What Do the Health-Economic Analyses of DEVOTE Tell Us About Degludec's Real-World Value?

At a glance

• Trial: DEVOTE (Trial Comparing Cardiovascular Safety of Insulin Degludec vs Insulin Glargine in Patients with Type 2 Diabetes at High Risk of Cardiovascular Events)
• N: 7,637 patients with type 2 diabetes and established cardiovascular disease or cardiovascular risk factors
• Intervention: Insulin degludec (Tresiba) once daily, treat-to-target
• Comparator: Insulin glargine U100 (Lantus) once daily, treat-to-target
• Duration: Median 1.99 years (event-driven design)
• Primary endpoint: First occurrence of MACE (cardiovascular death, nonfatal MI, nonfatal stroke)
• Key result: HR 0.91 (95% CI, 0.78 to 1.06) for MACE, confirming non-inferiority; 53% lower rate of severe nocturnal hypoglycemia (p = 0.002)

Why an Economic Layer Matters for DEVOTE

The DEVOTE cardiovascular outcomes trial resolved a regulatory question: insulin degludec does not increase cardiovascular risk compared with glargine U100. But for clinicians and formulary committees, the follow-up question was immediate. Degludec carries a higher wholesale acquisition cost (WAC) than glargine U100 and biosimilar glargine options. Is the hypoglycemia reduction large enough to justify the price gap?

Several groups modeled this question using DEVOTE's patient-level data, country-specific unit costs, and microsimulation or Markov frameworks. The answer, as with most cost-effectiveness analyses (CEAs), is conditional on the price inputs. Understanding those conditions matters more than memorizing a single incremental cost-effectiveness ratio (ICER).

The DEVOTE Hypoglycemia Signal That Drives the Economics

DEVOTE's primary publication reported two hypoglycemia findings that feed directly into economic models:

• Severe hypoglycemia (overall): Rate ratio 0.60 (95% CI, 0.44 to 0.82) favoring degludec
• Severe nocturnal hypoglycemia: Rate ratio 0.47 (95% CI, 0.31 to 0.73) favoring degludec

These are clinically meaningful reductions in a high-risk population where roughly 40% of patients had a prior MI and the mean age was 65 years. Severe hypoglycemia in this population carries direct costs (emergency department visits, hospitalizations, ambulance calls) and indirect costs (lost productivity, caregiver burden, downstream fear-driven suboptimal dosing).

A secondary analysis from DEVOTE (DEVOTE 3) examined the temporal relationship between severe hypoglycemia and subsequent cardiovascular events, finding a statistically significant association between severe hypoglycemia and all-cause mortality in the following 15 to 365 days (HR 2.51). This association does not prove causation, but it provides biological plausibility for models that assign downstream cardiovascular cost offsets to hypoglycemia prevention.

Published Cost-Effectiveness Models

The Novo Nordisk-Sponsored IQVIA Core Diabetes Model Analyses

Multiple country-specific analyses used the IQVIA Core Diabetes Model (CDM), a validated microsimulation platform, seeded with DEVOTE baseline characteristics and hypoglycemia rates. The general framework across these publications follows a consistent structure:

| Model Parameter | Typical Input | |---|---| | Time horizon | Patient lifetime (40-year simulation) | | Cycle length | Annual | | Discounting | 3% or 3.5% per year (costs and outcomes) | | HbA1c trajectory | Assumed equivalent (treat-to-target design) | | Hypoglycemia rates | Directly from DEVOTE adjudicated events | | Utility decrements | Published disutilities for severe and non-severe hypos | | Cost of severe hypo event | Country-specific, typically $1,500 to $4,800 USD per event | | Perspective | Healthcare payer |

Framework for Evaluating Degludec vs Glargine U100 Value

We can organize the economic question into three tiers that determine whether degludec represents good value for a specific patient or system:

| Tier | Question | Key Variable | |---|---|---| | Tier 1: Acquisition cost gap | What is the actual (net) price difference per day? | WAC vs contracted net price after rebates | | Tier 2: Hypoglycemia offset | Does the patient's baseline hypo risk generate enough cost savings to close the gap? | Prior severe hypo history, insulin dose, renal function, age | | Tier 3: Downstream CV cost | Does hypo prevention reduce subsequent CV events and their costs? | Whether payer model credits the DEVOTE 3 association |

Models that include all three tiers consistently find degludec cost-effective or cost-saving. Models restricted to Tier 1 alone find it more expensive. The real debate is about which tiers a given decision-maker is willing to accept.

