Advertisement

Diabetes Therapy

, Volume 10, Issue 5, pp 1969–1984 | Cite as

Meta-Analysis and Cost-Effectiveness Analysis of Insulin Glargine 100 U/mL Versus Insulin Degludec for the Treatment of Type 2 Diabetes in China

  • Wen Su
  • Chaoyun Li
  • Lei Zhang
  • Ziyi Lin
  • Jun Tan
  • Jianwei XuanEmail author
Open Access
Original Research

Abstract

Introduction

To evaluate the efficacy and safety as well as the long-term cost-effectiveness of insulin glargine 100 U/mL (IGlar) versus insulin degludec (IDeg) for the treatment of type 2 diabetes mellitus (T2DM) from the Chinese healthcare system perspective.

Methods

A systematic search of English and Chinese electronic databases for randomized controlled trials (RCTs) comparing IGlar with IDeg for the treatment of T2DM was performed, followed by a meta-analysis to compare the efficacy and safety of IGlar versus IDeg. The CORE Diabetes Model was used to estimate lifetime costs, quality-adjusted life years (QALYs) gained, and cost-effectiveness of IGlar versus IDeg. One-way and probabilistic sensitivity analyses were conducted to assess the underlying parameter uncertainty.

Results

Six RCTs were included in the meta-analysis. The IGlar group showed a statistically significant decrease in glycated hemoglobin (HbA1c) from baseline compared to the IDeg group (mean difference [MD] 0.08%, 95% confidence interval [CI] 0.01–0.14%, P = 0.02). Body mass index (BMI) control was numerically better in the IGlar group than in the IDeg group (MD 0.07 kg/m2, 95% CI − 0.01 to 0.14 kg/m2, P  = 0.08). In terms of hypoglycemia, the incidence of non-severe overall hypoglycemia was comparable between the IDeg and IGlar patient groups (P  > 0.05), while the incidence of non-severe nocturnal hypoglycemia (relative risk [RR 0.79], 95% CI 0.70–0.90, P < 0.01) and the event rates of non-severe overall (RR 0.91, 95% CI 0.85–0.97, P < 0.01) and non-severe nocturnal hypoglycemia (RR 0.91, 95% CI 0.85–0.97, P < 0.01) were lower in the IDeg group. The incidences and event rates of both severe overall and nocturnal hypoglycemia were similar for the two groups (P  > 0.05). The cost-effectiveness analysis showed that IGlar is the dominant treatment option compared with IDeg, with a lifetime savings of 1004 Chinese yuan in direct medical costs and a net gain of 0.015 QALYs per patient. Both one-way and probabilistic sensitivity analyses confirmed the robustness of the results.

Conclusions

IGlar is a cost-saving option with incremental effectiveness compared with IDeg for the treatment of T2DM in China.

Funding

Sanofi China.

Keywords

Cost-effectiveness Insulin degludec Insulin glargine Meta-analysis Type 2 diabetes 

Introduction

Type 2 diabetes mellitus (T2DM) is a progressive disease characterized by insulin resistance and the progressive loss of β-cell function, resulting in insulin deficiency. This disease accounts for 95% of all diabetes cases [1], and if uncontrolled, it can result in significant long-term morbidities and early mortality [2].

Globally, an estimated 422 million adults were living with diabetes in 2014, compared to 108 million in 1980. The age-standardized prevalence of diabetes has nearly doubled since 1980, increasing from 4.7 to 8.5% [2]. In China, the estimated overall prevalence of total diabetes was 10.9% among adults (114 million in total) in 2013 [3, 4]. Due to an aging population, it is expected that the absolute number of patients in China will continue to rise in future years.

In 2014, the total healthcare expenditure on diabetes was estimated to be $825 billion worldwide, the largest single share ($170 billion) of which was spent by China [5]. In 2014, the estimated direct cost per diabetic patient per year in China was around $446 [4]. The heavy economic burden of diabetes is mainly due to its complications, including heart attack, stroke, kidney failure, leg amputation, vision loss, and nerve damage. A 2009 Chinese study showed that 60.9% of patients with DM had at least one diabetic complication or related disease, with this percentage increasing to 71.2% in 2011 [6]. A study by Chen et al. revealed that the annual direct medical cost of DM patients with complications was 3.71-fold higher than that of those without a complication [7]. Therefore, both policy-makers and healthcare providers have an invested interest in preventing or slowing the progression of diabetes complications by achieving tight glycemic control .

Owing to the progressive nature of T2DM, a large number of patients require insulins to achieve glycemic control. Recent treatment guidelines have highlighted the importance of basal insulin therapy in people with T2DM [8, 9, 10, 11]. Insulin glargine 100 U/mL (IGlar) was the first once-daily, long-acting insulin analog to be marketed, and it has been in clinical use for more than 15 years [12]. Compared with older human insulin formulations, such as Neutral Protamine Hagedorn (NPH), IGlar achieves a similar reduction of glycated hemoglobin (HbA1c) from baseline but with a significantly lower rate of hypoglycemic events [13]. Insulin degludec (IDeg) is described as a second-generation basal insulin with an ultra-long and stable action profile and lower pharmacodynamic variability [14, 15]. Two meta-analyses investigating the efficacy and safety of IGlar versus IDeg have been recently published [16, 17]. Trials included in the meta-analysis conducted by Liu et al. [16] were a mix of different study designs (randomized controlled trials [RCTs] and cross-over trials) and different dosing regimens of IDeg (once daily or three times a week), which may have confounded the findings. The other meta-analysis was conducted by Roussel et al. [17] who only incorporated clinical trials from the BEGIN series programs and ignored other studies. Therefore, an updated meta-analysis is needed to comprehensively illustrate the comparative efficacy and safety between these two basal insulins.

In addition to clinical evidence, economic evidence is of great importance to decision-makers with the aim to optimize resource use and service delivery. Several cost-effectiveness studies [18, 19, 20, 21, 22, 23] comparing IGlar with IDeg for treating T2DM have been conducted in Western countries, such as the UK and Spain. All of these studies were conducted with a short time horizon (12 months) based on the rationale that the efficacy data were from treat-to-target trials with insulin doses adjusted to achieve similar glycemic control between treatments and, therefore, long-term modeling would not be informative. However, given that T2DM is a chronic progressive disease that could have severe and long-term complications, a long-term simulation model is needed to capture the full economic values of the medications.

