Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 May 3.
Published in final edited form as: Ann Intern Med. 2022 Oct 4;175(10):1392–1400. doi: 10.7326/M21-2941

First-Line Therapy for Type 2 Diabetes With Sodium–Glucose Cotransporter-2 Inhibitors and Glucagon-Like Peptide-1 Receptor Agonists

A Cost-Effectiveness Study

Jin G Choi 1,*, Aaron N Winn 2,*, M Reza Skandari 3, Melissa I Franco 4, Erin M Staab 5, Jason Alexander 6, Wen Wan 7, Mengqi Zhu 8, Elbert S Huang 9, Louis Philipson 10, Neda Laiteerapong 11
PMCID: PMC10155215  NIHMSID: NIHMS1891477  PMID: 36191315

Abstract

Background:

Guidelines recommend sodium–glucose cotransporter-2 (SGLT2) inhibitors and glucagon-like peptide-1 (GLP1) receptor agonists as second-line therapy for patients with type 2 diabetes. Expanding their use as first-line therapy has been proposed but the clinical benefits may not outweigh their costs.

Objective:

To evaluate the lifetime cost-effectiveness of a strategy of first-line SGLT2 inhibitors or GLP1 receptor agonists.

Design:

Individual-level Monte Carlo–based Markov model.

Data Sources:

Randomized trials, Centers for Disease Control and Prevention databases, RED BOOK, and the National Health and Nutrition Examination Survey.

Target Population:

Drug-naive U.S. patients with type 2 diabetes.

Time Horizon:

Lifetime.

Perspective:

Health care sector.

Intervention:

First-line SGLT2 inhibitors or GLP1 receptor agonists.

Outcome Measures:

Life expectancy, lifetime costs, incremental cost-effectiveness ratios (ICERs).

Results of Base-Case Analysis:

First-line SGLT2 inhibitors and GLP1 receptor agonists had lower lifetime rates of congestive heart failure, ischemic heart disease, myocardial infarction, and stroke compared with metformin. First-line SGLT2 inhibitors cost $43 000 more and added 1.8 quality-adjusted months versus first-line metformin ($478 000 per quality-adjusted life-year [QALY]). First-line injectable GLP1 receptor agonists cost more and reduced QALYs compared with metformin.

Results of Sensitivity Analysis:

By removing injection disutility, first-line GLP1 receptor agonists were no longer dominated (ICER, $327 000 per QALY). Oral GLP1 receptor agonists were not cost-effective (ICER, $823 000 per QALY). To be cost-effective at under $150 000 per QALY, costs for SGLT2 inhibitors would need to be under $5 per day and under $6 per day for oral GLP1 receptor agonists.

Limitation:

U.S. population and costs not generalizable internationally.

Conclusion:

As first-line agents, SGLT2 inhibitors and GLP1 receptor agonists would improve type 2 diabetes outcomes, but their costs would need to fall by at least 70% to be cost-effective.

Primary Funding Source:

American Diabetes Association.

In the United States, type 2 diabetes affects more than 30 million adults (1, 2) and costs approximately $327 billion annually, up from $174 billion in 2007 (3). This dramatic cost increase is partially attributable to the drug classes of sodium–glucose cotransporter-2 (SGLT2) inhibitors and glucagon-like peptide-1 (GLP1) receptor agonists (3–7). Large randomized controlled trials of SGLT2 inhibitors (8–10) and GLP1 receptor agonists (11–15) have demonstrated reductions in atherosclerotic cardiovascular disease (ASCVD), microvascular disease, and mortality in addition to improvements in glycated hemoglobin (HbA1c) and cardiovascular risk factors.

Based on these trials and other evidence, the 2020 American Diabetes Association (ADA) and European Association for the Study of Diabetes (EASD) guidelines recommended SGLT2 inhibitors or GLP1 receptor agonists as second-line therapy, after lifestyle intervention and metformin, in patients with type 2 diabetes without ASCVD, and as first-line therapy with metformin for patients with coexistent—or high risk for—ASCVD, heart failure, or chronic kidney disease (CKD) (16–19). Some experts have gone further to argue for SGLT2 inhibitors and GLP1 receptor agonists as first-line therapy (before metformin) in high-risk populations (20). Trials investigating safety have also suggested that some SGLT2 inhibitors may reduce heart failure admissions even without metformin (21, 22).

Although clinical trials have demonstrated the clinical effectiveness of these newer drugs, they are hundreds of times more expensive than other oral diabetes drugs. In recognition of their high costs, ADA guidelines do not recommend SGLT2 inhibitors and GLP1 receptor agonists as second-line therapy if “cost is a major issue” (19). In 2019, an Institute for Clinical and Economic Review evaluation not only found GLP1 receptor agonists as an add-on therapy to metformin to be cost-effective but also issued an “affordability alert” signaling that GLP1 receptor agonists put a large strain on payers, particularly state Medicaid programs (23). Importantly, the costs of these drug classes may be declining in the next few years because some SGLT2 inhibitors and GLP1 receptor agonists are scheduled to lose (or already have lost) market exclusivity.

Because of the tradeoffs between the clinical benefits of SGLT2 inhibitors and GLP1 receptor agonists and their currently high, but potentially falling, U.S. costs (20), we evaluated the cost-effectiveness of SGLT2 inhibitors and GLP1 receptor agonists as first-line therapies for U.S. patients with type 2 diabetes, compared with metformin as first-line therapy, to inform future clinical guidelines.

Methods

We used an individual patient-level Monte Carlo–based Markov model to simulate the lifetime incidence, prevalence, mortality, and costs associated with a U.S.-representative population with type 2 diabetes not being treated with diabetic medications (24). We compared first-line therapy metformin with using either SGLT2 inhibitors or injectable GLP1 receptor agonists as first-line treatment before metformin for drug-naive patients with type 2 diabetes. We compared drug class-level effects and not individual drug effects because ADA/EASD guidelines specify class level and not individual drug recommendations, and because insurance formularies usually do not cover all SGLT2 inhibitors or GLP1 receptor agonists.

Baseline Strategy: First-Line Metformin (ADA/EASD Guideline Strategy)

The baseline strategy followed the 2021 ADA/EASD guideline (Figure). Whenever HbA1c was at least 7.0%, treatment progression consisted of 1) initial metformin; 2–3) SGLT2 inhibitors or GLP1 receptor agonists with equal probability as second- or third-line options; 4–6) add-on sulfonylureas, thiazolidinediones, or dipeptidyl peptidase-4 inhibitors with a uniform distribution probability (that is, 33% with 3 options) until all recommended combinations were used (dipeptidyl peptidase-4 inhibitors and GLP1 receptor agonists were not combined); 7) add-on basal insulin and discontinued sulfonylureas, thiazolidinediones, and dipeptidyl peptidase-4 inhibitors; and 8) prandial insulin with HbA1c assumed to remain stable.

