Long-term cost-utility analysis of exenatide once weekly versus insulin glargine for the treatment of type 2 diabetes patients in the US

Abstract

Objective:

The purpose of this study was to estimate the long-term cost-utility of treating type 2 diabetes mellitus (T2DM) patients with exenatide once weekly (EQW) compared with insulin glargine (IG) from a US payer perspective.

Methods:

A validated computer simulation model, the CORE Diabetes Model, was used to project lifetime clinical outcomes and direct medical costs. Direct medical costs included pharmacy costs and costs associated with the management of diabetes and its complications. The model was populated using patient characteristics (mean age: 57.9 years; mean diabetes duration: 7.9 years; mean HbA1_c: 8.3%; mean body mass index [BMI]: 32.3 kg/m²) and clinical data from a phase 3 clinical trial that compared EQW with IG in T2DM patients on a background of metformin alone or a combination of metformin and a sulphonylurea (DURATION-3). All EQW patients were assumed to have stayed on treatment for 3 years before switching to IG. Health outcomes and costs were discounted at 3% per year. Complication costs were derived from published sources. A range of sensitivity analyses was performed.

Results:

Over a lifetime horizon, and compared with IG, EQW was associated with an incremental cost of $3914 (SD = 2923). EQW was projected to increase life expectancy by 0.135 (SD = 0.216) years and to improve quality-adjusted life expectancy by 0.246 (SD = 0.147) quality-adjusted life years (QALYs), generating an incremental cost-effectiveness ratio (ICER) of $15,936/QALY. Assuming a payer’s willingness to pay threshold of $50,000/QALY, EQW is therefore cost-effective compared to IG. One-way and probabilistic sensitivity analyses confirmed EQW’s cost-effective profile.

Limitations:

Short-term changes (26 weeks) in surrogate end-points (e.g., HbA1_c, weight, complications) from one clinical trial were used to project long-term future effects on clinical outcomes.

Conclusions:

Treatment with EQW is projected to be cost-effective compared to treatment with IG.

Keywords::

• Exenatide
• Insulin glargine
• Cost-utility
• Modeling
• Type 2 diabetes
• United States
• Costs
• Cost-effectiveness

Introduction

International guidelines recommend that type 2 diabetes mellitus (T2DM) be managed by maintaining glycosylated hemoglobin (HbA1_c) to a level below 7.0%, with the aim of preventing or slowing down the development of diabetes complications1–3. Alongside the development of new classes of glucose-lowering interventions, clinical research has increasingly focused on therapies that reduce T2DM risk factors, including obesity3. In countries such as the US, where the proportion of overweight and obese individuals has reached alarming levels4–6, potential benefits associated with weight loss from T2DM treatment could be particularly valuable.

In patients with suboptimal glycemic control despite maximum tolerated doses of metformin or combined metformin and sulphonylurea, early use of insulin can be recommended as a tier 1 treatment approach3,7. In the US, insulin glargine (IG) is a commonly used basal insulin8. However, treatment with IG is associated with side-effects such as hypoglycemia and weight gain8.

Glucagon-like peptide-1 (GLP-1), a peptide hormone that is secreted by intestinal cells after eating, enhances glucose-dependent insulin secretion, suppresses inappropriate postprandial glucagon secretion, slows gastric emptying, and decreases food intake. Exenatide twice daily (ExBID), a GLP-1 receptor agonist administered subcutaneously, was licensed in the US in 20059, where it is indicated for the improvement of glycemic control in patients with T2DM as monotherapy or as adjunctive therapy to metformin, a sulfonylurea (with or without metformin), a thiazolidinedione (with or without metformin), or IG.

A once-weekly exenatide formulation (EQW) is approved for use in the European Union and in the US. EQW uses the same active ingredient as ExBID, except that the exenatide is encapsulated in microspheres made from medical-grade poly-(d, l-lactide-co-glycolide) in the EQW formulation. After subcutaneous administration by the patient, the exenatide is slowly released as the polymer in the microspheres is hydrolyzed into lactic acid and glycolic acid, which are eliminated as carbon dioxide and water, and via diffusion10.

