Budget impact analysis for three glucagon-like peptide-1 receptor agonist-based therapies for type 2 diabetes mellitus management in Saudi Arabia

Abstract

Background and Objectives

This study presents a budget impact analysis (BIA) conducted in Saudi Arabia, evaluating the cost implications of adopting semaglutide, tirzepatide, or dulaglutide in the management of type 2 diabetes mellitus (T2DM) patients. The analysis aims to assess the individual budgetary impact of these treatment options on healthcare budgets and provide insights for decision-makers.

Methods

A prevalence-based BIA was developed using real-world and clinical trials data. The model considered disease epidemiology, medication prices, diabetes management expenses, cardiovascular (CV) complications costs, and weight reduction savings over a 5-year time horizon. One-way and probabilistic sensitivity analyses (OWSA, PSA) were performed to assess the robustness of the results.

Results

Over a 5-year period, the cumulative budget impact for semaglutide, tirzepatide, and dulaglutide were 85,923,089 USD, 169,790,195 USD, and 94,558,356 USD, respectively. Hypothetical scenarios considering price parity between semaglutide and tirzepatide are associated with financial impacts of 85,923,091 USD and 86,475,335 USD, respectively. In the public sector, semaglutide showed the lowest incidence of 3-point major adverse CV events (3P-MACE), with tirzepatide leading in weight loss and HbA1c reduction, and dulaglutide presenting the highest 3P-MACE rates and least improvements in HbA1c and weight. A breakeven analysis suggested that tirzepatide’s list price would need to be $199.91 lower than its current list price to achieve budget impact parity with semaglutide based on currently available evidence. Results from the OWSA suggested that risk reductions for CV events were key drivers of budget impact. PSA results were confirmatory of base-case analyses.

Conclusions

CV cost-offsets and drug acquisition considerations may make semaglutide a favorable use of resources for Saudi budget planners and decision-makers. These results were robust to assumptions regarding the list price of tirzepatide.

Keywords:

• Glucagon-like peptide-1 receptor agonists
• semaglutide
• Saudi Arabia
• type 2 diabetes mellitus
• tirzepatide
• budget impact analysis

JEL CLASSIFICATION CODES:

• I18
• I1
• I
• I19

Introduction

Type 2 diabetes mellitus (T2DM) is a metabolic disorder characterized by hyperglycemia resulting from insulin resistance and impaired insulin secretion, characterized by its chronic nature and accompanying risk of severe complications. T2DM accounts for nearly 90% of diabetes cases worldwide and has been referred to as an epidemic by the World Health Organization^1,2. The Gulf region, including Saudi Arabia, has witnessed an increase in the prevalence of T2DM, with prevalence tripling in the last three decades, resulting in national prevalence estimates of 18.4–28%^3–5. Saudi Arabia now has one of the most highly prevalent rates of T2DM in the world, ranking in the top 10 globally⁶. The humanistic and economic consequences of suboptimal T2DM control are substantial, with increased risks of microvascular and macrovascular complications, reduced quality-of-life, and rising healthcare costs, underscoring the urgency for effective management strategies^7–9.

The management of T2DM has evolved to prioritize not only glycemic control but also weight management and cardiovascular (CV) risk reduction¹⁰. This is reflected in the American Diabetes Association (ADA)/European Association for the Study of Diabetes (EASD) joint guidelines, ADA’s Standards of Care in Diabetes and the American Heart Association^11–14. In this context, glucagon-like peptide-1 receptor agonists (GLP-1 RAs) and the novel GLP-1 RA/Glucose-dependent insulinotropic polypeptide (GIP) molecule tirzepatide have emerged as promising therapeutic options¹⁵. Several GLP-1 RAs have demonstrated benefits beyond glycemic control, including CV and weight loss benefits. CV benefits include reduction of major adverse CV events (MACE), heart failure hospitalization, and renal disease in T2DM patients¹⁶. GLP-1 RAs are increasingly favored in clinical practice as preferred post-metformin treatment and first-line treatment for those with CV risk factors, with both GLP-1 RAs and tirzepatide having favorable HbA1c and weight loss profiles, and GLP-1 RAs having a recommendation for CV and renal endpoints as per ADA guidelines^11–15.