Figure.

Figure.

Open in a new tab

Medication treatment algorithms.

* A1c = hemoglobin A1c; ASCVD = atherosclerotic cardiovascular disease; CKD = chronic kidney disease; CKD4 = chronic kidney disease stage 4; eGFR = estimated glomerular filtration rate; DPP-4 = dipeptidyl peptidase-4 inhibitor; GLP1 = glucagon-like peptide-1 receptor agonist; SGLT2 = sodium-glucose cotransporter-2 inhibitor; SU = sulfonylureas; TZD = thiazolidinediones.

* The same population was compared using each of the 3 medication treatment algorithms.

† Patients with CKD4 should not start an SGLT2, and should stop one that has been started; SGLT2 should be stopped and not started.

‡ DPP-4 inhibitors were added in the first-line GLP1 receptor agonist strategy only if the GLP1 receptor agonist was discontinued due to adverse events.

For people developing ASCVD or CKD during follow-up, sulfonylureas, thiazolidinediones, and dipeptidyl peptidase-4 inhibitors were discontinued, and whenever HbA1c was at least 7.0%, treatment progression consisted of 1–2) SGLT2 inhibitors and GLP1 receptor agonists with 50/50 probability until both were used; 3–4) sulfonylureas or thiazolidinediones (with 50/50 probability) until both options were used; and 5) insulin following the ADA/EASD guideline strategy. In all scenarios, an estimated glomerular filtration rate (eGFR) less than 30 mL/min/1.73 m2 would lead to SGLT2 inhibitor discontinuation, and end-stage renal disease would lead to discontinuation of all medications except insulin.

First-Line SGLT2 (SGLT2 Strategy)

For comparison, whenever HbA1c was at least 7.0% (Figure), treatment progression consisted of an initial SGLT2 inhibitor and then metformin followed as described in the baseline strategy without subsequent SGLT2 inhibitors. For patients with or developing ASCVD or CKD and HbA1c of at least 7.0%, SGLT2 inhibitors would be added and progression consisted of 1) metformin, 2) GLP1 receptor agonists, and 3) treatment options following the remaining ADA/EASD guideline strategy for patients with ASCVD.

First-Line Injectable GLP1 (Injectable GLP1 Strategy)

Whenever HbA1c increased to at least 7.0% (Figure), treatment consisted of initial GLP1 receptor agonists and then metformin followed as described in the baseline strategy without subsequent GLP1 receptor agonists. For patients with HbA1c of at least 7.0% or developing ASCVD or CKD, GLP1 receptor agonists would be started, followed by 1) metformin, 2) SGLT2 inhibitors, and 3) the remaining ADA/EASD guideline strategy for patients with ASCVD.

Population Simulation Model

Our model was based on diabetes-related complication and mortality modules from the United Kingdom Prospective Diabetes Study Outcomes Model version 2 (UKPDS OM2) (Supplement Figure 1, available at Annals.org) with additional hypoglycemic event, quality of life (utility), and U.S. cost modules (25–29). Used in nearly all major type 2 diabetes simulation models (27–29), the UKPDS outcomes model equations have been internally (26) and externally validated (28).

Based on 26 individual-level characteristics, our model predicts the lifetime risk for diabetes-related microvascular complications (foot ulcer, blindness, renal failure, first and subsequent amputation) and macrovascular complications (first and subsequent myocardial infarction, ischemic heart disease, first and subsequent stroke) using 13 risk equations, and mortality using 4 risk equations. The UKPDS OM2 assumes that the mortality risk increases after all diabetic complications (except foot ulcer and blindness). The model predicts nondiabetes-related and diabetes-related death. We have internally validated our version of the UKPDS OM2 (24).

We modified the model in several ways. In addition to updating the medication module (30), we added the effects of GLP1 receptor agonists or SGLT2 inhibitors on systolic blood pressure, low-density lipoprotein, and weight (Table 1) and also added GLP1 receptor agonist effects on heart rate and SGLT2 inhibitor effects on eGFR. These estimated changes in cardiovascular risk factors were based on findings from our meta-analyses of these drugs versus placebo (28 trials for SGLT2 inhibitors; 17 trials for GLP1 receptor agonists) (31, 32). We assumed that these cardiovascular risk factor changes would only be present while patients were taking these medications. Because patients in our meta-analyses and in the UKPDS were already taking additional cardiovascular medications, we did not incorporate additional changes in cardiovascular risk factors for patients taking other diabetic drugs (for example, metformin, sulfonylureas) to avoid double-counting benefit. Third, our model includes GLP1 receptor agonists (31, 32) or metformin treatment discontinuation using predictive models based on our meta-analysis results for diabetic ketoacidosis and genital infections for SGLT2 inhibitors and gastrointestinal events (nausea, vomiting, or diarrhea).

Table 1.

Meta-Analytic* Effects of SGLT2 Inhibitors and GLP1 Receptor Agonists on Cardiovascular Risk Factors, Diabetic Complications, Mortality, and Adverse Events†

Variable Mean Difference (95% CI)/Risk Difference (95% CI)
SGLT2 Inhibitors vs. Placebo GLP1 Receptor Agonistsvs. Placebo
Cardiovascular risk factors
 Weight, kg −2.02 (−2.22 to −1.82) −1.84 (−2.37 to −1.30)
 Systolic blood pressure, mm Hg −3.62 (−4.22 to −3.01) −1.75 (−2.14 to −1.35)
 Heart rate, beats/min No effect 2.22 (1.69 to 2.75)
 Low-density lipoprotein
  mmol/L No effect −0.04 (−0.06 to −0.03)
  mg/dL No effect −1.55 (−2.32 to −1.16)
 High-density lipoprotein
  mmol/L 0.05(0.04 to 0.06) No effect
  mg/dL 1.93(1.55 to 2.32) No effect
 eGFR, mL/min/1.73 m2 −1.41 (−1.98 to −0.84) No effect
Macrovascular outcomes
 Stroke No effect −0.002 (−0.002 to −0.001)
 Any myocardial infarction −0.001 (−0.002 to −0.00004) No effect
 Heart failure −0.004 (−0.004 to −0.003) No effect
Microvascular outcomes
 End-stage renal disease No effect −0.003 (−0.003 to −0.002)
Mortality
 All-cause mortality −0.001 (−0.002 to −0.0006) −0.002 (−0.004 to −0.001)
Adverse events
 Any gastrointestinal event No effect 0.031 (0.014 to 0.058)
 Severe hypoglycemia −0.001 (−0.002 to −0.0001) No effect
 Diabetic ketoacidosis 0.0005(0.0001 to 0.001) No effect
 Genital yeast infection 0.043 (0.029 to 0.060) No effect
Open in a new tab

eGFR = estimated glomerular filtration rate; GLP1 = glucagon-like peptide-1; SGLT2 = sodium–glucose cotransporter-2.