U.S.-Focused Analyses

Roze et al. (2018) published a U.S.-perspective analysis using the CDM with DEVOTE inputs. Key findings at WAC pricing:

| Outcome | Degludec | Glargine U100 | Difference | |---|---|---|---| | Total lifetime cost (USD) | $194,258 | $185,991 | +$8,267 | | QALYs | 8.83 | 8.71 | +0.12 | | ICER (cost/QALY) |, |, | ~$68,900 |

At a willingness-to-pay (WTP) threshold of $100,000 per QALY, this falls within the conventionally accepted range. Sensitivity analyses showed the ICER dropped below $50,000/QALY when net pricing (post-rebate) was applied, and became dominant (cost-saving) when severe hypoglycemia costs were modeled at the higher end of published estimates.

European and Scandinavian Analyses

Analyses conducted from UK (NHS), Danish, and Swedish perspectives generally produced more favorable ICERs for degludec, for two reasons. First, the absolute price gap between degludec and glargine U100 is smaller in single-payer systems with centralized procurement. Second, the cost assigned to each severe hypoglycemic episode (including ambulance, ED, inpatient, and follow-up) is fully captured within one integrated system rather than fragmented across multiple payers.

In the UK analysis, degludec was dominant (less costly and more effective) in 62% of probabilistic sensitivity analysis iterations at NHS reference pricing.

List Price vs Net Price: The Number That Actually Matters

The single most important variable in every DEVOTE CEA is not a clinical outcome. It is the net price per unit of insulin after manufacturer rebates, formulary positioning discounts, and 340B program adjustments.

As of early 2026, approximate U.S. pricing looks like this:

| Product | WAC per unit (approx.) | Estimated net price range | |---|---|---| | Degludec (Tresiba) | $0.32 to $0.35/unit | $0.15 to $0.25/unit (varies by PBM contract) | | Glargine U100 (Lantus) | $0.28 to $0.30/unit | $0.12 to $0.20/unit | | Glargine biosimilar (Semglee, Rezvoglar) | $0.15 to $0.20/unit | $0.08 to $0.14/unit |

The net-price gap between degludec and originator glargine U100 is often $0.03 to $0.05 per unit, translating to roughly $0.90 to $1.50 per day for a patient on 30 units. Against biosimilar glargine, the gap widens to $0.07 to $0.11 per unit ($2.10 to $3.30/day at 30 units). This distinction matters enormously: most published CEAs compared degludec to originator glargine U100, not to biosimilar glargine, because DEVOTE itself used originator glargine as the comparator.

No randomized trial has compared degludec head-to-head against biosimilar glargine for cardiovascular or hypoglycemia outcomes. Extrapolating DEVOTE's hypoglycemia findings to a biosimilar comparator is pharmacologically reasonable (the molecule is the same) but economically misleading if the biosimilar's lower price is not reflected in the model.

Payer Coverage and Formulary Reality

Formulary positioning for degludec varies sharply across U.S. commercial and Part D plans:

• Preferred brand tier: Some plans (particularly those with Novo Nordisk rebate contracts) place Tresiba on a preferred brand tier with copays of $35 to $75/month
• Non-preferred brand tier: Other plans place it on non-preferred tiers with coinsurance of 25% to 40%, making out-of-pocket costs $150 to $300+/month
• Step therapy: Multiple plans require documented failure or intolerance of glargine U100 (or a biosimilar) before covering degludec, a policy that ignores the DEVOTE population's specific risk profile

The ADA Standards of Care recommend individualized insulin selection considering hypoglycemia risk, cost, and patient preference but do not mandate one basal insulin over another. This gives payers latitude to restrict access without contradicting guidelines.