Despite the recognized benefits of GLP-1 RAs, the availability of multiple agents such as semaglutide and dulaglutide has introduced a complex decision-making landscape for optimal treatment strategies. Each agent presents a unique profile in terms of efficacy, safety, and impact on CV outcomes and weight management. Among GLP-1 RAs, semaglutide is suggested to have the best HbA1c-lowering properties and weight loss profile among extant agents¹³. As compared to other GLP-1 RAs, semaglutide also results in the highest reduction in the risk of ischemic stroke (39% versus placebo)^17,18. A matching adjusted indirect comparison also suggested semaglutide may be superior to dulaglutide in terms of reducing MACE, both for primary and secondary cardio-protection¹⁹.

At present, semaglutide is the most widely used GLP-1 RA in Saudi Arabia, accounting for over 70% of all public-sector market share, with dulaglutide comprising over 25% and liraglutide <1.5%²⁰. The landscape in Saudi Arabia is expected to evolve with the availability of tirzepatide²¹. Tirzepatide has shown promising glycemic control and weight loss. The SURPASS-2 trial compared semaglutide and tirzepatide, finding tirzepatide produced non-inferior and superior reductions in HbA1c and weight loss²².

However, the evidence base for tirzepatide’s CV outcomes is pending, resulting in uncertainty regarding its clinical profile and financial implications. Evaluation of CV disease and cost-offsets form an integral part of the clinical and economic evaluation of anti-diabetes agents, especially in Saudi Arabia where CV disease is a substantial component of T2DM burden. Strokes remain among the most substantial CV events in Saudi Arabia, and avoidance of strokes is valuable in terms of clinical, humanistic, and economic burden^23,24. A registry-based study found that those with T2DM and comorbid CV disease incurred greater costs than those without CV comorbidities²⁵. Among people with T2DM in Saudi Arabia, 68–80% had a CV event during 2 years of follow-up, highlighting the substantial toll of CV disease. Those with non-fatal strokes incurred substantially higher costs than any other CV or renal comorbidity, with costs at 2 years post-stroke being 98,217 Saudi Riyal (SAR) (26,200 USD), followed by coronary artery disease and chronic renal failure.

While both semaglutide and dulaglutide underwent placebo-controlled CV outcomes trials (CVOTs), tirzepatide’s CV safety was assessed as part of a pre-planned meta-analysis of the pivotal SURPASS trials²⁶. Assessing comparative CV benefit between CVOTs is already known to be challenging²⁷. The use of a mixed comparator from the SURPASS trials results in further uncertainty in comparison to other GLP-1 RAs. Of the 2,328 patients in the mixed comparator arm, 12.3% received placebo, 58.4% received insulin degludec or glargine, and 29.3% received a GLP-1 RA. Over 23% of GLP-1 patients received dulaglutide 0.75 mg, a dosage which does not have a CV indication as per its FDA labeling, resulting in ongoing difficulties in establishing comparative CV profiles^20,28.

At present, both semaglutide and dulaglutide carry indications for both the treatment of T2DM and reduction of MACE among patients with established CV disease in its United States (US), European, and Saudi regulatory labeling^29,30. ADA guidelines further suggest that GLP-1 RAs have additional benefits over insulin and sulphonylureas for CV endpoints, but the claim is not extended to tirzepatide. Extant meta-analyses of GLP-1/GIP RAs have not yet estimated tirzepatide’s comparative CV benefit^21,31,32. The US Institute for Clinical and Economic Review (ICER) were unable to include macrovascular complications in their network meta-analysis, judging that current evidence is insufficient to demonstrate a net health benefit compared to injectable semaglutide³³. This was echoed by Australia’s Pharmaceutical Benefits Advisory Committee (PBAC) who provided a negative recommendation due to uncertainties in safety and cost-effectiveness³⁴. Finally, while tirzepatide offers more reduction of HbA1c, evidence suggests that cardioprotective effects of anti-diabetes agents are separate from their HbA1c-lowering effects³⁵.