The objective of our study was to investigate the efficacy, safety, and long-term cost-effectiveness of IGlar (100 U/ml) versus IDeg in patients with T2DM from the perspective of the Chinese healthcare system.

Methods

Systematic Review and Meta-Analysis

We searched ten English and Chinese electronic databases (PubMed, Embase, Cochrane Central Register of Controlled Trials [CENTRAL], Web of Science, Cochrane library, ClinicalTrials.gov, CNKI, Wanfang, VIP, Chinese Clinical Trial Registry [ChiCTR; chictr.org.cn]) using the search items ‘glargine’ OR ‘Lantus’ AND ‘degludec’ OR ‘Tresiba.’ The time horizon of the literature search was up to March 2018. Only RCTs comparing IGlar with IDeg among T2DM patients were considered. The exclusion criteria included: not a RCT (i.e. crossover or self-controlled trials were excluded); subgroup analysis if original article was already included; repeat publications; publications not in Chinese or English; or no full text available. A total of six clinical trials were ultimately identified [24, 25, 26, 27, 28, 29], among which two trials had extension phases. As several outcomes were not evaluated in the extension phase and their treatment durations were much longer than those of the other studies, we included only the original reports of the trials in our meta-analysis in order to reduce potential heterogeneity. The PRISMA flow chart and characteristics of the included trials are given in the Electronic Supplementary Material (ESM Fig. S1; ESM Table S1).

A total of 4219 T2DM patients who participated in the six clinical trials that compared IGlar (n = 1391) with IDeg (n = 2828) were included in the meta-analysis. The following outcomes were extracted for the meta-analysis: HbA1C reduction from baseline; changes in body mass index (BMI) and weight from baseline; and incidence and episode rates of hypoglycemia. The Cochrane Collaboration’s tool [30] was used to assess the risks of bias for the included studies. The meta-analysis was conducted using Cochrane’s Review Manager (RevMan 5.3). The continuous outcomes were measured by the mean difference (MD), dichotomous outcomes were measured by risk ratio (RR), and the hypoglycemia event rate (per patient year) was measured by log rate ratio.

Cost-Effectiveness Analysis

Model Overview

We employed a validated computer simulation model of diabetes (IQVIA CORE Diabetes model [CDM]) [31, 32] to estimate long-term health outcomes and the economic consequences of implementing interventions during the treatment of diabetes. The CDM allows results to be extrapolated from short-term trials to long-term outcomes. The model comprises 17 interdependent submodels that simulate the complications of diabetes, including angina, myocardial infarction, congestive heart failure, stroke, peripheral vascular disease, diabetic retinopathy, macular edema, cataract, hypoglycemia, ketoacidosis, lactic acidosis, nephropathy and end-stage renal disease, neuropathy, foot ulcer and amputation, pulmonary edema, and depression, in addition to nonspecific mortality. Each submodel is a Markov model using time, state, time-in-state, and diabetes type-dependent probabilities (where appropriate and available) to simulate the progress of patients through different states. For each cycle, the order in which the submodels run changes randomly. Monte-Carlo simulations are performed at the individual patient level using tracker variables to accommodate complex interactions between individual complication submodels. The CDM uses separate transition probabilities and management strategies for type 1 diabetes mellitus and T2DM, and source data for model parameters are obtained from a broad range of published clinical and epidemiological studies. The main sources of the long-term transition data are the Diabetes Control and Complications Trial (DCCT), United Kingdom Prospective Diabetes Studies (UKPDS), and Framingham studies, the former two of which informed the relationship between glycemic control and diabetes prognosis, while the Framingham studies are a series of long-term, ongoing cardiovascular cohort studies. The CDM allows direct and indirect costs to be estimated, adjusts for quality of life, and allows users to perform cost-effectiveness and cost-utility analyses. Access to the model’s technical documentation that lists all data sources is available from the authors on request. A more detailed description of the structure of the IQVIA CORE model is given in a previous publication [31].

In order to reflect the chronic nature of T2DM and capture both mortality and T2DM-related complications over patients’ lifetime, a lifetime (50 years) horizon was used.

Gain of quality-adjusted life-years (QALYs) and direct costs expressed in 2017 Chinese yuan (CNY) from the perspective of the Chinese healthcare system were calculated. Both costs and clinical benefits were discounted at an annual rate of 3.0%. One-way sensitivity analyses and probabilistic sensitivity analyses were conducted on the key parameters to assess the robustness of the results.

Clinical Inputs

A simulated cohort of patients was defined mainly based on the baseline characteristics of patients in a RCT comparing IGlar with IDeg in Chinese T2DM patients [33]. Other baseline characteristics were sourced from the literature [3, 33, 34, 35, 36, 37, 38, 39]. Baseline cohort characteristics are presented in Table 1.
Table 1

Baseline cohort characteristics

Cohort

Mean

SD

Source

Start age (years)

56.0

8.6

[33], Table 1

Duration of diabetes (years)

7.9

5.4

[33], Table 1

Proportion male (%)

50.8

[33], Table 1

HbA1c (%)

8.2

0.9

[33], Table 1

SBP (mmHg)

131.89

15.14

[34], Table 1

DBP (mmHg)

81.02

11.06

[34], Table 1

Total cholesterol (mg/dL)

181.60

39.10

[3], Table 1

HDL-holesterol (mg/dL)

52.30

14.90

[3], Table 1

LDL-cholesterol (mg/dL)

112.92

44.86

[34], Table 1

Triglyceride (mg/dL)

191.32

138.17

[34], Table 1

BMI (kg/m2)

25.0

2.9

[33], Table 1

eGFR (mL/min/1.73 m2)

104.9

0

[35], Table 1

HAEM (g/dL)

1.38

0

[35], Table 1

White blood cells (106/ml)

7.0

0

[35], Table 1

Heart rate (bpm)

72

12

[36], Fig. 3

Waist–hip rate

0.93

0

[37], Table A

UAER (mg/mmol)

2.0

0

[35], Table 1

Serum creatinine (mg/dL)

1.1

0

[37], Table A

Serum albumin (g/dL)

3.9

0

[37], Table A

Proportion smokers (%)