*

From references 31 and 32.

†

Cardiovascular risk factor changes and adverse events were added to the simulation model in the main cost-effectiveness analyses. In sensitivity analyses, additional effects on macrovascular and microvascular events and mortality were added to the simulation modeling to estimate the maximal potential effect of SGLT2 inhibitors and GLP1 receptor agonists.

Model Inputs

We created a nationally representative sample of persons with diabetes eligible to start first-line therapy using participants who self-reported diabetes or had HbA1c of more than 6.5% within the National Health and Nutrition Examination Survey (NHANES) from 2013 to 2016 (Supplement Figure 2, available at Annals.org) (33). We excluded participants with suspected autoimmune diabetes (for example, age at diagnosis <18 years or started insulin within 1 year of diagnosis). We only included participants who were not taking diabetes medications. For missing values that were necessary for the model, we took the average of 5 values generated using Markov chain Monte Carlo multiple imputation (34). Medical examination subsample weights were used to obtain national population estimates. We excluded participants with 4 or more missing model inputs.

Individual descriptive characteristics from NHANES were used as baseline data, which included age, sex, race, smoking status, diabetes duration, and self-reported medical conditions (coronary heart disease, angina, heart failure, myocardial infarction, stroke, dialysis in the last year, and retinopathy). Biomarkers included HbA1c, systolic blood pressure, eGFR, low-density lipoprotein, high-density lipoprotein, body mass index, hemoglobin, leukocyte count, heart rate, and albuminuria. For medical histories unavailable in NHANES (history of peripheral vascular disease, atrial fibrillation, amputation, blindness, and peripheral neuropathy), we imputed age- and sex-based probabilities for peripheral vascular disease and atrial fibrillation (35, 36) and assumed that no persons had amputations, blindness, or peripheral neuropathy at baseline.

Costs and Health Utility

We calculated costs associated with medication use, complications, hospital utilization, and self-monitoring (testing and supplies) (37–39). Medication costs were calculated using publicly available Medicare costs across drug classes (40). All costs were from a health care sector perspective and in 2019 U.S. dollars (Supplement Table 1, available at Annals.org).

To account for quality-of-life reductions due to complications from disease or treatment, we calculated quality-adjusted life expectancy (QALE) using a health utility assessment where 0 equates to death and 1 equates to perfect health. Using established utility reductions in the literature, the model incorporated quality-of-life reductions for any diabetic complication and use of glucose-lowering medications (oral and injectable) as quality-adjusted life-years (QALYs) (Supplement Table 1) (41–51). For persons developing more than 1 complication, we multiplied all utility values for that year when calculating the QALY (52).

Main Analysis

For the 3 initial treatment approaches—metformin, SGLT2 inhibitors, and injectable GLP1 receptor agonists, the model ran 2500 microsimulations for each NHANES participant and captured the average population lifetime diabetic complications rates, life expectancy, QALYs, and costs. Costs and QALYs were discounted at 3% annually. An efficiency frontier was also constructed.

The modeling incorporated patient-level heterogeneity as a first-order Monte Carlo, that is, accounting for the entire distribution of patient risk factors and their potential nonlinear effects on model outcomes. As such, the results reflect stochastic uncertainty (2500 runs for each person) and methodological uncertainty (threshold and scenario and sensitivity analyses, described in the next section) but not distributions of uncertainty around cost or utility variables or the risk equation coefficients in the UKPDS OM2 risk equations (53). Models were run using R 4.0.0 (54). The NHANES analyses were conducted using SAS version 9.4 (55). The Consolidated Health Economic Evaluation Reporting Standards (CHEERS) checklist and impact inventory are available in Supplement Tables 2 and 3 (available at Annals.org).

Statistical Analysis

Threshold and Scenario Analyses

The robustness of results was evaluated using several additional analyses.We performed a threshold analysis to determine the hypothetical medication costs required for a strategy to become cost-effective assuming a willingness-to-pay threshold of under $150 000 per QALY.

We conducted many additional scenario analyses including the following.

Maximal Treatment Benefits.

Our base case assumes that risk factor improvements mediate event rate reductions, so in this scenario, we assumed additional reductions in events according to our drug versus placebo meta-analyses, potentially double-counting benefits (Table 1)(30, 31).

No Medication Disutility.

In this scenario analysis, we assumed no disutility from taking any diabetes medications when estimating quality of life.

Orally Administered GLP1 Receptor Agonists.

For oral semaglutide, we assumed that the effectiveness and adverse events were identical to injectable GLP1 receptor agonists but included its higher cost and excluded the injection-associated disutility.

Medication cost estimates from the Federal Supply Schedule (FSS) (56) and Average Wholesale Price (AWP) with a 15% discount (39) were also included. Cost inputs are available in Supplement Table 4 (available at Annals.org).

Sensitivity Analyses

One-way deterministic sensitivity analyses were performed to analyze the effects of changes in discount rate (0% or 6%) and cost (±25%) variables, HbA1c thresholds for changing treatment (<6.5% or <8%), and the upper and lower bounds of medication effects (Table 1) on cardiovascular risk factors.

Role of the Funding Source

The funding sources had no role in the design and conduct of the study; collection, management, analysis, or interpretation of the data; or preparation, review, or approval of the manuscript.

Results

In NHANES 2013 to 2016, 493 participants met inclusion and exclusion criteria, representing about 7.3 million U.S. adults aged 18 years and older with likely type 2 diabetes and not reporting any diabetes medications (Supplement Table 5, available at Annals.org). The average age was 55 years and 50% were female. Participants had diabetes for an average of 4.2 years; 36% had a history of diabetic complications.