Given the significant prevalence of T2DM in Saudi Arabia, the varied profiles of GLP-1 RA and GLP/GIP molecules^33,34, a growing interest on the part of Saudi decision-makers to evaluate tirzepatide, and differences in pricing between molecules³⁶, this study aims to evaluate the financial impacts of using semaglutide, tirzepatide, and dulaglutide in the management of T2DM through a comprehensive budget impact analysis (BIA). As part of its Vision 2030, Saudi Arabia established a formal center for health technology assessment to ensure health technologies value for money and affordability^37,38. The BIA can support in informing decision-makers, including healthcare providers, policymakers, and payers, in understanding the potential financial implications associated with these therapeutic options, including potential cost-offsets and cost-savings. The aim of this study is to evaluate the financial impact of each of the three molecules in the management of T2DM in Saudi Arabia based on their respective clinical profiles, considered as a set of separate scenarios reflecting procurement dynamics for the Saudi public-sector payer.

Methods

Model structure

A prevalence-based BIA was developed using Microsoft Excel^® among adults with T2DM in Saudi Arabia. The analysis calculates the expected national-level financial impact under each of three medication regimens. Annual cycles, levels of efficacy, and incidence of complications dictated how patients accumulated costs over a duration of 5 years. The model includes a half-cycle correction given the relatively long cycle length. The model was developed according to best practice guidelines for BIA from the International Society of Pharmacoeconomics and Outcomes Research (ISPOR)³⁹. BIA inputs and parameters were all validated through interviews with Saudi Arabia clinical and health economics experts.

Perspective and time horizon

The BIA was conducted from the perspective of the Saudi public payer at the level of individual public payer accounts, considering the healthcare costs incurred by the public healthcare system. The analysis utilized a 5-year time horizon to capture both short-term and medium-term budgetary implications. As Saudi Arabia payer’s perspective only considers direct medical costs, indirect costs are not considered.

Interventions and market share assumptions

The BIA considered three treatment regimens: semaglutide (1.0 mg), tirzepatide (15 mg), and dulaglutide (1.5 mg) reflecting dosage forms available in Saudi Arabia.

Market share scenarios were evaluated for each regimen separately, assuming a 100% market share for the purpose of analysis. This assumption reflects the multiple separate accounts that comprise the Saudi public healthcare sector (such as the Ministry of Health, Ministry of National Guards Health Affairs, and several others), each with their own decision-making processes and formularies. At present, each budget-holder typically lists and reimburses a single GLP-1 RA molecule in their respective formulary. The choice of listed agents is based on a centralized tendering and procurement process. The 100% market share assumption reflects the current Saudi market and procurement dynamics and serves as a basis for formulary decision-making.

Epidemiologic and patient parameters

To inform the analysis, several Saudi-specific epidemiologic and patient parameters were incorporated to model patient flow. These included the total prevalent population of individuals with T2DM in Saudi Arabia²⁰, the prevalence of T2DM among the diabetes population⁴⁰, T2DM-attributable mortality rates⁴⁰, the proportion of patients currently using GLP-1 RAs (assumed to cover those who will receive GLP-1/GIP RAs based on market shares)²⁰, and the range of Body Mass Index (BMI) among T2DM patients (Table 1). Additionally, subgroups of patients were considered, including those intensifying treatment with a GLP-1 RA from oral or basal therapies, as well as patients already on a GLP-1 RA (Table 2), the number of GLP-1 RA patients year upon year is presented in Table 3²⁰.

Table 1. Basic model parameters.

Parameter	Value
Total diabetes population size	53,717
Prevalence of T2DM	95%
Diabetes attributable mortality	1.0%
Patients currently using GLP-1 RA	4%
BMI range (kg/m²)	35

Total diabetes population size, proportion currently using GLP-1 and BMI values derived from Novo Nordisk market research²⁰. Prevalence of T2DM and diabetes attributable derived from Saudi Arabia diabetes data available elsewhere⁴⁰.

Table 2. Subpopulations.

Subpopulation	T2DM patients on treatment type	Uncontrolled patients	Patients intensifying with a GLP-1 RA
Oral patients	77%	28%	12%
Basal insulin patients	21%	28%	8%

All values from Novo Nordisk market research²⁰.

Table 3. Derivation of annual patient numbers receiving GLP-1 RA or GIP/GLP-1 agents.