28

[3], Table 1

Cigarettes (n/days)

15.2

[38], Table 9.1

Alcohol consumption (oz./week)

4.36

[39], The first table in P264

BMI Body mass index, DBP diastolic blood pressure, eGFR estimated glomerular filtration rate, HbA1c glycosylated hemoglobin, HEAM blood haemoglobin, HDL high-density lipoprotein, LDL low-density lipoprotein, SBP systolic blood pressure, SD standard deviation, UAER urinary albumin excretion rate

Treatment effects incorporated in our analysis included the changes from baseline in HbA1c and in BMI. The changes in HbA1C and BMI from baseline and the risk of hypoglycemia events in the IGlar group were derived from Pan et al. [27] as the reference case, and the treatment effects in IDeg group were calculated based on the relative treatment effects between the two groups that were derived from our new meta-analysis. Details of applied treatment effects are presented in Table 2.
Table 2

Treatment effects applied in the analysis

Parameter

Mean (SD)

Source

IGlar

 Change from baseline HbA1c (%)

− 1.2 (1.0)

[27]

 Change from baseline BMI (kg/m2)

0.7 (1.1)

[27]

 Non-severe hypoglycemia (events/100 PYE)

96

[27]

 Non-severe nocturnal hypoglycemia (events/100 PYE)

24

[27]

 Severe hypoglycemia (events/100 PYE)

1

[27]

 Severe nocturnal hypoglycemia (events/100 PYE)

0

[27]

 Average daily dosage (IU)

31

[27]

IDeg

 Change from baseline HbA1c (%)

− 1.12 (1.0)

[27], meta-analysis

 Change from baseline BMI (kg/m2)

0.77 (1.1)

[27], meta-analysis

 Non-severe hypoglycemia (events/100 PYE)

87.27

[27], meta-analysis

 Non-severe nocturnal hypoglycemia (events/100 PYE)

16.90

[27], meta-analysis

 Severe hypoglycemia (events/100 PYE)

0.92

[27], meta-analysis

 Severe nocturnal hypoglycemia (events/100 PYE)

0

[27], meta-analysis

 Average daily dosage (IU)

31

[27]

IDeg Insulin degludec, IGlar insulin glargine 100 U/mL, PYE patient-year of exposure

Patients in the two groups were assumed to receive their corresponding treatment regimens until their HbA1c rose above 7.5% (58 mmol/L), then they were switched to basal-bolus therapy [10]. This assumption recognizes that intensification to basal-bolus therapy will be required for patients to maintain glycemic control over the long term. Following application of the treatment effects based on the trial data, all treatment variables were assumed to follow the natural progression algorithms built into the CDM (as described by Palmer et al. [31]). For BMI, the treatment effect upon reaching the HbA1c 7.5% threshold was assumed to remain constant during treatment; this was a conservative approach as in reality weight may be regained gradually over time. Upon reaching the 7.5% HbA1c threshold, the treatment effects of basal-bolus therapy were applied to all patients based on data from the published paper [40].

Costs

Direct medical costs included acquisition costs for IGlar and IDeg, costs of treatment associated with diabetes-related complications, and costs of routine patient management. All costs were expressed in 2017 CNY. Costs associated with self-monitoring of blood glucose, self-injection, and oral anti-hyperglycemic medications were not included in the analyses as these costs were assumed to be the same across treatment arms. The costs of diabetes-related complications in the year of the event and the annual follow-up costs (each year of the simulation subsequent to the event) were mainly calculated from a study on the direct medical costs of diabetes-related complications using the sampling claims data collected by China Health Insurance Research Association (CHIRA) [41]. The detailed calculation equations are presented in ESM Tables S2 and S3. The costs inputs are shown in Table 3.
Table 3

Costs inputs applied in the analysis

Parameter

Value

Source

IGlar unit cost (CNY per IU)

0.617

Average bidding prices [51]

IDeg unit cost (CNY per IU)

0.621

Average bidding prices [51]

Insulin lispro unit cost (CNY per IU)

0.254

Average bidding prices [51]

CVD complications

 MI 1st year cost

73,414

[41]

 MI 2nd + years cost

23,207

[41]

 Angina 1st year cost

35,486

[41]

 Angina 2nd+ years cost

10,039

[41]

 CHF 1st year cost

35,171

[41]

 CHF 2nd+ years cost

18,658

[41]

 Stroke 1st year cost

29,070

[41]

 Stroke 2nd+ years cost

14,381

[41]

 Stroke death within 30 days cost

15,887

[41]

 PVD 1st year cost

21,375

[41]

 PVD 2nd+ years cost

3358

[41]

Renal complications

 HD costs 1st year

144,238

[41]

 Annual costs HD 2+ years

116,005

[41]

 PD costs 1st year

59,882

[52]

 Annual costs PD 2+ years

48,842

[52]

 RT costs 1st year

259,208

[52]

 Annual costs RT 2+ years

68,572

[52]

Acute events

 Non-severe hypoglycemia cost

801

[41]

 Severe hypoglycemia cost

12,557

[41]

 Keto event cost

12,128

[41]

 Lactic acid event cost

8300

[41]

 Edema onset cost

561

[41]

Eye disease

 Laser treatment cost

6287

[41]

 Cataract operation cost

10,339

[41]

 Following cataract operation cost

362

[41]

 Blindness—year of onset cost

2119

[52]

 Blindness—following years cost

700

[52]

Neuropathy/foot ulcer/amputation

 Neuropathy 1st year cost

17,091

[52]

 Neuropathy 2nd  years cost

6557

[52]

 Amputation (event based) cost

15,905

[52]

 Amputation prosthesis (event based) cost

14,283

[52]

 Gangrene treatment cost

14,959

[41]

 After healed ulcer cost

5402

[52]

 Infected ulcer cost

23,080

[52]

 Standard uninfected ulcer cost

19,207

[52]

CHF Congestive heart failure, CVD cardiovascular disease, CYN Chinese yuan, HD hemodialysis, MI myocardial infarction, PVD peripheral vascular disease, PD peritoneal dialysis, RT renal transplant

Utilities and Disutilities

Utilities and disutilities (measures of the impact on quality of life) associated with complications of diabetes were obtained from published sources as shown in ESM Table S4 [42, 43, 44, 45, 46].

