The economic impact of two cohorts, patients uncontrolled on 60 units/day or less of insulin and patients uncontrolled on OADs as background treatments, was assessed using a budget impact model created in Microsoft Excel 2010 (Microsoft Corp, Redmond, WA). The insulin dose in background therapy was determined on the basis of clinical trials, where the maximum dose of iGlarLixi was 60 units/20 µg (60 units of insulin glargine and 20 µg of lixisenatide) [22,23,24]. The budget impact was calculated as a difference between the costs of new and reference scenarios in which patients in both cohorts either switched to or added on respective intensification treatments, i.e., either GLP-1RA, iGlar, RAI, or iGlarLixi, depending on clinical guidelines. The model framework is presented in Fig. 1. Analysis was conducted among a hypothetical plan population of 1,000,000 over a time horizon of 1 year from a US healthcare payer perspective. Medication costs, dispensing costs, resource costs (needle, tests strip, and lancet), and cost benefits due to absolute reduction of HbA1c values while being on specific treatment compared to baseline HbA1c values were also included (diabetes-related drug costs, acute care costs, and diabetes-related outpatient costs) for computing the annual treatment costs. Additionally, co-payment, coinsurance, and inflation rates were applied to assess the actual costs of treatment. Monitoring costs for usage and administration of therapies were not included in the calculations.

Budget impact model framework. GLP-1RA glucagon-like peptide 1 receptor agonists, HbA1c glycated hemoglobin, iGlar insulin glargine, iGlarLixi fixed-ratio combination of basal insulin glargine and lixisenatide, OAD oral antidiabetic drug, PMPM per member per month, RAI rapid-acting insulin, T2DM type 2 diabetes mellitus, BMI body mass index. Patients on metformin in background therapy could only add on intensification therapy and no switching was allowed, and patients on DPP4 in background therapy could not switch to iGlarLixi and GLP-1RA as treatment intensification options

The target population of adults diagnosed with and treated for T2DM was divided into patients on either insulin or OADs in background therapy, where the model index date was defined as the point of intensification (add-on or switch) with iGlarLixi/iGlar/GLP-1RA/RAI following the most recent uncontrolled HbA1c value while treated on their background therapy for at least 3 months. An incident-based modelling approach was employed, in which patients were included in analysis only if they were not previously treated with products considered as intensification therapy. Accordingly, total healthcare costs were calculated on an annual basis for all incident patients in the target population. Results were reported in 2022 US dollars (USD) for total annual costs of new and reference scenarios and budget impact per patient per month (PMPM).

According to 2022 ADA estimates, 90–95% of patients diagnosed with diabetes have T2DM [25]. The midpoint (92.5%) was used as the default value in the model. Other diabetes-related efficacies, events, or safety outcomes aside from HbA1c were not included in analysis. For the combination insulins/OADs, the unit costs were taken as the averages of the unit costs of each brand of insulin/OAD. The unit cost of GLP-1RA as a class was taken as an average of the unit costs of all individual GLP-1RAs. Patient cost-sharing (co-pay or coinsurance) was assumed to be paid once per refill frequency.

A retrospective analysis of the Optum database was conducted among a population of patients with T2DM who met certain inclusion criteria, i.e., patients with at least one diagnosis of T2DM (International Classification of Diseases, 10th Revision [ICD-10] starting with E11 or International Classification of Diseases, 9th Revision [ICD-9] 250.00, 250.02, 250.10, 250.12, 250.20, 250.22, 250.30, 250.32, 250.40, 250.42, 250.50, 250.52, 250.60, 250.62, 250.70, 250.72, 250.80, 250.82, 250.90, 250.92) and had undergone treatment of T2DM either via insulins (basal insulin 60 units or less per day) or OADs between January 1, 2016 and March 31, 2022 (study period). Identified patients must also have had HbA1c values that were clinically relevant and feasible (5–13%), uncontrolled HbA1c values (≥ 7%), and had undertaken background therapy at least 3 months prior to uncontrolled HbA1c. Only patients who intensified with either iGlarLixi, RAI, GLP-1RA, or iGlar for the first time following the uncontrolled HbA1c reading were included. Patients had to also have been continuously enrolled in the health plan for at least 12 months prior to the index date, be at least 18 years of age, with a body mass index (BMI) between 10 and 40 kg/m2 pre-index, and not be pregnant (specific to female patients). An upper limit of 40 kg/m2 was set for BMI as bariatric surgery becomes a treatment option for these patients; thus, accounting for pharmacological treatment only would not adequately capture the clinical complexity of such cases [26]. Likewise, 15.5% of patients with T2DM are estimated to have BMI ≥ 40 kg/m2 [26, 27], so their exclusion from analysis would not majorly impact model results.