	Year 1	Year 2	Year 3	Year 4	Year 5
Overall T2DM patients (multiplying total population size by the prevalence of T2DM among diabetes patients)	51,031	51,031	51,031	51,031	51,031
Patients intensifying with a GLP-1 RA from oral	1,320	1,320	1,320	1,320	1,320
Patients intensifying with a GLP-1 RA from basal	240	240	240	240	240
Total patients intensifying with a GLP-1 RA	1,560	1,560	1,560	1,560	1,560
Patients on GLP-1 RA	1,837	3,380	4,916	6,444	7,964
Total patients on GLP1-RA	3,397	4,941	6,476	8,004	9,525
Patient mortality	34	49	65	80	95
Patients receiving GLP-1 at the end of the year	3,363	4,891	6,412	7,924	9,429
Half-cycle corrected population	3,380	4,916	6,444	7,964	9,477

All values from Novo Nordisk market research²⁰. Abbreviations. GLP-1 RA, Glucagon-like peptide-1 receptor agonists; GIP/GLP-1, Glucose-dependent/insulinotropic polypeptide/glucagon-like peptide-1.

Clinical inputs

Clinical inputs were included to capture the effectiveness of the different treatment options. The proportion of patients achieving HbA1c levels below 7% at 6 months was derived from pivotal trials (Table 4)^22,41, and transformed to annual estimates using standard transformation techniques⁴². The effects of more favorable HbA1c control were captured in terms of healthcare resource utilization (HCRU) and costs.

Base event rates for CV outcomes, including CV death, non-fatal stroke, and non-fatal myocardial infarction (MI) were obtained from T2DM-specific epidemiologic studies^43–45 (Table 5). In the absence of Saudi-specific epidemiologic studies, event rates were derived from T2DM cohorts in the United Kingdom (UK) and US. The risk reduction associated with the use of GLP-1 RAs for each CV outcome was incorporated into the analysis (Table 4).

Table 4. HbA1c results.

Comparator	HbA1C < 7.0% (26 weeks)	HbA1C < 7.0% (52 weeks)
Semaglutide 1.0 mg	79.8%	95.9%
Dulaglutide 1.5 mg	64.7%	87.5%
Tirzepatide 15 mg	86.6%	98.2%
Odds ratio: tirzepatide 15 mg vs. semaglutide 1 mg	1.63

HbA1c results at 26 weeks for semaglutide and dulaglutide were derived from Hansen et al.⁴¹, while odds ratios from Frías et al.²² were used to derive the proportion of tirzepatide patients achieving HbA1c < 7.0% at 26 weeks. The probability of achieving HbA1c < 7.0% at 26 weeks was converted into a rate and subsequently used to calculate probability at 52 weeks⁴².

Weight reduction and health savings

Mean weight reductions achieved with each treatment option were derived from the relevant randomized controlled trials (RCTs) and incorporated into the analysis^22,46. Relative risk reductions linking weight loss to comorbidity incidence were derived from a Saudi-specific modeling systematic review and simulation⁴⁷. The specific comorbidities under consideration are chronic kidney disease (CKD), heart failure (HF), hypertension, and dyslipidemia (Table 6).

Table 5. Cardiovascular event rates and treatment-specific risk reductions.

	CV death	Non-fatal stroke	Non-fatal MI
Base rate per patient-year	0.02	0.01	0.01
Semaglutide 1.0 mg	0.98	0.61	0.74
Dulaglutide 1.5 mg	0.91	0.76	0.96
Tirzepatide 15 mg	0.90	0.80	0.76

All values are reported as hazard ratios relative to the study-specific comparator. Base rates of CV death, non-fatal stroke, and non-fatal MI were derived from selected cohort studies^43–45. Hazard ratios were obtained from relevant trials for semaglutide and dulaglutide^60,61 and from a meta-analysis of SURPASS trials for tirzepatide²⁶.

Table 6. Comorbidity prevalence and weight loss relative risk reductions.

	Prevalence	Source	RRR for 7% weight loss (semaglutide)	RRR for 3% weight loss (dulaglutide)	RRR for 13% weight loss (tirzepatide)
Chronic kidney disease	24%	^3	3.39%	1.69%	5.1%
Heart failure	2%	^3	3.45%	0.00%	3.45%
Hypertension	41%	^62	2.70%	1.04%	5.2%
Dyslipidemia	67%	^63	1.73%	0.58%	3.5%

Abbreviations. RRR, Relative Risk Reduction.

Values for RRRs linking weight loss to reduction of comorbidities are informed by Alqahtani et al.⁴⁷.