In the lifetime simulation modeling, the first-line metformin strategy had a cumulative absolute reduction in macrovascular complications (congestive heart failure, ischemic heart disease, myocardial infarction, and stroke) of 4.4% to 5.2% for SGLT2 inhibitors and GLP1 receptor agonists (Table 2) but only a 0.1% benefit in reduced microvascular outcomes. The unweighted results are available in Supplement Table 6 (available at Annals.org).

Table 2.

Estimated Lifetime Complication Rates due to First-Line Metformin, SGLT2 Inhibitor, and GLP1 Receptor Agonist Strategies for the U.S. Population With Type 2 Diabetes

Complication, % First-Line Metformin First-Line SGLT2 Inhibitor First-Line GLP1 Receptor Agonist
Macrovascular outcomes
 Heart failure 14.2 13.2 13.1
 Ischemic heart disease 18.8 17.3 17.2
 Myocardial infarction 27.0 26.0 25.5
 Stroke 17.2 16.3 16.2
Microvascular outcomes
 Amputation 3.5 3.5 3.5
 Blindness 11.5 11.4 11.5
 End-stage renal disease 2.6 2.6 2.5
Adverse events
 Foot ulcer 2.9 2.9 2.9
Open in a new tab

GLP1 = glucagon-like peptide-1; SGLT2 = sodium–glucose cotransporter-2.

First-line SGLT2 inhibitors or GLP1 receptor agonists increased undiscounted life expectancy by 3.0 months and 3.4 months, respectively, versus first-line metformin (Table 3). The first-line SGLT2 inhibitor increased discounted QALE by 33 days, but GLP1 receptor agonists decreased discounted QALE by 22 days, respectively, versus first-line metformin, due to discounting and treatment-associated disutility.

Table 3.

Comparison of Estimated LE, QALE, Cost, ICER, and Scenario Analyses for First-Line Metformin, SGLT2 Inhibitor, and GLP1 Receptor Agonist Strategies for the U.S. Population With Type 2 Diabetes

Variable LE, y Incremental LE, y QALE, y Incremental QALE, y Lifetime Cost, U.S. Dollars* Incremental Cost, U.S. Dollars* ICER vs. Metformin, U.S. Dollars*
Base case
 Metformin 24.33 - 12.75 - 145 000 - -
 SGLT2 24.58 0.25 12.84 0.09 188 000 43 000 478 000
 GLP1 24.61 0.28 12.69 −0.06 194 000 49 000 Inferior†
Orallyadministered GLP1 receptor agonists
 Metformin 24.33 - 12.75 - 148 000 - -
 SGLT2 24.58 0.25 12.84 0.09 191 000 43 000 478 000
 GLP1 24.61 0.28 13.01 0.26 362 000 214000 823 000
No medication disutility
 Metformin 24.33 - 13.17 - 145 000 - -
 SGLT2 24.58 0.25 13.31 0.14 188 000 43 000 307 000
 GLP1 24.61 0.28 13.32 0.15 194 000 49 000 327 000
Additional treatment benefits beyond risk-factor changes
 Metformin 24.90 - 12.98 - 217 000 - -
 SGLT2 25.46 0.56 13.21 0.23 300 000 83 000 361 000
 GLP1 25.60 0.7 13.11 0.13 323 000 106 000 815 000
Federal SupplySchedule
 Metformin 24.33 - 12.75 - 135 000 - -
 SGLT2 24.58 0.25 12.84 0.09 167 000 32 000 316 000
 GLP1 24.61 0.28 12.69 -0.06 179 000 44 000 Inferior†
AverageWholesale Price (15% discount)
 Metformin 24.33 - 12.75 - 99 000 - -
 SGLT2 24.58 0.25 12.84 0.09 132 000 33 000 296 000
 GLP1 24.61 0.28 12.69 −0.06 110 000 11 000 Inferior†
Open in a new tab

GLP1 = glucagon-like peptide-1; ICER = incremental cost-effectiveness ratio comparing each strategy to baseline metformin strategy; LE = life expectancy; QALE = quality-adjusted life expectancy; SGLT2 = sodium–glucose cotransporter-2 inhibitor.

*

Costs are discounted at 3%, and ICER is based on discounted costs.

†

These strategies were dominated because they were more expensive and had lower QALE versus metformin.

First-line metformin had the lowest average per-person lifetime cost and GLP1 receptor agonists the highest (Table 3) with 60% to 70% of the cost difference being due to medication costs (initial metformin: $92 000 vs. $135 000 for the SGLT2 inhibitor and $141 000 for the GLP1 receptor agonist). Stroke complications represented 10% to 14% of costs (about $20 000) for each strategy. The incremental cost-effectiveness ratio (ICER) of SGLT2 inhibitors was $478 000 per QALY gained compared with metformin, and injectable GLP1 receptor agonists cost more and shortened QALE (that is, was inferior to metformin) (Table 3).

An efficiency frontier identifies the most efficient use of resources when considering first-line metformin, SGLT2 inhibitor, and injectable and oral GLP1 receptor agonist (Supplement Figure 3, available at Annals.org). Beyond metformin, no strategies fell below the willingness-to-pay threshold of $150 000 per QALY. The SGLT2 inhibitors had an ICER of $478 000 per QALY compared with metformin; oral GLP1 receptor agonists had an ICER of $1 024 000 per QALY compared with SGLT2 inhibitors. First-line injectable GLP1 receptor agonists cost more and yielded lower QALYs.

Threshold Analyses

Based on a willingness-to-pay threshold of under $150 000 per QALY, SGLT2 inhibitors and oral GLP1 receptors would become cost-effective at $1800 per year ($5 per day) and $2100 per year ($6 per day), respectively, requiring cost reductions of at least 70% for SGLT2 inhibitors and at least 90% for oral GLP1 receptor agonists.

Scenario Analyses

The oral GLP1 receptor agonist (semaglutide) increased QALE by 3 months but had an ICER of $823 000 per additional QALY gained versus metformin (Table 3). Excluding medication disutility, first-line SGLT2 inhibitors or GLP1 receptor agonists remained expensive with ICERs exceeding $300 000 per QALY (Table 3). When adding beneficial changes in event rates, first-line SGLT2 inhibitors or injectable GLP1 receptor agonists increased QALY versus metformin by 2.8 months and 1.6 months, respectively, but ICERs versus metformin remained high at $361 000 per QALY for SGLT2 inhibitors and $815 000 per QALY for GLP1 receptor agonists (Table 3). With either the lower FSS or discounted AWP costs, first-line injectable GLP1 receptor agonists remained more expensive with a lower QALY. First-line SGLT2 inhibitors versus metformin using FSS or discounted AWP yielded ICERs of $316 000 and $296 000 per QALY gained, respectively (Table 3).