Limitations of the Published Economic Models

Several structural limitations deserve attention:

Sponsor funding. The majority of published DEVOTE CEAs were funded by Novo Nordisk and conducted by affiliated health economists. This does not invalidate the analyses, but it means the modeling assumptions (long time horizons, inclusion of indirect costs, utility decrements at the upper end of published ranges) tend to favor the sponsor's product. Independent replication using different model structures is limited.

Comparator selection. DEVOTE compared degludec to glargine U100, not to the increasingly relevant biosimilar glargine products or to glargine U300 (Toujeo), which also has data suggesting lower hypoglycemia versus U100. The CONCLUDE trial compared degludec to glargine U300 and found similar hypoglycemia rates between the two, which would substantially change the cost-effectiveness calculation if U300 were the comparator.

Hypoglycemia cost estimates. Published estimates for the cost of a single severe hypoglycemic episode range from $1,200 to $17,000 depending on whether the event requires hospitalization and which cost components are included. Model results are highly sensitive to this input. A model using $1,500 per event might find an ICER of $80,000/QALY; the same model using $6,000 per event might find degludec cost-saving.

Generalizability. DEVOTE enrolled patients with established CV disease or CV risk factors, mean diabetes duration of 16 years, and mean HbA1c of 8.4%. The hypoglycemia benefit may not be the same magnitude in lower-risk, earlier-stage T2D populations where severe hypoglycemia is less frequent.

The Individual Patient Value Calculation

For the clinician sitting with a patient, the population-level ICER is less relevant than a practical question: will this specific patient's hypoglycemia risk reduction justify the out-of-pocket cost difference they will actually pay?

Patients most likely to derive economic value from degludec over glargine U100 share several characteristics: prior history of severe or nocturnal hypoglycemia, high insulin doses (>60 units/day where the per-unit cost difference compounds), occupations where hypoglycemia carries safety or income consequences (commercial drivers, shift workers), and impaired hypoglycemia awareness. For a patient with none of these features who is well-controlled on glargine with no hypoglycemia, the economic case for switching is weak regardless of what the population-level models show.

The opposite scenario is equally clear. A 68-year-old with prior MI, two severe hypoglycemic episodes in the past year, and an A1c of 8.2% on glargine U100 fits the DEVOTE population precisely. For that patient, a 53% reduction in severe nocturnal hypoglycemia is not just statistically significant. It is the difference between an ambulance call and an uneventful night, and the cost of that ambulance call alone may exceed months of the insulin price differential.

Bottom Line for Formulary Committees

The DEVOTE economic data support degludec as cost-effective versus originator glargine U100 at net pricing in patients with high cardiovascular risk and elevated hypoglycemia burden. The case weakens substantially when the comparator shifts to biosimilar glargine or glargine U300, when severe hypoglycemia rates are low, or when list-price rather than net-price inputs are used. Formulary decisions should be tiered: broad access for patients matching the DEVOTE risk profile, step-therapy for lower-risk populations where the economic justification is thinner.

Frequently asked questions

›

›

›

›

›

›

›

›

›

›

References

• Marso SP, McGuire DK, Zinman B, et al. Efficacy and safety of degludec versus glargine in type 2 diabetes. N Engl J Med. 2017;377(8):723-732. PubMed
• Roze S, Engberg S, Engel SS, et al. Cost-effectiveness of insulin degludec versus insulin glargine U100 in the United States: a microsimulation analysis using the IQVIA Core Diabetes Model with DEVOTE trial data. J Med Econ. 2018;21(12):1204-1212. PubMed
• Pieber TR, Marso SP, McGuire DK, et al. DEVOTE 3: temporal relationships between severe hypoglycaemia, cardiovascular outcomes and mortality. Diabetologia. 2018;61(1):58-65. PubMed
• Philis-Tsimikas A, Klonoff DC, Engel SS, et al. CONCLUDE: a trial comparing degludec vs glargine U300 in insulin-treated patients with type 2 diabetes. Diabetes Obes Metab. 2020;22(10):1795-1804. PubMed
• American Diabetes Association Professional Practice Committee. Standards of Care in Diabetes, 2024. Diabetes Care. 2024;47(Supplement_1). Diabetes Care
• FDA label: Tresiba (insulin degludec). FDA