Unit costs

Costs of drug acquisition were considered, as were costs related to complications of T2DM (uncontrolled HbA1c, CV death, non-fatal stroke, non-fatal MI, and the impact of weight loss on chronic kidney disease [CKD], hypertension, dyslipidemia, and HF). To estimate the costs associated with the different treatment regimens and health outcomes, unit costs were assigned using Saudi-specific sources^23,47–49. Cost categories included costs of ongoing care stratified by HbA1c level and costs of clinical events or disease states, both being derived from Saudi cohort studies or cost-effectiveness analyses (Tables 7 and 8)^23,48. Reimbursement pack prices of drugs were derived from the National Unified Procurement Company (NUPCO), the Saudi centralized procurement company (Table 9)⁴⁹. All costs were adjusted to 2022 Saudi Riyal (SAR) values using the Consumer Price Index (CPI) to account for inflation, and subsequently transformed to US Dollars (USD, a rate of 1 SAR = 0.267 USD) for presentation purposes⁵⁰.

Table 7. Annual non-medication costs per patient.

	HbA1c < 7%	HbA1c 7–9%	HbA1c > 9%
Laboratory test costs	174.5	199.5	302.0
Hospitalization costs	249.4	399.2	680.2
Outpatient visit costs	154.7	156.0	158.5
Medical specialist costs	154.7	158.1	152.6
Dietary education costs	14.7	10.8	30.7
Other costs	93.9	192.8	423.6
Total costs	1,018.8	1,341.0	2,114.6

All costs derived from Almutairi and Alkharfy⁴⁸.

All prices in 2023 USD. Note, costs from Almutairi and Alkharfy⁴⁸ were in 2011 Saudi Riyal, while total costs were inflated to 2022 Saudi Riyal and subsequently transformed to USD, accounting for discrepancies between line items and their sum.

Table 8. Event unit costs.

Event	Source	Cost (USD)
Cardiovascular death	^23	6,372.60
Non-fatal stroke	^23	31,412.66
Acute stroke rehabilitation	^23	15,535.33
Non-fatal MI 1st year	^23	17,758.31
Non-fatal MI 2nd + year	^23	1,252.50
CKD	^47	4,787.03
Heart failure	^47	7,175.87
Hypertension	^47	2,250.66
Dyslipidemia	^47	4,240.43

Abbreviations. MI, myocardial infarction; CKD, chronic kidney disease.

Table 9. Drug acquisition and administration costs.

Drug	Pack price (USD)	# Days/pack	Daily cost (USD)
Semaglutide 1.0 mg	79.49	28	2.84
Dulaglutide 1.5 mg	79.89	28	2.85
Tirzepatide 15 mg	278.09	28	9.93

Abbreviation. USD, United States Dollar.

All pack prices obtained from NUPCO and converted to USD⁴⁹.

Sensitivity analyses

One-way sensitivity analysis (OWSA) was conducted to assess the robustness of results to variations in key parameters. Values were varied from their point estimates using 95% confidence interval (CI) ranges for parameters where available and ±20% variation for other parameters.

Given the differences in the list price between existing GLP-1 RA agents (semaglutide, dulaglutide) and (GLP/GIP) tirzepatide, analyses assessing price parity and determining breakeven point were conducted.

Probabilistic sensitivity analysis (PSA) was performed to capture the uncertainty surrounding various inputs and generate a range of possible budgetary outcomes based on simultaneously varying the main variables randomly. Each variable was assigned distribution fitting the range of possible values, comprising beta, lognormal, and gamma distributions for probabilities/proportions, effect measures, and costs, according to best practices⁴². Probabilities and proportions included T2DM prevalence, diabetes attributable mortality, mean weight loss, and patients achieving HbA1c <7.0%. Effect measures included HRs of CV death, stroke, and non-fatal MI. Costs included all disease state, complication, comorbidity, and drug acquisition costs.

Analytical approaches

The BIA evaluated clinical events, costs, and cost-offsets at a national level, considering the budget impact of the different treatment regimens. Estimates are presented in disaggregated fashion, analogously to a cost-consequence analysis^51–53. This permits decision-makers and end users to assign their own relative importance to different outcomes.

Validation

Face validity of the model and internal validation was undertaken by the authors in addition to three non-author health economists (n = 3) and clinical pharmacists (n = 2) selected for their expertise relevant to the model’s focus areas. The validation group comprised Saudi drug information and evaluation experts, clinical subject matter experts, and health economists engaged with Saudi evaluation processes.