Sensitivity Analyses

One-way sensitivity analysis and probabilistic sensitivity analysis were performed on selected key variables to explore the robustness of base-case results in relation to parameter uncertainties and modeling assumptions.

The following one-way sensitivity analyses were conducted: discounting rate (0, 5, or 8%), time horizon (10 or 20 years), the HbA1C threshold for changing treatment line (7 or 8%), hypoglycemic event rate (the same between two groups), BMI (the same between two groups), drug acquisition costs (± 30%), and disease management costs (± 20%).

A Monte Carlo simulation was used to perform probabilistic sensitivity analysis (PSA) with parameter inputs (utilities, costs, treatment effects, and cohort characteristics) sampled from fixed distributions with the mean and standard deviation values. For costs data, a range of ± 10% was applied. The PSA used 1000 simulated patients over 1000 iterations to ensure stability of the results.

Compliance with Ethics Guidelines

This research is entirely based on secondary data sources and does not contain any studies with human participants or animals performed by any of the authors.

Results

Meta-Analysis

The meta-analysis indicated that HbA1c reduction (%) from baseline to study end was significantly greater in patients on IGlar versus those on IDeg (Fig. 1a; MD 0.08%, 95% confidence interval [CI] 0.01–0.14%). In addition, both body weight gain (Fig. 1b; MD 0.17 kg, 95% CI − 0.05 to 0.38 kg) and BMI (Fig. 1c; MD 0.07 kg/m2, 95% CI − 0.01 to 0.14 kg/m2) favored the IGlar group, but the differences between groups were not statistically significant.
Fig. 1

ac Forest plots of meta-analysis results of insulin degludic (IGlar) vs. insulin glargine 100 U/mL (IDeg) for glycated hemoglobin reduction (a), weight gain (kg) (b), and body mass index (kg/m2) (c). CI Confidence interval, IV inverse variance, SD standard deviation

In terms of hypoglycemia, the incidences (percentage of patients who experienced hypoglycemia) of non-severe overall hypoglycemia (Fig. 2a; RR 0.96, 95% CI 0.89–1.02), severe overall hypoglycemia (Fig. 2c; RR 0.68, 95% CI 0.40–1.14), and severe nocturnal hypoglycemia (Fig. 2d; RR 0.98, 95% CI 0.35–2.75) were comparable between the IDeg and IGlar groups, while the incidence of non-severe nocturnal hypoglycemia was significantly lower in IDeg group (Fig. 2b; RR 0.79, 95% CI 0.70–0.90).
Fig. 2

ag Forest plots of meta-analysis results of IGlar vs. IDeg for hypoglycemia in terms of incidence (%) of non-severe overall hypoglycemia (a), incidence (%) of non-severe nocturnal hypoglycemia (b), incidence (%) of severe overall hypoglycemia (c), incidence (%) of severe nocturnal hypoglycemia (d), events rate (PYE) of non-severe overall hypoglycemia (e), events rate (PYE) of non-severe nocturnal hypoglycemia (f), events rate (PYE) of severe overall hypoglycemia (g). M-H Mantel-Haenszel method, PYE patient-year of exposure

The events rates (rate per patient-year) of non-severe overall hypoglycemia (Fig. 2e; RR 0.91, 95% CI 0.85–0.97) and non-severe nocturnal hypoglycemia (Fig. 2f, RR 0.70, 95% CI 0.61–0.82) were significantly lower in IDeg group than in IGlar group. The events rate of severe hypoglycemia was comparable between IDeg and IGlar (Fig. 2g; RR 0.92, 95% CI 0.53–1.59). Due to the rareness of the severe nocturnal hypoglycemia event reported in the clinical studies, the meta-analysis of severe nocturnal hypoglycemia event rate could not be performed.

Cost-Effectiveness Analysis

Base-Case Analysis

As illustrated in Table 4, patients treated with IGlar were projected with a net increase of 0.015 (95% CI 0.006–0.025) QALY gains over a patient’s lifetime compared with those treated with IDeg. The clinical benefit was the result of the reduced cumulative incidence of diabetes-related complications in the IGlar group (ESM Tables S5, S6). The mean lifetime discounted direct cost for the IGlar group was CNY 1004 lower (95% CI − 1758 to − 251) than that for the IDeg group, primarily owing to differences in the costs of drug acquisition and treatment of complications (i.e. cardiovascular disease and renal diseases). Consequently, the use of IGlar was found to predominate relative to the use of IDeg for the treatment of T2DM in China.
Table 4

Base-case analysis

Parameter

IGlar

IDeg

Difference

IGlar vs. IDeg

 Discounted life expectancy (years)

14.554

14.531

0.023

 Discounted QALYs

9.526

9.511

0.015

 Discounted direct costs (CNY)

468,515

469,519

− 1004

  Drug acquisition

121,307

121,244

− 63

  Management

1065

1063

2

  CVD

135,739

136,541

− 802

  Renal

17,824

18,301

− 477

  Ulcer/amputation/neuropathy

100,792

100,888

− 96

  Eye

5084

5148

− 64

  Hypoglycemia

86,704

86,334

370

 ICER (life expectancy)

IGlar dominant

  

 ICER (QALYs)

IGlar dominant

  

ICER Incremental cost-effectiveness ratio, QALY quality-adjusted life year

One-Way Sensitivity Analysis

One-way sensitivity analyses showed that the results were robust to parameter changes (Table 5). IGlar remained dominant in all scenarios tested, with the exception of the time horizon of 10 years, in which IGlar was the dominant medication. Within the range of all parameter changes, the results were most sensitive to the reduction of the discount rate to 0%.
Table 5

One-way sensitivity analyses

Parameters

QALY

Cost (CNY)

ICER

IGlar

IDeg

Difference

IGlar

IDeg

Difference

Discount rate 0%

13.611

13.581

0.03

729,228

730,440

− 1212

Dominant

Discount rate 5%

7.81

7.799

0.01

365,673

366,540

− 867

Dominant

Discount rate 8%

6.064

6.058

0.005

266,543

267,239

− 696

Dominant

Time horizon 10 years

5.377

5.381

− 0.001

205,164

205,506

− 343

77,843 (IDeg is cost-effective)