Weekly administration of 2 mg of exenatide was compared with daily administration of IG (starting at 10 IU, titrated to achieve fasting glucose levels of 72–99 mg/dL) in a 26-week, randomized, controlled, phase 3 trial (n = 456) known as ‘DURATION-3’7. Patients included in this study had sub-optimal glycemic control, despite the use of maximum tolerated doses of metformin or of a combination of metformin and a sulphonylurea for at least 3 months. Results from this study showed that both EQW and IG were associated with statistically significant reductions in HbA1_c; however, the reduction with EQW was significantly greater than with IG (treatment difference, −0.16%; 95% CI: −0.29 to −0.03). In addition, EQW was associated with weight loss (mean: −2.6 kg), whereas IG induced weight gain (mean: 1.4 kg) (difference: 4.0 kg; p < 0.0001). The results of this study are published elsewhere7.

Current study objectives

The objective of this study was to estimate the long-term cost-utility of EQW compared with IG from the payer’s perspective in the US, based on the results of the DURATION-3 study7.

Patients and methods

IMS CORE diabetes model

The CORE Diabetes Model (CDM) was developed to estimate the long-term health outcomes and economic consequences of interventions in type 1 and type 2 diabetes mellitus. CDM inputs include baseline cohort characteristics, history of complications, current and future management of diabetes and concomitant medications, treatment effects, and changes in physiological parameters. A series of sub-models within the CDM simulate the major complications of diabetes including myocardial infarction, congestive heart failure, stroke, peripheral vascular disease, neuropathy, foot ulcer, amputation, retinopathy, macular edema, cataract, and nephropathy. Each of the sub-models uses Markov Monte Carlo simulation. Transition probabilities for the model were obtained from published epidemiological and clinical studies, including the Diabetes Control and Complications Trial and the United Kingdom Prospective Diabetes Study11,12. The CDM is consistent with American Diabetes Association computer-based modeling guidelines for assessing the long-term costs and clinical outcomes for various diabetes treatments13. Detailed descriptions of the CDM have been published previously14,15.

Simulation cohort

Characteristics of the modeled cohort were derived from baseline demographics, clinical parameters, and racial characteristics from the DURATION-3 intention-to-treat (ITT) population7. Baseline demographic characteristics, including diabetes-related complication rates, were averaged across both treatment arms, to create a combined cohort (Table 1).

Table 1.  Baseline patient characteristics.

	Mean	SD
Patient demographics
Baseline age, years	57.9	9.4
Duration of diabetes, years	7.9	6.0
Proportion male	53.3	NA
Baseline risk factors
HbA1c, %	8.3	1.1
Systolic blood pressure, mmHg	134.5	16.7
Total cholesterol, mg/dL	186.2	40.9
High-density lipoprotein cholesterol, mg/dL	46.2	11.7
Low-density lipoprotein cholesterol, mg/dL	103.6	35.0
Triglyceride, mg/dL	162.6	86.9
Body mass index, kg/m^2	32.3	5.1
Racial characteristic, %
White	83.1	NA
Black	0.7	NA
Hispanic	10.3	NA
Native American	0	NA
Asian/Pacific Islander	5.9	NA
Baseline CVD complications, %
Myocardial infarction	8.2	NA
Angina	0	NA
Peripheral vascular disease	0	NA
Stroke	4.9	NA
Congestive heart failure	3.7	NA
Atrial fibrillation	0	NA
Left ventricular hypertrophy	0	NA
Baseline renal complications, %
Microalbuminuria	0.6	NA
Gross proteinuria	0.1	NA
End-stage renal disease	0	NA
Background diabetic retinopathy	15.9	NA
Proliferative diabetic retinopathy	1.8	NA
Severe vision loss	0	NA
Baseline macular edema, %	0	NA
Baseline cataract, %	0	NA
Baseline retinopathy complications, %
Background diabetic retinopathy	15.9	NA
Proliferative diabetic retinopathy	1.8	NA
Baseline neuropathy, %	6.5	NA

NA, not applicable.

Model outcomes

In the base case analysis, health outcomes of interest included the cumulative incidence of complications, life years (LYs), quality-adjusted life years (QALYs), and annual cumulative costs per patient. Key incremental health and economic outcomes as well as an incremental cost-effectiveness ratio (ICER) were estimated.

Intervention effects

In the base case analysis, it was assumed that the treatment effects of exenatide would last for 3 years, after which patients would switch to treatment with basal insulin because of disease progression. The initial clinical effects of EQW 2 mg and IG were derived from the 26-week ITT results of the DURATION-3 trial and used as inputs to populate the CDM (Table 2). Detailed results of DURATION-3 have been previously published7.

Table 2.  Base case 26-week clinical data for exenatide once weekly and insulin glargine.