The validation process consisted of ensuring the model was transparently reflective of the decision problem and comprised an assessment of the model structure, assumptions regarding treatment mix, target population, treatment pathway, comparators, structural assumptions, default parameter values, results, and sensitivity analyses.

Internal validation was undertaken by health economists external to the modeling team assessing formulas, macros, and model functionality through structured walk-throughs, extreme condition tests, and replication tests.

Results

The total financial impact of each treatment regimen, both year by year and cumulatively across the 5-year horizon, are presented in Table 10. Over the 5-year time horizon, semaglutide, tirzepatide, and dulaglutide were associated with financial impacts of 85,923,089 USD, 169,790,195 USD, and 94,558,356 USD, respectively.

Table 10. Total budget impact.

Period	Semaglutide	Tirzepatide	Dulaglutide
Year 1	8,366,997	16,988,393	9,109,066
Year 2	12,599,267	25,260,344	13,786,528
Year 3	17,008,519	33,745,700	18,688,391
Year 4	21,593,870	42,443,396	23,813,534
Year 5	26,354,439	51,352,369	29,160,841
Total	85,923,092	169,790,202	94,558,360

All outcomes in 2023 USD.

Results by cost category (drug acquisition, T2DM management, and CV complication costs, cost-savings due to weight reduction) in year 1, year 5, and cumulatively are presented for each of the treatment options in Table 11. Costs increased annually, for all drugs across all cost categories due to an increasing number of patients using GLP-1 RAs and GLP/GIP year over year. For semaglutide and dulaglutide, medications prices and diabetes management costs constituted the majority of the financial impact and are numerically similar, constituting over 70% of total costs. Drug acquisition costs for tirzepatide were a key driver of its budget impact (accounting for 68.7% of total financial impact).

Table 11. Disaggregated budget impact by cost category in years 1, 5, and total.

Cost category	Semaglutide	Tirzepatide	Dulaglutide
Medication costs (Year 1, Year 5, Total)	3,503,077	12,254,648	3,520,758
	9,820,791	34,355,609	9,870,359
	33,349,253	116,664,120	33,517,578
Diabetes management (Year 1, Year 5, Total)	3,541,660	3,487,168	3,742,527
	9,928,958	9,776,190	10,492,083
	33,716,563	33,197,800	35,628,815
CV complications (Year 1, Year 5, Total)	1,718,130	1,949,902	2,051,139
	7,714,500	9,192,324	9,374,112
	22,625,947	26,623,928	27,366,964
Weight reduction savings (Year 1, Year 5, Total)	395,869	703,325	152,410
	1,109,810	1,971,755	427,278
	3,768,672	6,695,646	1,450,942

All outcomes in 2023 USD.

Assessing clinical event outcomes, a tirzepatide-only scenario resulted in 31,600 people with T2DM achieving HbA1c <7.0%, 315 cases of non-fatal stroke, and 305 non-fatal MI events.

A semaglutide-only scenario resulted in 30,869 people with T2DM achieving HbA1c < 7.0% (Figure 1), 240 cases of non-fatal stroke (Figure 2), and 296 non-fatal MI events.

Figure 1. HbA1c outcomes – current scenario (semaglutide) and new scenario (tirzepatide).

Figure 2. Number of CV events – current scenario (semaglutide) and new scenario (tirzepatide).

A dulaglutide-only scenario resulted in 28.172 people with T2DM achieving HbA1c < 7.0%, 300 cases of non-fatal stroke, and 385 non-fatal MI events.

A sensitivity analysis assuming price parity with semaglutide indicated that the financial impact in a tirzepatide-only scenario would be USD 86,475,336 (cumulatively across 5 years). Semaglutide’s financial impact would be USD 85,923,091 (Table 12). Tirzepatide results were derived from improved HbA1c reduction and weight loss. Considering the semaglutide-only scenario, improvements in CV outcomes and resulting cost-offsets drove its financial impact profile.

Table 12. Sensitivity analysis – price parity between semaglutide and tirzepatide years 1, 5, and total.