Time horizon 20 years

8.196

8.195

0.001

366,787

367,446

− 659

Dominant

Change in line when HbA1c reaches at 7%

9.506

9.479

0.027

474,562

481,634

− 7072

Dominant

Change line when HbA1c reaches at 8%

9.541

9.526

0.015

458,073

464,290

− 6217

Dominant

Same hypoglycemia event rate

9.526

9.509

0.017

468,515

469,734

− 1219

Dominant

No difference in BMI

9.526

9.513

0.013

468,515

469,573

− 1058

Dominant

Drug costs increase 30%

9.526

9.511

0.015

504,907

505,892

− 985

Dominant

Drug costs decrease 30%

9.526

9.511

0.015

432,123

433,146

− 1023

Dominant

Management costs increase 20%

9.526

9.511

0.015

537,956

539,174

− 1218

Dominant

Management costs decrease 20%

9.526

9.511

0.015

399,073

399,863

− 790

Dominant

The treatment effects in IDeg group from Pan’s study [33]

9.526

9.525

0.001

468,515

470,420

− 1905

Dominant

Probabilistic Sensitivity Analysis

With 1000 Monte Carlo simulations, assuming a willingness-to-pay threshold of CNY 178,980 per QALY gained (threefold the gross domestic product per capita in 2017 in China), the probabilistic sensitivity analyses indicated that IGlar was dominant (more effective and less costly) in 22.4% of the simulations and cost-effective in 55.7% of the simulations. The incremental cost-effectiveness plane is presented in Fig. 3, and the resulting cost-effectiveness acceptability curve is presented in Fig. 4.
Fig. 3

Incremental cost-effectiveness plane in probabilistic sensitivity analysis. CYN Chinese yuan

Fig. 4

Cost-effectiveness acceptability curve

Discussion

This is the first long-term cost-effectiveness analysis comparing IGlar with IDeg for the treatment of patients with T2DM. Our study indicated that, compared with IDeg, IGlar was the dominant medication with a greater QALY gain and at a lower medical cost for T2DM patients in China. One-way and probabilistic sensitivity analyses pointed to the robustness of the results.

The availability of rising numbers of noninsulin antidiabetic agents has fostered a reluctance to use insulin among both physicians and patients. Nevertheless, in terms of achieving good glycemic control, the use of insulin, sooner rather than later, significantly reduces the risk of diabetic complications and may also slow or even halt diabetes progression [47]. Therefore, basal insulin is recommended for the treatment of newly-diagnosed T2DM patients with a HbA1c of  ≥ 9.0% or fasting plasma glucose of ≥ 11.1 mmol/L, or patients with  a HbA1c of ≥ 7.0% after 3 months of oral antidiabetic drug treatment [8]. When choosing basal insulin, physicians need to balance the efficacy of glycemic control and hypoglycemia risk.

HbA1c is the gold-standard of treatment efficacy as it measures glycemic control over several months and has a predictive value for diabetes complications [17]. The most recent meta-analysis [16] revealed that IGlar achieved a significantly greater HbA1c reduction from baseline than did IDeg, which is similar to our results. Several other previously published meta-analyses [48, 49, 50] reported non-inferior efficacy of HbA1c reduction between IGlar and IDeg. In our meta-analysis, IGlar was associated with significantly greater HbA1c reduction compared with IDeg. Apart from the efficacy, side effects, such as hypoglycemia, which is commonly associated with insulin treatments, also have a major impact on a patient’s life and pose a substantial cost burden through increased treatment costs and reduced productivity [23]. In our study, the incidence of non-severe overall hypoglycemia was comparable between the IDeg and IGlar patient groups, while the incidence of non-severe nocturnal hypoglycemia and the event rates of non-severe overall and non-severe nocturnal hypoglycemia were lower in the IDeg group. The incidence and event rate of several hypoglycemia were comparable between the two groups.

Long-term cost-effectiveness analysis that weighs the benefits of HbA1c reduction and risks of hypoglycemic episodes provides an integrated benefit–risk profile and a unique perspective that is needed for decision-making. In our study, after extrapolating the short-term efficacy data to predict long-term clinical outcomes and corresponding costs, we found that IGlar was associated with more QALY gains and lower costs for T2DM patients in China compared with IDeg over a lifetime horizon. As mentioned above, several other published economic analyses have explored the short-term cost-effectiveness of IGlar versus IDeg and concluded that IDeg is more cost-effective primarily due to the lower risk of hypoglycemic events [18, 19, 20, 21, 22, 23]. However, for a chronic progressive disease that is prone to severe and long-term complications, a long-term simulation model is more likely to capture the full clinical and economic values of the medications. Another possible reason why our results differ from those of published papers is that the current study modeled multiple complications rather than only hypoglycemic events. For people with diabetes, complications, such as cardiovascular disease and foot ulcer, tend to consume more medical resources than does the management of hypoglycemic events.

There are several limitations to our study. First, the transition probabilities used in the CORE model were mainly derived from clinical trials and epidemiological studies conducted in Western populations. Potential differences may exist between Western and Chinese patient populations. In the absence of any long-term followup data on Chinese diabetes patients, the transition probabilities from Western populations are still the best data currently available. Secondly, this study only considered direct costs, and indirect costs, such as costs associated with lost productivity, were not included. If a societal perspective is adopted, the overall benefit of IGlar may be underestimated due to the lower incidence of diabetic complications. Lastly, even though our study indicated that IGlar was the dominant treatment option compared with IDeg, the QALYs gained were fairly small, as illustrated through the CE plane that centered the cloud not far from 0 in the PSA. Consequently, caution is advised in interpreting these results.

Conclusion

Compared with IDeg, IGlar appears to be an agent of choice for the treatment of T2DM patients in China. Over the lifetime treatment horizon, treatment with IGlar is projected to result in a small but significant (95% CI 0.006–0.025) QALY gain at a lower treatment cost.

Notes

Acknowledgements

Funding

Sponsorship for this study and the journal’s Rapid Service Fee were funded by Sanofi China. All authors had full access to all of the data in this study and take complete responsibility for the integrity of the data and accuracy of the data analysis.

Editorial Assistance

X. Henry Hu, MD, Ph.D. provided language help and proofread the article.

Authorship

All named authors meet the International Committee of Medical Journal Editors (ICMJE) criteria for authorship for this article, take responsibility for the integrity of the work as a whole, and have given their approval for this version to be published.