Sensitivity Analyses

In all sensitivity analyses, ICERs for SGLT2 inhibitors exceeded $200 000 per QALY compared with metformin even when reducing the discount rate to 0%, the SGLT2 inhibitor costs by 25%, or varying SGLT2 inhibitor effectiveness to its upper bound. The GLP1 receptor agonists were either dominated or had ICERs under $150 000.

Discussion

Our model results suggest that first-line SGLT2 inhibitors and GLP1 receptor agonists are not cost-effective for the U.S. population, compared with the 2021 ADA/EASD guidelines, primarily due to their high medication costs. The model projected improved life expectancy by 3.0 months to 3.4 months due to reduced rates of macrovascular disease from first-line SGLT2 inhibitors and GLP1 receptor agonists (24.4% to 25.2% cumulative absolute reduction). However, the current drug costs would be too high to encourage their adoption as first line for usual clinical practice, requiring a price reduction of 70% for SGLT2 inhibitors and 90% for oral GLP1 receptor agonists. Based on our scenario analyses, eliminating the injection disutility for oral GLP1 receptor agonists, eliminating medication disutilities, modeling maximal treatment benefits, or identifying alternative medication price sources did not alter the absence of cost-effectiveness conclusions.

Diabetes drug prices, including generic prices for newer drugs, have been increasing (3, 57). Our threshold analyses suggest that the cost of the SGLT2 inhibitors and oral GLP1 receptor agonists must be reduced at least 70% and 90% to become cost-effective at a willingness-to-pay threshold of $150 000 per QALY. Generic SGLT2 inhibitors could enter the market (1 of 2 dapagliflozin patents died in October 2020) and approval for generic alternatives has been sought from the U.S. Food and Drug Administration. Although promising, it may take decades for medication prices to drop low enough to become affordable. For example, a generic GLP1 receptor agonist became available in 2017, but its costs remain high. Without external incentives, limited access to these drug classes will likely persist (for example, due to higher copays or requirements for prior authorizations) (58), as will further diabetes disparities—for decades into the future—because of differential access to care due to insurance (for example, private vs. public), which often tracks race and ethnicity (59–61).

The results of our simulation model and subgroup analyses provide greater clarity about the driving forces of the cost-effectiveness of the various treatment options. The major driving factors were their high costs and a decrease in quality of life associated with injections. The first-line SGLT2 inhibitor strategy and GLP1 receptor agonist strategies (both oral and injectable) had greater efficacy in terms of life expectancy and QALE compared with the first-line metformin strategy. Specifically, GLP1 receptor agonists were the most effective strategy to maximize life expectancy, but the injectable form of GLP1 receptor agonists reduced QALE. Our main analysis included the injection disutility as a nontrivial source of concern from the patients’ perspective (62).

To our knowledge, no other studies have compared the cost-effectiveness of using GLP1 receptor agonists or SGLT2 inhibitors as first-line therapy versus the 2021 ADA guidelines. Other recently published U.S. cost-effectiveness analyses have examined drug classes rather than medication algorithms (63–65). Therefore, our conclusions that SGLT2 inhibitors and GLP1 receptor agonists as first-line therapy may not be cost-effective despite improved clinical outcomes solely reflect the incremental benefit of using the newer agents as first-line agents rather than second-line, that is, a delayed use of the newer agents.

Our modeling analysis has limitations. First, there are known limitations of the UKPDS model, which include the overestimation of macrovascular complications (66–68). This limitation could potentially lead to an artificial increase in the costs and disutilities associated with these complications. However, these issues are present in all comparisons, which minimize their effect. Second, incomplete demographic and heart failure history and a limited sample of NHANES participants precluded additional analyses. In addition, the NHANES population not currently receiving diabetic medications is a mixed population of newly diagnosed diabetes and diabetes for approximately 10 years. The latter population, with an extensive history of diabetes without self-reported diabetes medication, may be a socioeconomically disadvantaged group, which could affect the generalizability of not only our results but also a population of interest given the economic barrier to adequate treatment. Furthermore, the use of Medicare drug costs biases the analysis against the SGLT2 inhibitor and the GLP1 receptor agonist drug classes (Table 3). However, the SGLT2 inhibitor and GLP1 receptor agonist strategies remained not cost-effective regardless of the medication cost source in our scenario analyses. Lastly, our analyses focus on the U.S. population and costs and cannot be generalized to other countries.

Due to their associated lower risk for macrovascular events, SGLT2 inhibitors and GLP1 receptor agonists are potentially effective first-line agents to treat type 2 diabetes. However, these drug classes are not cost-effective at their current costs. In the interest of improving access to high-quality care in the United States, our study results indicate the need to reduce SGLT2 inhibitor and GLP1 receptor agonist medication costs substantially for patients with type 2 patients to improve health outcomes and prevent exacerbating diabetes health disparities.

Supplementary Material

Supplement

Grant Support:

This research was supported by the American Diabetes Association (1-18-JDF-037). Dr. Choi was supported by National Institutes of Health National Institute on Aging grant 5T35AG029795-13. Drs. Laiteerapong, Huang, Wan, Staab, and Franco are members of the National Institute of Diabetes and Digestive and Kidney Diseases Chicago Center for Diabetes Translation Research at the University of Chicago (P30DK092949).

Footnotes

Reproducible Research Statement: Study protocol and Data set: Not available. Statistical code: Available to interested readers by contacting Dr. Laiteerapong at nlaiteer@medicine.bsd.uchicago.edu.

No other persons contributed to the scientific content or provided technical support.

Author contributions are available at Annals.org.

Contributor Information

Jin G. Choi, Section of General Internal Medicine, Department of Medicine, University of Chicago, Chicago, Illinois.

Aaron N. Winn, Medical College of Wisconsin, Milwaukee, Wisconsin.

M. Reza Skandari, Centre for Health Economics & Policy Innovation, Imperial College Business School, Imperial College London, London, United Kingdom.

Melissa I. Franco, Section of General Internal Medicine, Department of Medicine, University of Chicago, Chicago, Illinois.

Erin M. Staab, Section of General Internal Medicine, Department of Medicine, University of Chicago, Chicago, Illinois.

Jason Alexander, Section of General Internal Medicine, Department of Medicine, University of Chicago, Chicago, Illinois.

Wen Wan, Section of General Internal Medicine and Center for Chronic Disease Research and Policy, Department of Medicine, University of Chicago, Chicago, Illinois.