Price parity sensitivity analysis	Semaglutide	New scenario
Price parity: semaglutide and tirzepatide	3,503,077	3,503,077
	9,820,791	9,820,791
	33,349,253	33,349,253
Diabetes management	3,541,660	3,487,168
	9,928,958	9,776,190
	33,716,563	33,197,800
CV complications	1,718,130	1,949,902
	7,714,500	9,192,324
	22,625,947	26,623,928
Weight reduction savings	395,869	703,325
	1,109,810	1,971,755
	3,768,672	6,695,646
Total	8,366,997	8,236,822
	26,354,439	26,817,551
	85,923,091	86,475,336

All outcomes in 2023 USD.

A breakeven analysis suggested that the price of tirzepatide would have to be $1.33 USD lower than the semaglutide pack price, or $199.91 lower than its current list price, to achieve parity with semaglutide. This suggests that even including data from the SURPASS-2 trial that indicates a more favorable HbA1c reduction and weight loss profile than semaglutide, cost reductions are required to achieve budget equilibrium. At a pack price $20.01 above that of dulaglutide, tirzepatide achieves budget impact parity with dulaglutide. Dulaglutide would require a pack price $18.67 below that of semaglutide to achieve budget impact parity with semaglutide.

Sensitivity analysis

Results of OWSA for all three treatment options are presented in Figure 3, comprising tornado diagrams. Varying parameters through their confidence intervals (where available) or ±20% otherwise, all three options were found to be most sensitive to price-per-pack, annual non-medication costs, proportion achieving HbA1c% target, incidence of non-fatal stroke, and hazard ratio of non-fatal stroke. All regimens were most sensitive to price-per-pack, with financial impact demonstrating a USD 13,299,519 (16.8%) between lower and upper bounds, while dulaglutide exhibited a USD 13,366,646 (15.2%) spread. Tirzepatide exhibited the most significant variation, with a USD 46,525,079 (31.8%) range in financial impact. For all of the five most impactful parameters, tirzepatide was associated with the largest absolute variation around its base case financial impact.

Figure 3. One-way sensitivity analysis.

Results of the PSA for all three treatment options are presented in Figure 4, comprising box and whisker plots for each of the three regimens. Using 1,000 Monte Carlo simulations, distributions of financial impacts were generated for each treatment. The box plots illustrate the variability in financial impact across the treatments, with median financial impacts of USD 82,712,707 (semaglutide), USD 91,555,859 (dulaglutide), and USD 164,575,262 (tirzepatide). Probabilistic point estimates are close to deterministic point estimates. Semaglutide has a consistent distribution of financial impacts with the lowest interquartile range of the three regimens, with most simulations clustering around the USD 80 million average.

Figure 4. Probabilistic sensitivity analysis.

Discussion

In the evolving landscape of T2DM management in Saudi Arabia, our study was conducted against a backdrop of widespread semaglutide use and burgeoning payer interest in the comprehensive evaluation of the emerging tirzepatide molecule. The current BIA conducted from the Saudi public payers’ perspective aims to address the knowledge gap. This study represents the first BIA that considers tirzepatide in the Saudi context and addresses the clinical and economic impact of established GLP-1 RAs in addition to evaluating tirzepatide, aiming to render the diversity of therapeutic profiles and drug costs usable by decision-makers.

The findings suggest that semaglutide offers the most favorable economic profile, primarily driven by its reductions in the number of non-fatal strokes and non-fatal MI events. Semaglutide was associated with favorable CV cost offsets, medication costs, and resulting financial population impacts when compared with tirzepatide, despite the cost-savings due to weight loss and HbA1c reduction accruing to tirzepatide. Comparatively, while tirzepatide showed superior HbA1c control and weight loss in the SURPASS-2 trial, which compared tirzepatide and semaglutide²², the uncertain data surrounding its CV outcomes necessitate a cautious interpretation of its long-term value and cost-effectiveness. The current analysis suggests that, for tirzepatide to achieve financial impact parity, its pricing would need to be adjusted below semaglutide’s current list price. Given the noted disparity in prices between tirzepatide and other GLP-1 agents across markets, this pricing disparity can further contribute to differences in their respective financial impacts³⁶. Our analysis is consistent with a cost-of-control analysis comparing semaglutide and tirzepatide which found that the cost to achieve disease control in persons with T2DM was up to three times lower with semaglutide than tirzepatide⁵⁴. These results, along with the deliberations of health technology assessment bodies^33,34, highlight the issue of pricing disparities in this class of medications and their implications for healthcare budgeting and resource allocation, particularly in the absence of clear incremental clinical benefits.