Authorship Contributions

Wen Su, Chaoyun Li, Lei Zhang and Ziyi Lin participated in the study design, analysis, discussion, and preparation of the manuscript. Jun Tan and Jianwei Xuan participated in the analysis, discussion, and preparation of the manuscript.

Disclosures

Chaoyun Li is an employee of Sanofi China. Wen Su, Lei Zhang, Ziyi Lin, Jun Tan, and Jianwei Xuan have nothing to disclose.

Compliance with Ethics Guidelines

This article is based on previously conducted studies and does not contain any studies with human participants or animals performed by any of the authors.

Data Availability

Qualified researchers may request access to patient level data and related study documents, including the clinical study report, study protocol with any amendments, blank case report form, statistical analysis plan, and dataset specifications. Patient-level data will be anonymized, and study documents will be redacted to protect the privacy of trial participants. Further details on Sanofi’s data sharing criteria, eligible studies, and process for requesting access can be found at “ https://doi.org/10.1007/s13300-015-0096-0”.

Open Access

This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Supplementary material

13300_2019_683_MOESM1_ESM.docx (104 kb)
Supplementary material 1 (DOCX 104 kb)

References

  1. 1.
    Hu H, Sawhney M, Shi L, Duan S. A systematic review of the direct economic burden of type 2 diabetes in China. Diabetes Ther. 2015;6(1):7–16.  https://doi.org/10.1007/s13300-015-0096-0.CrossRefPubMedPubMedCentralGoogle Scholar
  2. 2.
    World Health Organization. Global report on diabetes. Geneva: World Health Organization. https://www.who.int/substance_abuse/publications/global_alcohol_report/en/.. Accessed 8 Aug 2018.
  3. 3.
    Wang L, Gao PZM. Prevalence and ethnic pattern of diabetes and prediabetes in China in 2013. JAMA. 2017;317(24):2515–23.  https://doi.org/10.1001/jama.2017.7596.CrossRefPubMedPubMedCentralGoogle Scholar
  4. 4.
    International Diabetes Federation. IDF diabetes atlas 8th edition. Brussels: International Diabetes Federation. https://diabetesatlas.org/resources/2017-atlas.html. Accessed 8 Aug 2018.
  5. 5.
    NCD Risk Factor Collaboration (NCD-RisC). Worldwide trends in diabetes since 1980: a pooled analysis of 751 population-based studies with 4.4 million participants. Lancet. 2008;387(10027):1513–30.  https://doi.org/10.1016/s0140-6736(16)00618-8.CrossRefGoogle Scholar
  6. 6.
    Huang Y, Vemer P, Zhu J, Postma MJ, Chen W. Economic burden in Chinese patients with diabetes mellitus using electronic insurance claims data. PLoS One. 2016;11(8):e0159297.  https://doi.org/10.1371/journal.pone.0159297.CrossRefPubMedPubMedCentralGoogle Scholar
  7. 7.
    Chen XB, Tang L, Chen HY, Zhao LYHS. Assessing the impact of complications on the costs of type 2 diabetes in urban China. Chin J Diabetes. 2003;11(4):238–41.Google Scholar
  8. 8.
    Chinese Diabetes Society. Treatment guidelines for type 2 diabetes in China. Chin J Diabetes Mellit. 2018;10(1):4–67.Google Scholar
  9. 9.
    Ganda OP, Segal A, Blair E, Beaser R, Gaglia J, Halprin E. GRM of the JCOC. CHAPTER 5. Clinical guideline for pharmacological management of adults with type 2 diabetes. Am J Manag Care. 2018;24(7):SP253–62.  https://doi.org/10.1111/jdi.12102.CrossRefPubMedGoogle Scholar
  10. 10.
    Internal Clinical Guidelines Team. Type 2 diabetes in adults: management. NICE guideline, no 28. London: National Institute for Health and Care Excellence (NICE);2015.Google Scholar
  11. 11.
    Inzucchi SE, Bergenstal RM, Buse JB, et al. Management of hyperglycemia in type 2 diabetes: a patient-centered approach: position statement of the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetes Care. 2012;35(6):1364–79. CrossRefPubMedPubMedCentralGoogle Scholar
  12. 12.
    Owens DR, Biol FI. Insulin preparations with prolonged effect. Diabetes Technol Ther. 2011;13:S5–14. https://doi.org/10.2337/dc12-0413.CrossRefPubMedGoogle Scholar
  13. 13.
    Horvath K, Jeitler K, Berghold A, et al. Long-acting insulin analogues versus NPH insulin (human isophane insulin) for type 2 diabetes mellitus (review). Cochrane Database Syst Rev. 2007;18(2):CD005613.Google Scholar
  14. 14.
    Heise T, Hermanski L, Nosek L, Feldman A, Rasmussen S, Haahr H. Insulin degludec: four times lower pharmacodynamic variability than insulin glargine under steady-state conditions in type 1 diabetes. Diabetes Obes Metab. 2012;14(9):859–64.  https://doi.org/10.1111/j.1463-1326.2012.01627.x.CrossRefPubMedGoogle Scholar
  15. 15.
    Heise T, Nosek L, Bøttcher SG, Hastrup HHH. Ultra-long-acting insulin degludec has a flat and stable glucose-lowering effect in type 2 diabetes. Diabetes Obes Metab. 2012;14(10):944–50.CrossRefPubMedGoogle Scholar
  16. 16.
    Liu W, Yang X, Huang J. Efficacy and safety of insulin degludec versus insulin glargine: a systematic review and meta-analysis of fifteen clinical trials. Int J Endocrinol. 2018;2018:8726046.  https://doi.org/10.1155/2018/8726046.CrossRefPubMedPubMedCentralGoogle Scholar
  17. 17.