Change from baseline	Exenatide QW 2 mg		Insulin Glargine*
	Mean	SD	Mean	SD
HbA1c, %	−1.47	0.76	−1.31	0.90
Systolic blood pressure, mmHg	−3.03	16.57	−0.63	14.88
Total cholesterol, mg/dL	−4.64	35.42	−1.55	34.65
High-density lipoprotein cholesterol, mg/dL	0.00	5.90	0.39	5.77
Low-density lipoprotein cholesterol, mg/dL	−1.93	29.51	1.55	28.87
Triglycerides, mg/dL	−11.51	101.86	−15.94	112.49
Body mass index, kg/m^2	−0.94	1.09	0.51	1.06
Minor hypoglycemia rate/100 person years	43	–	127	–
Major hypoglycemia rate/100 person years	0	–	0	–

*Starting dose 10 IU, target glucose range 72.1–99.1 mmol/L.

Health state utilities

For T2DM and its complications, health state utilities/disutilities were derived wherever possible from the UKPDS, with data gaps filled with the use of a variety of other relevant sources (Table 3)16–25.

Table 3.  Non treatment-specific health state utilities and disutilities for complications.

Event/state	Utility/ disutility	Reference
Type 2 diabetes no complications	0.814	24
Myocardial infarction, year of event	−0.129	24
Myocardial infarction, years 2+ after event	0.736	24
Angina	0.682	24
Congestive heart failure	0.633	24
Stroke, year of event	−0.181	24
Stroke, years 2+ after event	0.545	24
Peripheral vascular disease	0.570	19
Microalbuminuria	0.814^a	–
Gross proteinuria	0.814^a	–
Hemodialysis	0.604	18
Peritoneal dialysis	0.612	18
Kidney transplant	0.782	25
Background diabetic retinopathy	0.790	20
Proliferative diabetic retinopathy	0.790^b	–
Macular edema	0.790	20
Severe vision loss/blindness	0.570	22
Cataract	0.620	23
Neuropathy	0.630	22
Healed diabetic ulcer	0.814^a	–
Active ulcer	0.750	21
Amputation, year of event	−0.109	24
Amputation, years 2+ after event	0.680	24

^aNo state-specific health utility identified, conservatively assumed to be equivalent to complication-free utility.

^bAssumption.

Disutilities associated with increases in body mass index (BMI), hypoglycemia, and specific adverse events were applied. A loss of utility of 0.0038 was included for each unit of BMI increase above a BMI of 25 kg/m² 26. For hypoglycemia, each value captured the disutility associated with both the experience of a hypoglycemic event and the fear associated with the experience16. Disutilities associated with adverse events such as nausea/vomiting and injection site reaction were estimated based on the incidences of the events from DURATION-3 and disutility values derived from the literature27–29. Disutility caused by nausea was applied only to the first year of treatment (−0.0026 for EQW and −0.0003 for IG)29. A disutility associated with injection site reaction for treatment with EQW of −0.0014 and IG of −0.0002 was assumed in all years of treatment7,28. A utility of 0.0223 was used to capture the benefit of lower administration frequency associated with EQW compared with that of IG (with a utility of 0) in the first year28.

Costs and perspective

The analysis was conducted from a healthcare payer’s perspective, and involved direct costs associated with the pharmacological treatment of diabetes (i.e., EQW and IG pharmacy cost) and with the management of diabetes complications.

Annual pharmacy costs were based on the Wholesale Acquisition Costs (WAC) [source: PriceRx Medispan]30. For EQW, an annual cost of $4215 was used based on $323.44 per package of four single doses of 2 mg each. These costs were published early in 2012, after FDA approval. Pharmacy costs of insulin glargine were estimated based on the 2010 WAC costs. A different cost was estimated for the first year of treatment with insulin glargine ($92.85/1000 units)30, based on a 31.1 units daily dose (from Duration-3 RCT); for the second and subsequent years, costs were based on a 47.2 units daily dose31. Consequently, the annual cost of treatment with insulin glargine was estimated at $1053.99 in year 1 and $1559.62 in subsequent years of treatment.

Costs of complications were derived from published studies32,33, except for the annual costs of treating complications with aspirin (81 mg), simvastatin (80 mg), and lisinopril (40 mg) (Table 4). Annual pharmacy costs for these drugs were estimated with the use of the 2010 PriceRx Medispan Wholesale Acquisition Costs (WAC)30. Original values from the literature were inflated using the healthcare component of Consumer Price Index to reflect 2010 US dollars34.