Mengqi Zhu, Section of General Internal Medicine, Department of Medicine, University of Chicago, Chicago, Illinois.

Elbert S. Huang, Section of General Internal Medicine and Center for Chronic Disease Research and Policy, Department of Medicine, University of Chicago, Chicago, Illinois.

Louis Philipson, Sections of Adult and Pediatric Endocrinology, Diabetes & Metabolism, Department of Medicine, University of Chicago, Chicago, Illinois.

Neda Laiteerapong, Section of General Internal Medicine and Center for Chronic Disease Research and Policy, Department of Medicine, University of Chicago, Chicago, Illinois.

References

  • 1.Saeedi P, Petersohn I, Salpea P, et al. ; IDF Diabetes Atlas Committee. Global and regional diabetes prevalence estimates for 2019 and projections for 2030 and 2045: results from the International Diabetes Federation Diabetes Atlas, 9th edition. Diabetes Res Clin Pract. 2019;157:107843. doi: 10.1016/j.diabres.2019.107843 [DOI] [PubMed] [Google Scholar]
  • 2.Centers for Disease Control and Prevention. National Diabetes Statistics Report. U.S. Department of Health & Human Services; 2020. [Google Scholar]
  • 3.American Diabetes Association. Economic costs of diabetes in the U.S. in 2017. Diabetes Care. 2018;41:917–928. doi: 10.2337/dci18-0007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Zhuo X, Zhang P, Kahn HS, et al. Change in medical spending attributable to diabetes: national data from 1987 to 2011. Diabetes Care. 2015;38:581–7. doi: 10.2337/dc14-1687 [DOI] [PubMed] [Google Scholar]
  • 5.Marcum ZA, Driessen J, Thorpe CT, et al. Regional variation in use of a new class of antidiabetic medication among Medicare beneficiaries: the case of incretin mimetics. Ann Pharmacother. 2015;49:285–92. doi: 10.1177/1060028014563951 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Brown P New diabetes drugs: better outcomes at a hefty cost. Medpage Today; 27 June 2016. Accessed at www.medpagetoday.com/cardiology/prevention/58780 on 1 October 2020.
  • 7.National Institute for Health and Care Excellence (NICE). Canagliflozin, dapagliflozin, and empagliflozin as monotherapies for treating type 2 diabetes. Technology appraisal guidance TA390. NICE; 25 May 2016. [Google Scholar]
  • 8.Zinman B, Wanner C, Lachin JM, et al. ; EMPA-REG OUTCOME Investigators. Empagliflozin, cardiovascular outcomes, and mortality in type 2 diabetes. N Engl J Med. 2015;373:2117–28. doi: 10.1056/NEJMoa1504720 [DOI] [PubMed] [Google Scholar]
  • 9.Neal B, Perkovic V, Mahaffey KW, et al. ; CANVAS Program Collaborative Group. Canagliflozin and cardiovascular and renal events in type 2 diabetes. N Engl J Med. 2017;377:644–657. doi: 10.1056/NEJMoa1611925 [DOI] [PubMed] [Google Scholar]
  • 10.Wiviott SD, Raz I, Bonaca MP, et al. ; DECLARE–TIMI 58 Investigators. Dapagliflozin and cardiovascular outcomes in type 2 diabetes. N Engl J Med. 2019;380:347–357. doi: 10.1056/NEJMoa1812389 [DOI] [PubMed] [Google Scholar]
  • 11.Marso SP, Bain SC, Consoli A, et al. ; SUSTAIN-6 Investigators. Semaglutide and cardiovascular outcomes in patients with type 2 diabetes. N Engl J Med. 2016;375:1834–1844. doi: 10.1056/NEJMoa1607141 [DOI] [PubMed] [Google Scholar]
  • 12.Marso SP, Daniels GH, Brown-Frandsen K, et al. ; LEADER Steering Committee. Liraglutide and cardiovascular outcomes in type 2 diabetes. N Engl J Med. 2016;375:311–22. doi: 10.1056/NEJMoa1603827 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Pfeffer MA, Claggett B, Diaz R, et al. ; ELIXA Investigators. Lixisenatide in patients with type 2 diabetes and acute coronary syndrome. N Engl J Med. 2015;373:2247–57. doi: 10.1056/NEJMoa1509225 [DOI] [PubMed] [Google Scholar]
  • 14.Gerstein HC, Colhoun HM, Dagenais GR, et al. ; REWIND Investigators. Dulaglutide and cardiovascular outcomes in type 2 diabetes (REWIND): a double-blind, randomised placebo-controlled trial. Lancet. 2019;394:121–130. doi: 10.1016/S0140-6736(19)31149-3 [DOI] [PubMed] [Google Scholar]
  • 15.Holman RR, Bethel MA, Mentz RJ, et al. ; EXSCEL Study Group. Effects of once-weekly exenatide on cardiovascular outcomes in type 2 diabetes. N Engl J Med. 2017;377:1228–1239. doi: 10.1056/NEJMoa1612917 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Hyperglycemia Laakso M. and cardiovascular disease in type 2 diabetes. Diabetes. 1999;48:937–42. doi: 10.2337/diabetes.48.5.937 [DOI] [PubMed] [Google Scholar]
  • 17.Buse JB, Wexler DJ, Tsapas A, et al. 2019 update to: management of hyperglycemia in type 2 diabetes, 2018. A consensus report by the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetes Care. 2020;43:487–493. doi: 10.2337/dci19-0066 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Davies MJ, D’Alessio DA, Fradkin J, et al. Management of hyperglycemia in type 2 diabetes, 2018. A consensus report by the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetes Care. 2018;41:2669–2701. doi: 10.2337/dci18-0033 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.American Diabetes Association. 9. Pharmacologic approaches to glycemic treatment: standards of medical care in diabetes-2021. Diabetes Care. 2021;44:S111–S124. doi: 10.2337/dc21-S009 [DOI] [PubMed] [Google Scholar]
  • 20.Marx N, Grant PJ, Cosentino F. Compelling evidence for SGLT2 inhibitors and GLP-1 receptor agonists as first-line therapy in patients with diabetes at very high/high cardiovascular risk. Eur Heart J. 2020;41:329–330. doi: 10.1093/eurheartj/ehz853 [DOI] [PubMed] [Google Scholar]
  • 21.Zaccardi F, Khunti K, Marx N, et al. First-line treatment for type 2 diabetes: is it too early to abandon metformin. Lancet. 2020;396:1705–1707. doi: 10.1016/S0140-6736(20)32523-X [DOI] [PubMed] [Google Scholar]