    Roussel R, Ritzel R, Boëlle-Le Corfec E, Balkau B, Rosenstock J. Clinical perspectives from the BEGIN and EDITION programmes: trial-level meta-analyses outcomes with either degludec or glargine 300 U/mL vs glargine 100 U/mL in T2DM. Diabetes Metab. 2018;44(5):402–9.  https://doi.org/10.1016/j.diabet.2018.02.002.CrossRefPubMedGoogle Scholar
  18. 18.
    Evans M, Mehta R, Gundgaard J, Chubb B. Cost-effectiveness of insulin degludec vs. insulin glargine U100 in type 1 and type 2 diabetes mellitus in a UK setting. Diabetes Ther. 2018;9(5):1919–30.  https://doi.org/10.1016/j.diabet.2018.02.002.CrossRefPubMedPubMedCentralGoogle Scholar
  19. 19.
    Lalić N, Russel-Szymczyk M, Culic M, Tikkanen CKCB. Cost-effectiveness of insulin degludec versus insulin glargine U100 in patients with type 1 and type 2 diabetes mellitus in Serbia. Diabetes Ther. 2018;9(3):1201–16.  https://doi.org/10.1007/s13300-018-0426-0.CrossRefPubMedPubMedCentralGoogle Scholar
  20. 20.
    Mezquita-Raya P, Darbà J, Ascanio M, Ramírez de Arellano A. Cost-effectiveness analysis of insulin degludec compared with insulin glargine U100 for the management of type 1 and type 2 diabetes mellitus—from the Spanish National Health System perspective. Expert Rev Pharmacoecon Outcomes Res. 2017;17(6):587–95.  https://doi.org/10.1080/14737167.2017.1345628.CrossRefPubMedGoogle Scholar
  21. 21.
    Evans M, Chubb B, Gundgaard J. Cost-effectiveness of insulin degludec versus insulin glargine in adults with type 1 and type 2 diabetes mellitus. Diabetes Ther. 2017;8(2):275–91.  https://doi.org/10.1007/s13300-017-0236-9.CrossRefPubMedPubMedCentralGoogle Scholar
  22. 22.
    Pollock RFTC. A short-term cost-utility analysis of insulin degludec versus insulin glargine U100 in patients with type 1 or type 2 diabetes in Denmark. J Med Econ. 2017;20(3):213–20.  https://doi.org/10.1080/13696998.2016.1245663.CrossRefPubMedGoogle Scholar
  23. 23.
    Evans M, Wolden M, Gundgaard J, Chubb B, Christensen T. Cost-effectiveness of insulin degludec compared with insulin glargine for patients with type 2 diabetes treated with basal insulin—from the UK health care cost perspective. Diabetes Obes Metab. 2014;16(4):366–75.  https://doi.org/10.1111/dom.12250.CrossRefPubMedGoogle Scholar
  24. 24.
    Gough SC, Bhargava A, Jain R, Mersebach H, Rasmussen SBR. Low-volume insulin degludec 200 units/ml once daily improves glycemic control similarly to insulin glargine with a low risk of hypoglycemia in insulin-naive patients with type 2 diabetes: a 26-week, randomized, controlled, multinational, treat-to-target trial. Diabetes Care. 2013;36(9):2536–42.  https://doi.org/10.2337/dc12-2329.CrossRefPubMedPubMedCentralGoogle Scholar
  25. 25.
    Meneghini L, Atkin SL, Gough SC, et al. BKN-3668 (BEGIN FTI)The efficacy and safety of insulin glargine and insulin degludec dosed at the same time daily. Diabetes Care. 2013;36(4):858–64.  https://doi.org/10.2337/dc12-1668.CrossRefPubMedPubMedCentralGoogle Scholar
  26. 26.
    Onishi Y, Iwamoto Y, Yoo SJ, Clauson P, Tamer SCPS. Insulin degludec compared with insulin glargine in insulin-naıve patients with type 2 diabetes: a 26-week, randomized, controlled, Pan-Asian, treat-to-target trial. J Diabetes Investig. 2013;4(6):605–12.  https://doi.org/10.1111/jdi.12102.CrossRefPubMedPubMedCentralGoogle Scholar
  27. 27.
    Pan C, Gross JL, Yang W, Lv X. A multinational, randomized, open-label, treat-to-target trial comparing insulin degludec and insulin glargine in insulin-naive patients with type 2 diabetes mellitus naive. Drugs R&D. 2016;16(2):239–49.  https://doi.org/10.1007/s40268-016-0134-z.CrossRefGoogle Scholar
  28. 28.
    Garber AJ, King AB, Del Prato S, et al. Insulin degludec, an ultra-longacting basal insulin, versus insulin glargine in basal-bolus treatment with mealtime insulin aspart in type 2 diabetes (BEGIN basal-bolus type 2): a phase 3, randomised, open-label, treat-to-target non-inferiority trial. Lancet. 2012;379(9825):1498–507.  https://doi.org/10.1016/S0140-6736(12)60205-0.CrossRefPubMedGoogle Scholar
  29. 29.
    Zinman B, Philis-Tsimikas A, Cariou B, et al. MCN-3579 (BEGIN OLTI)Insulin degludec versus insulin glargine in insulin-naive patients with type 2 diabetes: a 1-year, randomized, treat-to-target trial (BEGIN Once Long). Diabetes Care. 2012;35(12):2464–71.  https://doi.org/10.2337/dc12-1205.CrossRefPubMedPubMedCentralGoogle Scholar
  30. 30.
    Higgins JPT, Green S (editors). Cochrane handbook for systematic reviews of interventions. Chichester: John Wiley & Sons, 2011.Google Scholar
  31. 31.
    Palmer AJ, Roze S, Valentine WJ, et al. The CORE diabetes model: projecting long-term clinical outcomes, costs and costeffectiveness of interventions in diabetes mellitus (types 1 and 2) to support clinical and reimbursement decision-making. Curr Med Res Opin. 2004;20 Suppl 1:S5–26.  https://doi.org/10.1185/030079904X1980.CrossRefPubMedGoogle Scholar
  32. 32.
    McEwan P, Foos V, Palmer JL, Lamotte M, Lloyd A, Grant D. Validation of the IMS CORE diabetes model. Value Health. 2014;17(6):714–24.  https://doi.org/10.1016/j.jval.2014.07.007.CrossRefPubMedGoogle Scholar
  33. 33.
    Mu Y, Guo L, Li L, et al. The efficacy and safety of insulin degludec versus insulin glargine in insulin-naive subjects with type 2 diabetes: results of a Chinese cohort from a multinational randomized controlled trial. Chin J Intern Med. 2017;56(9):660–6.Google Scholar