Table 4.  Direct unit cost of diabetes complications or events, adjusted to $US 2010.

Description of complication or event	Annual costs	Reference
Myocardial infarction, year of event	41,695	33
Myocardial infarction, each subsequent year	2200	32
Angina, year of onset	9850	33
Angina, each subsequent year	2039	32
Congestive heart failure, year of onset	30,066	33
Congestive heart failure, each subsequent year	3418	32
Stroke, year of event	14,104	32
Stroke, each subsequent year	14,104	33
Peripheral vascular disease, onset	9800	33
Hemodialysis	25,150	33
Peritoneal dialysis	29,609	33
Kidney transplant, first year	22,465	33
Kidney transplant, each subsequent year	22,465	33
Retinal photocoagulation	879	32
Severe vision loss/blindness, year of onset	3357	33
Severe vision loss/blindness, each subsequent year	3357^a	–
Cataract extraction	1094	33
Neuropathy, onset	3657	33
Uninfected ulcer	11,042^b	–
Infected ulcer	11,042	33
Gangrene	25,125	33
Amputation, year of event	7371^c	33
Amputation, prosthesis	1271	32
Annual cost of aspirin	4^d	30
Annual cost of statins	144^d	30
Annual costs of ACE-I	50^d	30
Major hypoglycemic event	454	33
Keto event	239	33
Lactic acid event	24,344	33
Edema onset	3116	33
Costs of screening for retinopathy	87	32
Costs of screening for nephropathy	29	32
Costs of non-standard ulcer treatment	172	32

^aAssumption that the cost in subsequent years equals to the cost in 1st year.

^bSet equal to the cost of infected ulcer.

^cEstimated as weighted average of various types of amputations.

^dAnnual costs based on Medispan drug unit prices and dose assumptions agreed on with manufacturer.

Discounting and time horizon

Health outcomes and costs were discounted at a rate of 3% per annum, and the time horizon for the analysis was 35 years in order to capture the effect of treatment on costs and outcomes over a lifetime.

Sensitivity analyses

A range of one-way sensitivity analyses was performed with input parameters varied to assess their effect on clinical and cost outcomes. For example, simulation time horizon was set to 5, 10, 15, 20, and 30 years, and the HbA1_c change from baseline associated with exenatide treatment was changed to both the higher and lower ends of the 95% CI (−1.37% and −1.57%). Several sensitivity analyses were performed in which the effect of BMI on QALYs was waived; disutility associated with nausea, injection site reaction, and the convenience/inconvenience resulting from a once weekly/daily injection with EQW/IG were removed one at a time; and all treatment-specific disutilities in the model (injection-site reaction, nausea/vomiting, and frequency of administration) were removed simultaneously. Two sensitivity analyses were performed to assess the impact of discounting on the results, by using a 5% discount rate and no discount rate (rate set at 0%). A probabilistic sensitivity analysis (PSA) was conducted to assess the combined uncertainty of all model inputs on the cost-effectiveness.

Results

Risks of complications

Table 5 shows the 35-year cumulative incidence of 16 diabetes complications for the two treatment groups, as well as the risk of each complication for EQW relative to IG. Compared with IG, the ratio of cumulative incidence of all complications but one (first stroke: 1.005) were projected to be lower for EQW. Projected lower relative risks ranged from 0.929 for proliferative retinopathy to 0.989 for peripheral vascular disease.

Table 5.  Cumulative incidence and relative risk of diabetes complications for exenatide once weekly vs insulin glargine (35-year time horizon).

Complication	Exenatide QW incidence (%)	Insulin glargine incidence (%)	Relative risk
Cardiovascular disease
Congestive heart failure	36.566	37.926	0.964
Acute myocardial infarction	24.662	25.391	0.971
First stroke	26.550	26.430	1.005
Angina	13.328	13.526	0.985
Peripheral vascular disease	12.363	12.504	0.989
Renal disease
Microalbuminuria	26.925	27.667	0.973
Gross proteinuria	5.007	5.273	0.950
End-stage renal disease	0.671	0.720	0.932
Eye disease
Background retinopathy	19.274	19.921	0.968
Proliferative retinopathy	1.185	1.275	0.929
Severe vision loss	7.294	7.530	0.969
Macular edema	15.135	15.808	0.957
Cataract	9.407	9.626	0.977
Ulcer, neuropathy
First foot ulcer	19.804	20.130	0.984
First amputation	4.662	4.750	0.981
Onset of neuropathy	47.644	48.785	0.977