The proportion of patients coming from a background therapy of insulin or OADs (the insulin and OAD cohorts, respectively) and intensifying to respective treatments were based on the inclusion criteria specified above and are depicted in the consort diagram (Fig. 2). The list of background therapies and possible treatment intensifications for both cohorts are presented in Tables 1 and 2, respectively. Additionally, the average HbA1c reductions due to iGlarLixi/iGlar/ GLP-1RA were captured from the literature for both insulin [28] and OAD [29] cohorts.

Real-world analysis consort diagram. BMI body mass index, GLP-1RA glucagon-like peptide 1 receptor agonists, HbA1c glycated hemoglobin, iGlar insulin glargine, iGlarLixi fixed-ratio combination of basal insulin glargine and lixisenatide, NPH neutral protamine Hagedorn, OAD oral antidiabetic drug, RAI rapid-acting insulin, T2DM type 2 diabetes mellitus

The unit costs for the drugs were the wholesale acquisition costs (WAC), obtained from data on file (Table 3) [30]. The cost of dispensing was assumed to be the same for all the treatment arms, i.e., $13.85 (adjusted to 2022 USD) [31]. The model also included the annual diabetes-related cost savings of $629.53 (adjusting to 2022 USD) for every 1-percentage point reduction in HbA1c among patients with an index HbA1c ≥ 7% [32].

Data for insulin units per day was also retrieved from the retrospective database analysis. The units of all insulins and iGlarLixi were assumed to be 1500 U, whereas the prescription units of all OADs were assumed to be 90 tablets (1 tablet per day) over a 3-month prescription period. The prescription units of GLP-1RA as a class were 42.75 mg based on an average of units prescribed for individual GLP-1RA regimen drugs. GLP-1RA as a class comprised Bydureon® (AstraZeneca group of companies), Ozempic® (Novo Nordisk), Trulicity® (Eli Lilly and Company), Byetta® (AstraZeneca group of companies), Victoza® (Novo Nordisk), Adlyxin® (Sanofi‐Aventis US LLC), and Rybelsus® (Novo Nordisk).

The inputs for co-payment and coinsurance were obtained from Kaiser’s employer health benefits 2022 annual survey based on each drug’s tier placement [33]. Tier placement for the branded drugs was tier 2, while for generic/unbranded drugs it was tier 1.

The costs of test strips, lancets, and needles were included for insulins and GLP-1RA [34,35,36]. The number of these resources consumed per day was based on the package inserts for these drugs.

To evaluate the effect of alternative intensifying market share proportions on healthcare budgets, hypothetical scenarios were analyzed. Table 4 lists the scenarios for the insulin and OAD cohorts.

To investigate the impact of individual parameters on model results, one-way sensitivity analyses (OWSA) were conducted. All parameters were varied by ± 10% of the base case. Analysis was conducted separately for the insulin and OAD cohorts. Variables included in the OWSA were population inputs, baseline HbA1c, dosage, costs, costs related to HbA1c reductions, co-payment, and coinsurance.

The study was performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments. Internal review board approval was not sought, as all analyses were applied to anonymized retrospective patient data. The electronic health record datasets analyzed during the current study are licensed by Sanofi from Optum, and permission to access and use data not publicly available from Optum was obtained.