  • 22.Bin Hussain AK, Abdelgadir E, Rashid F, et al. Should metformin still be the first-line of treatment in type 2 diabetes mellitus? A comprehensive review and suggested algorithm. Diabetes Metab Syndr. 2019;13:1935–1942. doi: 10.1016/j.dsx.2019.04.028 [DOI] [PubMed] [Google Scholar]
  • 23.Institute for Clinical and Economic Review (ICER). ICER Issues Final Report and Policy Recommendations for Add-on Therapies to Treat Type 2 Diabetes. ICER; 9 December 2019. [Google Scholar]
  • 24.Laiteerapong N, Cooper JM, Skandari MR, et al. Individualized glycemic control for U.S. adults with type 2 diabetes: a cost-effectiveness analysis. Ann Intern Med. 2018;168:170–178. doi: 10.7326/M17-0537 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Clarke PM, Gray AM, Briggs A, et al. ; UK Prospective Diabetes Study (UKDPS) Group. A model to estimate the lifetime health outcomes of patients with type 2 diabetes: the United Kingdom Prospective Diabetes Study (UKPDS) outcomes model (UKPDS no. 68). Diabetologia. 2004;47:1747–59. doi: 10.1007/s00125-004-1527-z [DOI] [PubMed] [Google Scholar]
  • 26.Hayes AJ, Leal J, Gray AM, et al. UKPDS outcomes model 2: a new version of a model to simulate lifetime health outcomes of patients with type 2 diabetes mellitus using data from the 30 year United Kingdom Prospective Diabetes Study: UKPDS 82. Diabetologia. 2013;56:1925–33. doi: 10.1007/s00125-013-2940-y [DOI] [PubMed] [Google Scholar]
  • 27.Palmer AJ, Clarke P, Gray A, et al. ; Mount Hood 5 Modeling Group. Computer modeling of diabetes and its complications: a report on the Fifth Mount Hood challenge meeting. Value Health. 2013;16:670–85. doi: 10.1016/j.jval.2013.01.002 [DOI] [PubMed] [Google Scholar]
  • 28.Lundqvist A, Steen Carlsson K, Johansen P, et al. Validation of the IHE Cohort Model of Type 2 Diabetes and the impact of choice of macrovascular risk equations. PLoS One. 2014;9:e110235. doi: 10.1371/journal.pone.0110235 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.McEwan P, Foos V, Palmer JL, et al. Validation of the IMS CORE diabetes model. Value Health. 2014;17:714–24. doi: 10.1016/j.jval.2014.07.007 [DOI] [PubMed] [Google Scholar]
  • 30.Inzucchi SE, Bergenstal RM, Buse JB, et al. Management of hyperglycemia in type 2 diabetes, 2015: a patient-centered approach: update to a position statement of the American Diabetes Association and the European Association for the Study of Diabetes. Diabetes Care. 2015;38:140–9. doi: 10.2337/dc14-2441 [DOI] [PubMed] [Google Scholar]
  • 31.Alexander JT, Staab EM, Wan W, et al. The longer-term benefits and harms of glucagon-like peptide-1 receptor agonists: a systematic review and meta-analysis. J Gen Intern Med. 2022;37:415–38. doi: 10.1007/s11606-021-07105-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Alexander JT, Staab EM, Wan W, et al. Longer-term benefits and risks of sodium-glucose cotransporter-2 inhibitors in type 2 diabetes: a systematic review and meta-analysis. J Gen Intern Med. 2022;37:439–48. doi: 10.1007/s11606-021-07227-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Centers for Disease Control and Prevention National Center for Health Statistics (CDC NCHS). National Health and Nutrition Examination Survey Data (NHANES). U.S. Department of Health & Human Services; 2016. [Google Scholar]
  • 34.Gelman A, Rubin DB. Markov chain Monte Carlo methods in biostatistics. Stat Methods Med Res. 1996;5:339–55. doi: 10.1177/096228029600500402 [DOI] [PubMed] [Google Scholar]
  • 35.Naccarelli GV, Varker H, Lin J, et al. Increasing prevalence of atrial fibrillation and flutter in the United States. Am J Cardiol. 2009;104:1534–9. doi: 10.1016/j.amjcard.2009.07.022 [DOI] [PubMed] [Google Scholar]
  • 36.Selvin E, Erlinger TP. Prevalence of and risk factors for peripheral arterial disease in the United States: results from the National Health and Nutrition Examination Survey, 1999–2000. Circulation. 2004;110:738–43. doi: 10.1161/01.CIR.0000137913.26087.F0 [DOI] [PubMed] [Google Scholar]
  • 37.Ward A, Alvarez P, Vo L, et al. Direct medical costs of complications of diabetes in the United States: estimates for event-year and annual state costs (USD 2012). J Med Econ. 2014;17:176–83. doi: 10.3111/13696998.2014.882843 [DOI] [PubMed] [Google Scholar]
  • 38.Javor KA, Kotsanos JG, McDonald RC, et al. Diabetic ketoacidosis charges relative to medical charges of adult patients with type I diabetes. Diabetes Care. 1997;20:349–54. doi: 10.2337/diacare.20.3.349 [DOI] [PubMed] [Google Scholar]
  • 39.Micromedex RED BOOK. Micromedex Healthcare Series [database online]. Truven Health Analytics. [Google Scholar]
  • 40.Centers for Medicare & Medicaid Services (CMS). Medicare Part D Spending by Drug. CMS; 1 December 2021. [Google Scholar]
  • 41.Harris S, Mamdani M, Galbo-Jørgensen CB, et al. The effect of hypoglycemia on health-related quality of life: Canadian results from a multinational time trade-off survey. Can J Diabetes. 2014;38:45–52. doi: 10.1016/j.jcjd.2013.09.001 [DOI] [PubMed] [Google Scholar]
  • 42.Clarke P, Gray A, Holman R. Estimating utility values for health states of type 2 diabetic patients using the EQ-5D (UKPDS 62). Med Decis Making. 2002;22:340–9. doi: 10.1177/0272989X0202200412 [DOI] [PubMed] [Google Scholar]
  • 43.Bagust A, Beale S. Modelling EuroQol health-related utility values for diabetic complications from CODE-2 data. Health Econ. 2005;14:217–30. doi: 10.1002/hec.910 [DOI] [PubMed] [Google Scholar]
  • 44.Wasserfallen JB, Halabi G, Saudan P, et al. Quality of life on chronic dialysis: comparison between haemodialysis and peritoneal dialysis. Nephrol Dial Transplant. 2004;19:1594–9. doi: 10.1093/ndt/gfh175 [DOI] [PubMed] [Google Scholar]