Incremental health and cost benefits

In the base case analysis, treatments with EQW and IG were respectively associated with a mean life expectancy (LE) of 12.137 and 12.003 years, and a mean quality-adjusted life expectancy (QALE) of 8.476 and 8.231 years. EQW therefore increased mean LE by 0.135 years and QALE by 0.246 years compared with IG (Table 6). EQW was also associated with higher total costs, of $77,382 compared to $73,468 with IG, resulting in an incremental $3914 cost per patient over a lifetime. This additional cost is primarily due to the higher pharmacy costs ($4795) incurred by patients treated with EQW instead of IG in the first 3 years (Table 7). The analyses, however, projected that costs associated with the management of diabetes complications were reduced with EQW due to lower cumulative incidences of cardiovascular diseases and neuropathic complications (e.g., ulcer, amputation).

Table 6.  Base case clinical and economic outcomes associated with exenatide once weekly vs insulin glargine (35-year time horizon).

Clinical and economic outcomes	Exenatide QW	Insulin glargine	Incremental
Life expectancy, mean (SD), years	12.137 (0.163)	12.003 (0.161)	0.135 (0.216)
Quality-adjusted life expectancy, mean (SD)	8.476 (0.114)	8.231 (0.107)	0.246 (0.147)
Lifetime direct medical costs, mean (SD), $	77,382 (2,141)	73,468 (2,065)	3,914 (2,923)
Incremental cost-effectiveness ratio, $/QALY	–	–	15,936

Table 7.  Base case direct medical costs associated with exenatide once weekly vs insulin glargine treatment ($US 2010, 35-year time horizon).

	Exenatide QW	Insulin glargine	Incremental
Total costs	77,382	73,468	3914
Treatment	24,473	19,678	4795
Management	2444	2421	23
Cardiovascular disease	44,266	45,011	−745
Renal	234	256	−22
Ulcer/amputation/neuropathy	4451	4537	−86
Eye	1514	1566	−52

Over a patient’s lifetime, the ICER was $15,936. Assuming a payer’s willingness to pay threshold of $50,000, EQW is therefore cost-effective compared to IG.

Sensitivity analyses

In the base case analysis, a T2DM patient treated with EQW for the first 3 years of treatment was projected to gain an incremental average of 0.246 QALY over a 35-year period compared with a patient treated with IG. Assuming shorter time horizons, the additional QALY gained was smaller: an incremental 0.096, 0.133, 0.176, 0.203, and 0.237 QALY for horizons of 5, 10, 15, 20, and 30 years, respectively. In the base case, EQW was associated with an additional cost of $3914 compared with IG; this incremental cost changed to $4349, $3883, $3578, $3274, and $3712 for horizons of 5, 10, 15, 20, and 30 years, respectively. Corresponding ICERs ranged from $45,077/QALY at a 5-year time horizon to $15,687/QALY at a 30-year time horizon.

Setting the HbA1_c reduction from baseline for EQW to the lower boundary of the 95% confidence interval (−1.57%), corresponding to better disease control, resulted in an incremental QALY of 0.271, compared with an incremental QALY of 0.25 in the base case. In contrast, setting it to the upper boundary (−1.37%), corresponding to poorer HbA1_c control, resulted in an incremental QALY of 0.225. Using these values for HbA1_c improvement resulted in estimations of ICERs of incremental costs of $14,650 (−1.57%) and $18,831 (−1.37%).

The utility parameters that captured a change in BMI and the convenience of less frequent administration of EQW were the utility inputs that impacted incremental QALY and ICER most: when these utility values were not applied in the analyses, the incremental QALY dropped from 0.246 to ∼ 0.182 in both instances, resulting in ICERs of $21,446 and $21,475, respectively, compared with $15,936 in the base case analysis. This change was due to less frequent administration of EQW as well as to the weight gain associated with IG treatment. Removing the disutility associated with injection-site reaction and with nausea had very little effect on the results; incremental QALY of 0.247 and 0.248 were generated (compared with 0.246 in the base case). Overall, the impact of removing all treatment-specific disutilities (convenience of less frequent administration, injection-site reaction and nausea) on incremental QALY (0.186) and ICER ($21,060) was mainly driven by the parameter capturing the increased quality-of-life for patients treated once weekly only.