  34. 34.
    Ji L-N, Lu J-M, Guo X-H, et al. Glycemic control among patients in China with type 2 diabetes mellitus receiving oral drugs or injectables. BMC Public Health. 2013;13(1):602.  https://doi.org/10.1186/1471-2458-13-602.CrossRefPubMedPubMedCentralGoogle Scholar
  35. 35.
    Yang X, Ma RC, So W-Y, et al. Development and validation of a risk score for hospitalization for heart failure in patients with type 2 diabetes mellitus. Cardiovasc Diabetol. 2008;7(1):9.  https://doi.org/10.1186/1475-2840-7-9.CrossRefPubMedPubMedCentralGoogle Scholar
  36. 36.
    Hayes AJ, Leal J, Gray AM, Holman RR, Clarke PM. 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(9):1925–33.CrossRefGoogle Scholar
  37. 37.
    Folsom AR, Chambless LE, Duncan BB, Gilbert AC, Pankow JS. Prediction of coronary heart disease in middle-aged adults with diabetes. Diabetes Care. 2003;26(10):2777–84.  https://doi.org/10.2337/diacare.26.10.2777.CrossRefPubMedGoogle Scholar
  38. 38.
    Yang Yan, Nan Yi, Mengwu Tu, Wang Jijiang, Lili Wang YJ. Major finding of 2015 China adults tobacco survery. Chin J Health Manag. 2015;2016(2):85–7.Google Scholar
  39. 39.
    World Health Organization. Country profiles. Global status report on alcohol and health 2014. Geneva: World Health Organization.Google Scholar
  40. 40.
    Freemantle N, Mamdani M, Vilsbøll T, Kongsø JH, Kvist K, Bain SC. IDegLira versus alternative intensification strategies in patients with type 2 diabetes inadequately controlled on basal insulin therapy. Diabetes Ther. 2015;6(4):573–91.  https://doi.org/10.1007/s13300-015-0142-y.CrossRefPubMedPubMedCentralGoogle Scholar
  41. 41.
    Duan X, Li C, Li YLQ. Epidemiological characteristics, medical cost and healthcare resource utilization of diabetes-related complications among chinese patients with type 2 diabetes mellitus. Value Health 2018;21:S40. CrossRefGoogle Scholar
  42. 42.
    Beaudet A, Clegg J, Thuresson P-O, Lloyd A, McEwan P. Review of utility values for economic modeling in type 2 diabetes. Value Health. 2014;17(4):462–70.  https://doi.org/10.1016/j.jval.2014.03.003.CrossRefPubMedGoogle Scholar
  43. 43.
    Goldney RD, Phillips PJ, Fisher LJWD. Diabetes, depression, and quality of life: a population study. Diabetes Care. 2004;27(5):1066–70.CrossRefPubMedGoogle Scholar
  44. 44.
    Marc E, Kamlesh K, Muhammad M, et al. Health-related quality of life associated with daytime and nocturnal hypoglycaemic events: a time trade-off survey in five countries. Health Qual Life Outcomes. 2013;11(90):1–9.  https://doi.org/10.1186/1477-7525-11-90.CrossRefGoogle Scholar
  45. 45.
    Marrett E, Radican L, Davies MJ, Zhang Q. Assessment of severity and frequency of self-reported hypoglycemia on quality of life in patients with type 2 diabetes treated with oral antihyperglycemic agents : a survey study. BMC Res Notes. 2011;4:251–8.  https://doi.org/10.1186/1756-0500-4-251.CrossRefPubMedPubMedCentralGoogle Scholar
  46. 46.
    Matza LS, Boye KS, Yurgin N, Brewster-Jordan J, Mannix S, Shorr JMBB. Utilities and disutilities for type 2 diabetes treatment-related attributes. Qual Life Res. 2007;16(7):1251–65.  https://doi.org/10.1007/s11136-007-9226-0.CrossRefPubMedGoogle Scholar
  47. 47.
    Lovre D, Fonseca V. Benefits of timely basal insulin control in patients with type 2 diabetes. J Diabetes Complications. 2015;29(2):295–301. https://doi.org/10.1016/j.jdiacomp.2014.11.018. CrossRefPubMedGoogle Scholar
  48. 48.
    Rodbard HW, Gough S, Lane W. Reduced risk of hypoglycemia with insulin degludec versus insulin glargine in patients with type 2 diabetes requiring high doses of basal insulin: a meta-analysis of 5 randomized begin trials. Endocr Pract. 2014;20(4):285–92.  https://doi.org/10.4158/ep13287.or.CrossRefPubMedGoogle Scholar
  49. 49.
    Vora J, Christensen T, Rana A, Bain SC. Insulin degludec versus insulin glargine in type 1 and type 2 diabetes mellitus: a meta-analysis of endpoints in phase 3a trials. Diabetes Ther. 2014;5(2):435–46.  https://doi.org/10.1007/s13300-014-0076-9.CrossRefPubMedPubMedCentralGoogle Scholar
  50. 50.
    Einhorn D, Handelsman Y, Bode BW, Endahl LA, Mersebach H, King AB. Patients achieving good glycemic control (HBA1c < 7%) experience a lower rate of hypoglycemia with insulin degludec than with insulin glargine: a meta-analysis of phase 3A trials. Endocr Pract. 2015;21(8):917–26.  https://doi.org/10.4158/EP14523.OR.CrossRefPubMedGoogle Scholar
  51. 51.
    Yaozhi database. China’s provincial drug bidding price [DB/OL]. https://db.yaozh.com/yaopinzhongbiao.
  52. 52.
    Wu J, Xiaoning HYL. Cost-effectiveness analysis of insulin aspart 30 versus insulin glargine in patients with type 2 diabetes in China. Chin Pharm J. 2016;51(5):242–7.Google Scholar

Copyright information

© The Author(s) 2019

Authors and Affiliations

  • Wen Su
    • 1
  • Chaoyun Li
    • 2
  • Lei Zhang
    • 3
  • Ziyi Lin
    • 3
  • Jun Tan
    • 3
  • Jianwei Xuan
    • 1
    Email author
  1. 1.Health Economics Research InstituteSun Yat-sen UniversityGuangzhouChina
  2. 2.Health Economics and Outcome ResearchSanofiShanghaiChina
  3. 3.Shanghai Centennial ScientificShanghaiChina

Personalised recommendations