  • 45.Currie CJ, Morgan CL, Poole CD, et al. Multivariate models of health-related utility and the fear of hypoglycaemia in people with diabetes. Curr Med Res Opin. 2006;22:1523–34. doi: 10.1185/030079906X115757 [DOI] [PubMed] [Google Scholar]
  • 46.Peasgood T, Brennan A, Mansell P, et al. The impact of diabetes-related complications on preference-based measures of health-related quality of life in adults with type I diabetes. Med Decis Making. 2016;36:1020–33. doi: 10.1177/0272989X16658660 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Rajan N, Boye KS, Gibbs M, et al. Utilities for type 2 diabetes treatment-related attributes in a South Korean and Taiwanese population. Value Health Reg Issues. 2016;9:67–71. doi: 10.1016/j.vhri.2015.11.006 [DOI] [PubMed] [Google Scholar]
  • 48.Coffey JT, Brandle M, Zhou H, et al. Valuing health-related quality of life in diabetes. Diabetes Care. 2002;25:2238–43. doi: 10.2337/diacare.25.12.2238 [DOI] [PubMed] [Google Scholar]
  • 49.National Center for Health Statistics (NCHS). National Health Interview Survey Diabetes Supplement. NCHS; 2006. [Google Scholar]
  • 50.Centers for Medicare & Medicaid Services (CMS). 2012 Physician Fee Schedule. Accessed at www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/PhysicianFeeSched/PFS-Relative-Value-Files-Items/CMS1254038 on 10 November 2017.
  • 51.Beaudet A, Clegg J, Thuresson PO, et al. Review of utility values for economic modeling in type 2 diabetes. Value Health. 2014;17:462–70. doi: 10.1016/j.jval.2014.03.003 [DOI] [PubMed] [Google Scholar]
  • 52.Hanmer J, Lawrence WF, Anderson JP, et al. Report of nationally representative values for the noninstitutionalized US adult population for 7 health-related quality-of-life scores. Med Decis Making. 2006;26:391–400. doi: 10.1177/0272989X06290497 [DOI] [PubMed] [Google Scholar]
  • 53.Dakin HA, Leal J, Briggs A, et al. Accurately reflecting uncertainty when using patient-level simulation models to extrapolate clinical trial data. Med Decis Making. 2020;40:460–473. doi: 10.1177/0272989X20916442 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.R Core Team. R: A language and environment for statistical computing. 4.0.0 ed. R Foundation for Statistical Computing; 2010. Accessed at www.R-project.org. [Google Scholar]
  • 55.SAS Institute. The SAS system for Windows. 9.4 ed. SAS Institute; 2013. [Google Scholar]
  • 56.Office of Procurement, Acquisition and Logistics (OPAL) U.S. Department of Veterans Affairs (VA). VA Federal Supply Schedule Service. VA; 2021. [Google Scholar]
  • 57.McEwen LN, Casagrande SS, Kuo S, et al. Why are diabetes medications so expensive and what can be done to control their cost. Curr Diab Rep. 2017;17:71. doi: 10.1007/s11892-017-0893-0 [DOI] [PubMed] [Google Scholar]
  • 58.McCoy RG, Van Houten HK, Deng Y, et al. Comparison of diabetes medications used by adults with commercial insurance vs Medicare Advantage, 2016 to 2019. JAMA Netw Open. 2021;4:e2035792. doi: 10.1001/jamanetworkopen.2020.35792 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Manuel JI. Racial/ethnic and gender disparities in health care use and access. Health Serv Res. 2018;53:1407–1429. doi: 10.1111/1475-6773.12705 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Hu R, Shi L, Rane S, et al. Insurance, racial/ethnic, SES-related disparities in quality of care among US adults with diabetes. J Immigr Minor Health. 2014;16:565–75. doi: 10.1007/s10903-013-9966-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61.Basu J, Hanchate A, Bierman A. Racial/ethnic disparities in readmissions in US hospitals: the role of insurance coverage. Inquiry. 2018;55:46958018774180. doi: 10.1177/0046958018774180 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62.Boye KS, Matza LS, Walter KN, et al. Utilities and disutilities for attributes of injectable treatments for type 2 diabetes. Eur J Health Econ. 2011;12:219–30. doi: 10.1007/s10198-010-0224-8 [DOI] [PubMed] [Google Scholar]
  • 63.Pawaskar M, Bilir SP, Kowal S, et al. Cost-effectiveness of intensification with SGLT2 inhibitors for type 2 diabetes. Am J Manag Care. 2021;27:e269–e277. doi: 10.37765/ajmc.2021.88728 [DOI] [PubMed] [Google Scholar]
  • 64.Ehlers LH, Lamotte M, Monteiro S, et al. The cost-effectiveness of empagliflozin versus liraglutide treatment in people with type 2 diabetes and established cardiovascular disease. Diabetes Ther. 2021;12:1523–1534. doi: 10.1007/s13300-021-01040-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 65.Hung A, Jois B, Lugo A, et al. Cost-effectiveness of diabetes treatment sequences to inform step therapy policies. Am J Manag Care. 2020;26:e76–e83. doi: 10.37765/ajmc.2020.42639 [DOI] [PubMed] [Google Scholar]
  • 66.Pagano E, Konings SRA, Di Cuonzo D, et al. Prediction of mortality and major cardiovascular complications in type 2 diabetes: external validation of UK Prospective Diabetes Study outcomes model version 2 in two European observational cohorts. Diabetes Obes Metab. 2021;23:1084–1091. doi: 10.1111/dom.14311 [DOI] [PubMed] [Google Scholar]
  • 67.Tao L, Wilson EC, Griffin SJ, et al. ; ADDITION-Europe study team. Performance of the UKPDS outcomes model for prediction of myocardial infarction and stroke in the ADDITION-Europe trial cohort. Value Health. 2013;16:1074–80. doi: 10.1016/j.jval.2013.06.001 [DOI] [PubMed] [Google Scholar]
  • 68.van Dieren S, Peelen LM, Nöthlings U, et al. External validation of the UK Prospective Diabetes Study (UKPDS) risk engine in patients with type 2 diabetes. Diabetologia. 2011;54:264–70. doi: 10.1007/s00125-010-1960-0 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement
NIHMS1891477-supplement-Supplement.pdf (489.4KB, pdf)

RESOURCES