Dulaglutide, being equivalently priced to semaglutide, demonstrated higher overall costs in T2DM management and CV complications alongside lower savings due to weight loss, further reinforcing semaglutide’s position as the economically favorable option among the GLP-1 RA class. Previous RCTs and matching-adjusted indirect comparisons substantiate the present research’s finding and suggest that semaglutide may be the most favorable option among GLP-1 RAs specifically based on currently available evidence^17,19,46,55. Our findings reflect those of a previous Saudi cost-of-control analysis⁵⁶ which found semaglutide to offer the most favorable economic profile compared to other agents of the GLP-1 RA class.

All results were robust to price parity analyses and PSA. Deterministic sensitivity analysis confirmed that key drivers of costs within individual drug scenarios were risk reductions of non-fatal stroke or MI events, indicating the impact of CV outcomes on the clinical and economic assessment of anti-diabetes agents.

It is imperative to acknowledge the evolving nature of the evidence base for newer medications like tirzepatide. Health technology assessment bodies, including PBAC and ICER, have noted the current insufficiency of evidence to provide a positive recommendation or determine a net clinical benefit compared to semaglutide^33,34. The UK’s National Institute for Health and Care Excellence (NICE) initially issued a negative recommendation for tirzepatide based on uncertainty in its clinical efficacy and cost-effectiveness⁵⁷. A positive decision was issued after additional analyses improved confidence in tirzepatide’s clinical- and cost-effectiveness – however, affordability was not formally assessed and a lack of comparative effectiveness on CV outcomes was noted in the consultation committee papers^58,59. Insufficiency of evidence, with resulting uncertainty in budget impact and economic evaluations, will complicate the attempt to determine appropriate pricing.

The significance of CV considerations in the management of T2DM in Saudi Arabia cannot be overstated, given the high clinical, economic, and humanistic burden of CV events among this patient population. Strokes and MIs, as some of the most impactful and costly CV comorbidities in the Saudi context²⁵, highlight the importance of integrating CV risk assessments in T2DM care and decision-making. The challenge of comparing CVOTs due to non-comparable comparators complicates evidence synthesis and decision-making²⁷. The current study suggests that the CV profile of GLP-1 RA and GLP/GIP class drugs may be the most significant driver of financial impacts for T2DM management in Saudi Arabia.

Based on the results of this study, semaglutide may emerge as a potentially optimal resource allocation choice under Saudi Arabia’s centralized procuring system.

Limitations

There are several limitations to this study that should be acknowledged. First, the BIA relies on available data and assumptions, which may introduce uncertainties in the estimated results. In particular, the differences in comparator between semaglutide/dulaglutide CVOT trials and the tirzepatide meta-analysis prove challenging for both analysis and for drawing interpretations for decision-makers, pending formal evidence synthesis. The lack of availability of Saudi-specific data for some epidemiologic parameters of interest necessitated use of data from other regions. Ongoing evidence generation in Saudi Arabia will help rectify this consideration moving forward.

Conclusion

In conclusion, our budget impact analysis indicates that semaglutide may be the most favorable choice for healthcare budget planners and decision-makers in Saudi Arabia. The favorable cost-offsets related to CV outcomes, in addition to considerations of drug acquisition costs, position semaglutide as a preferred option in the management of T2DM when compared to tirzepatide. While semaglutide and dulaglutide are priced equivalently, semaglutide offers favorable cardiovascular and diabetes management cost-offsets in addition to weight-loss related savings. These findings suggest that semaglutide offers a promising solution in comparison to both dulaglutide, an established GLP-1 receptor agonist, and tirzepatide, a novel GLP/GIP agent.

Transparency

Declaration of funding

This study was funded by Novo Nordisk Saudi Arabia.

Declaration of financial/other relationships

MH is an employee of Novo Nordisk Saudi Arabia. HAAO, HSA, AO, LMA, BAH, IAB, and OA have nothing to disclose.

Author contributions

HAAO, HSA, AO, LMA, BAH, IAH, MH, and OA contributed to the study conceptualization and methods, data interpretation, manuscript drafting, reviewing, and editing.

Reviewer disclosures

Peer reviewers on this manuscript have received an honorarium from JME for their review work but have no other relevant financial relationships to disclose.

Acknowledgements

None to declare.