. Another required consideration is the uncertainty as to whether the weight reduction and improvement of CV risk factors translate into reductions in complications and mortality over the long term in people with long-standing obesity. Although there is evidence to support this hypothesis, the reported results are based on significant weight losses achieved through bariatric surgery or lifestyle changes [81, 82]; consequently, they might not be fully applicable to semaglutide 2.4 mg. In this regard, semaglutide 2.4 mg has been shown to reduce the risk of major adverse cardiovascular events (CV mortality, non-fatal MI or non-fatal stroke) by 20% [48]. Nevertheless, long-term studies would be required to generate stronger certainty. On the other hand, since the risk equations used in the analysis were estimated using algorithms in populations from the UK, there could be some uncertainty about their application in the Spanish population. However, consolidated risk equations widely used in the literature were selected as an approximation (QRisk3, QDiabetes and UKPDS 82). Sex-specific analyses were not performed because of the absence of robust, sex-stratified efficacy estimates from the clinical trials. The available evidence does not support clinically meaningful heterogeneity in treatment effects by sex, and trials were not designed or powered to support separate economic evaluations by sex. Results should therefore be interpreted as average effects applicable to the indicated population, which is consistent with common practice in HTA-oriented cost-effectiveness analyses. Future research, including adequately powered subgroup analyses and real-world evidence, may help to further elucidate potential sex-related differences in clinical outcomes and economic value. This represents a limitation of the current analysis and an area for further investigation. It should be noted that OSA was not considered an independent health condition, as it may occur together with other obesity-related complications. Furthermore, it was assumed that OSA does not impact mortality nor the probability of transition between health states, although OSA has been identified as a significant predictor of mortality in a univariate analysis [83]. This decision was made in order to reduce the complexity of the model and, in any case, exclusion represented a conservative approach, since semaglutide 2.4 mg led to a lower incidence of OSA. Although cognitive behavioural interventions are recommended as part of a multidisciplinary approach to obesity management, their implementation in routine clinical practice within the Spanish NHS remains limited and its cost was not considered in the analysis. Finally, the analysis did not consider a social perspective, which could be of interest to decision-makers. Given the clinical outcomes of semaglutide 2.4 mg, the inclusion of indirect costs (such as those arising from productivity losses) could improve the ICER obtained. The aim of the present analysis was to establish a direct comparison between treatment with semaglutide 2.4 mg in combination with D&E and a D&E alone strategy. To ensure clarity of the analysis, second-line therapies were not included in the assessed model. In the context of routine clinical practice, it is common for some patients to choose bariatric surgery. However, estimating the annual percentage of adults with obesity who present complications and are subject to the waiting lists in Spain represents a significant challenge. Beyond the limitations discussed above, the model explicitly incorporates a stopping rule for those patients who do not reach minimum weight loss criteria, and it includes a subanalysis assessing the impact of not including such individuals. This approach is particularly relevant, because considering discontinuation and partial or total reversal of the effects after treatment discontinuation aligns better with the reality of adherence and the clinical course in daily practice. Consequently, the model does not assume that the benefit is maintained permanently but incorporates a more realistic view in terms of response and therapeutic continuity; this brings greater external validity to the results and makes them easier to interpret for decision-making. To the best of our knowledge, there have been no previous publications concluding that pharmacological treatment for obesity is cost-effective in the Spanish setting. In this regard, the results of the present study show that semaglutide 2.4 mg is cost-effective for the treatment of obesity in Spain, and help to fill this gap of evidence, providing information for decision-making in the context of the Spanish NHS. Furthermore, the consistency of the findings in the sensitivity analyses and scenario analyses performed on clinically relevant subpopulations, as well as in longer-term treatment durations with semaglutide 2.4 mg, supports the robustness of the model and of the conclusions. Conclusions On the basis of the model-based analysis, semaglutide 2.4 mg in combination with D&E was estimated to be cost-effective compared with D&E alone for patients with BMI ≥ 30 kg/m2 in the Spanish setting, under a willingness-to-pay threshold of €30,000 per QALY gained. Similarly, the results were found to be robust in the evaluated scenarios involving clinically relevant subgroups (such as patients with BMI ≥ 35 kg/m2 or BMI ≥ 30 kg/m2 and at least one CV event), with the ICER remaining below the established threshold. Supplementary Information Below is the link to the electronic supplementary material.Supplementary file1 (PDF 506 KB) Supplementary file1 (PDF 506 KB)