Setting discount rate to 0% for both costs and health benefits resulted in incremental QALYs gained with EQW rising from 0.246 in the base case to 0.347 and incremental costs from $3914 to $4181, resulting in a reduction of the ICER from $15,936 to $12,062. When increasing the discount rate to 5%, incremental QALYs and costs decreased to 0.203 and $3828, resulting in a higher ICER of $18,869.

Assuming a willingness-to-pay threshold of $50,000, the PSA demonstrated that there is 56.3% likelihood that EQW is cost-effective compared with IG.

Discussion

The results of the DURATION-3 study showed that EQW treatment was associated with significantly greater improvement in HbA1_c than IG treatment. In addition, EQW treatment was associated with weight loss, whereas IG treatment was associated with weight gain7. The current analysis used the baseline characteristics of the patients in the DURATION-3 study, as well as the clinical results of that study, as inputs to populate a validated economic model to estimate the cost-effectiveness of EQW compared with IG. This analysis estimated that, compared with IG, EQW would result in a lifetime incremental 0.246 QALY and additional $3914 lifetime direct cost per patient, generating an ICER of $15,936.

Additional total costs associated with treatment with EQW reflected the higher pharmacy price of EQW compared with IG. The additional cost of the treatment is offset by the reduced cost of the management of T2DM complications in patients on EQW. The cost savings associated with EQW therapy resulted primarily from lower cumulative incidences of cardiovascular diseases and neuropathic complications. Sensitivity analyses were performed for several parameters including time horizon. In the base case, the time horizon was 35 years. When the time horizon was shortened to 5, 10, 15, 20, and 30 years, the incremental QALY gains were estimated to be 0.096, 0.133, 0.176, 0.203, and 0.237 QALY, respectively (compared with 0.246 QALY in the base case). Varying the time horizon also changed the estimates of incremental cost savings to $4349, $3883, $3578, $3274, and $3712, respectively (compared with $3914 in the base case).

Dose frequency and dose flexibility (e.g., ability to take the dose at various times of day) and changes in BMI have been shown to influence utility scores of injectable anti-hyperglycemic drugs26,28. These utilities/disutilities were modeled in the base case analysis. When these utility values were excluded from the analysis, the incremental QALY decreased from 0.246 to 0.183 and 0.182 years, respectively, generating ICERs of $21,446 and 21,475 respectively, compared with $15,936 in the base case. This result suggests that the less frequent administration as well as the improvement in BMI associated with EQW treatment, compared with IG treatment, may be important drivers of the health benefit associated with EQW treatment.

Varying the discount rate applied to both costs and health benefits in two sensitivity analyses (0% and 5%) did not change the cost-effective profile of EQW vs IG.

To our knowledge, this study is the first to assess, from a US payer perspective, the cost-effectiveness of EQW compared to IG over a lifetime. However, a study that also used the CDM to estimate clinical and economic outcomes for EQW compared to IG from the UK National Health Service payer perspective projected clinical outcomes similar to those in this study. In addition, assuming a price for EQW equivalent to liraglutide 1.2 mg, the authors estimated that EQW was cost-effective compared to IG, with a cost per QALY gained of £10,597 35.

There are limitations to the current study. Relatively short-term changes (26 weeks) in surrogate end-points (e.g., HbA1_c, weight, complications) from one clinical trial were used to project long-term future effects on clinical outcomes. However, this caveat is inherent to any economic evaluation using trial data to project incremental costs and health benefits over a patient’s lifetime. However, transition probabilities used in the CDM have been validated against clinical and epidemiological data15.

Furthermore, DURATION-3 is a multi-national study that involved patients from sites outside the US (including in the European Union, Russia, Australia, Korea, Taiwan, and Mexico), limiting the generalizability of these results to a US-only population.

Conclusion

From a payer’s perspective, treatment of patients with T2DM with EQW was projected to be a cost-effective option compared to treatment with IG.

Transparency

Declaration of funding

The funding for the study and manuscript preparation was provided by Amylin Pharmaceuticals, Inc., USA.

Declaration of financial/other relationships

The sponsor had no decisive role in the conduct of the cost-consequence analysis. The sponsor did provide editorial assistance for the manuscript. Although results, interpretations, and conclusions were not dictated by the sponsor, the paper did receive sponsor approval prior to submission. Two of the authors (JHB, SCB) are employees of Amylin who sponsored the preparation of the manuscript. The other authors (AL, A-L G, and YS) are employees of IMS Health and have no relationship (financial, employment, other significant/relevant relationships) with the sponsor.