- Email: [email protected]
Cost-Effectiveness of Switching Patients with Type 2 Diabetes from Insulin Glargine to Insulin Detemir in Chinese Setting: A Health Economic Model Based on the PREDICTIVE Study
VALUE IN HEALTH 15 (2012) S56 –S59
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/jval
Cost-Effectiveness of Switching Patients with Type 2 Diabetes from Insulin Glargine to Insulin Detemir in Chinese Setting: A Health Economic Model Based on the PREDICTIVE Study Li Yang, PhD1,*, Torsten Christensen, MSc2, Fengyu Sun, MD3, Jinghua Chang, MSc3 1 Department of Health Policy and Management, School of Public Health, Peking University, China; 2Novo Nordisk A/S, Virum, Denmark; 3Novo Nordisk (China) Pharmaceuticals Co., Ltd., Beijing, China
A B S T R A C T
Objectives: To evaluate the long-term cost-effectiveness of switching from insulin glargine (IGla) to insulin detemir (IDet) in patients with type 2 diabetes in the setting of Chinese secondary and tertiary hospitals. Methods: A published and validated computer simulation model of diabetes (the Center for Outcomes Research model) was used to make the long-term (30 years) projection of health economic outcomes. Patient demographic information and clinical end points were derived from a subgroup analysis of the Predictable Results and Experience in Diabetes through Intensification and Control to Target: an International Variability Evaluation (PREDICTIVE) study. Baseline risk factors and racial characteristic data were obtained from Chinese cohort studies. The diabetes management and complications costs were obtained from Chinese published data and adjusted to 2010 values by using the Chinese consumer price index. An annual discounting rate of 3% was used for both health and cost outcomes, and one-way sensitivities analysis was performed, which illustrated that the results were
Introduction The effective prevention and management of diabetes mellitus (DM) is an increasing serious challenge to China. The China National Diabetes and Metabolic Disorders Study estimated that the age-standardized prevalence of diabetes and prediabetes was 9.7% and 15.5%, respectively, accounting for 92.4 million adults with diabetes and 148.2 million adults with prediabetes in China [1]. According to another survey conducted by the Ministry of Health [2], type 2 DM (T2DM) constitutes about 95% of all patients and the total costs of DM in China were estimated to be $20.5 billion for 2007, which accounted for 14.2% of national health-care expenditures [3]. In China, DM-related complications are the key driver of medical costs. The results of the first national DM-related complications survey based on a sample of 24,496 patients with diabetes in China showed that 73.2% of patients with diabetes had one or more complications [4]. The research conducted by Chen et al. [5] revealed that the annual direct medical cost of patients with complications was increased by 3.71 times compared with that of those without complications.
robust. Results: Conversion to IDet from IGla was projected to improve patient life expectancy by 0.06 year and 0.48 quality-adjusted life-years. Drug costs and management costs of diabetes mellitus were increased by US$368 (US$17,466 vs. US$17,097) and US$31 (US$5464 vs. US$5433), respectively. However, the costs of complications, including cerebrovascular disease, renal complications, ulcer/ amputation/neuropathy, eye complications, and hypoglycemia events, were reduced by US$819 (US$21,294 vs. US$22,114), resulting in a total direct medical cost saving of US$420 when converting to IDet. Conclusions: Conversion to IDet from an IGla regimen improved life expectancy and was a cost-saving treatment approach in a Chinese setting. Keywords: Chinese setting, cost-effectiveness, cost-utility, insulin detemir. Copyright © 2012, International Society for Pharmacoeconomics and Outcomes Research (ISPOR). Published by Elsevier Inc.
In the treatment of DM, insulin plays an important role. Earlier use of insulin in patients with T2DM will help to better control the glucose level and reduce the risk of complications. Detemir and glargine are both long-acting human insulin analogues used for maintaining the basal level of insulin. They have been proved to more closely reflect physiological basal insulin and are associated with a significantly less risk of hypoglycemia and weight gain [6,7]. Two pharmacoeconomic studies conducted in Germany [8] and the United States [9] evaluated the cost-effectiveness of detemir based on the clinical results of the German PREDICTIVE study. Moreover, other researches from the United Kingdom [10], Canada [11], and Europe [12] also systematically evaluated the long-term health benefits of detemir caused by the reduction in the hemoglobin A1c (Hb A1c) level. Insulin detemir came to the Chinese market in 2010 and is being used in secondary and tertiary hospitals in China. To better inform the insurance coverage decision makers and the doctors whether it is value for money in the long run, it is necessary to assess the long-term cost-effectiveness of insulin detemir com-
Conflicts of interest: The authors have indicated that they have no conflicts of interest with regard to the content of this article. * Address correspondence to: Li Yang, PhD, Department of Health Policy and Management, School of Public Health, Peking University, No. 38 Xueyuan Road, Haidian District, Beijing 100191, China. E-mail: [email protected] or [email protected]. 1098-3015/$36.00 – see front matter Copyright © 2012, International Society for Pharmacoeconomics and Outcomes Research (ISPOR). Published by Elsevier Inc. doi:10.1016/j.jval.2011.11.018
VALUE IN HEALTH 15 (2012) S56 –S59
pared with that of the existing long-acting human insulin analogues such as insulin glargine. This study was conducted to quantify the long-term cost-effectiveness of insulin detemir compared with that of insulin glargine for the treatment of T2DM in China from a societal perspective, and to assess the sensitivity of results to disutility values.
Methods Model We used a validated computer simulation model of diabetes (the Center for Outcomes Research [CORE] model) specifically for diabetes to convert the short-term clinical effects to long-term economic outcomes [13]. The model is based on a series of submodels that simulate the important complications of DM. Each submodel is a Markov model using Monte Carlo simulation. The CORE model takes into account baseline cohort characteristics, history of complications, current and future diabetes management and concomitant medications, screening strategies, and changes in physiological parameters over time. The development of complications, life expectancy, quality-adjusted life-year (QALY), and total costs within populations can be calculated. Furthermore, the reliability of simulated outcomes has been tested [13,14].
S57
The management costs and DM-related complications costs were referred to the results of the survey conducted by the IMS Health in six tertiary and three secondary hospitals in China [21]. Complication costs refer to medical expenditure that resulted from diabetes-related complications. Management costs refer to administration devices and blood glucose monitoring and took into account OAD usage and variations in insulin dosage as noted in the PREDICTIVE study. The outpatient visit services and others including screening services were included. All the costs were inflated to 2010 values by using the consumer price index of health care, which was 3.2% in 2009, 2.9% in 2008, and 2.1% in 2007, respectively [22]. The annual medication costs were calculated and are shown in Table 2 (Cost data for management and diabetesrelated complications in China for 2010) in Supplemental Materials found at doi:10.1016/j.jval.2011.11.018.
Quality of life and utilities Quality-adjusted life expectancy was incorporated into the analysis by using diabetes-related health state utility and event disutility values published by Palmer et al. [13]. A disutility value of ⫺0.0035 was applied for each minor hypoglycemic event (an event not requiring hospitalization) recorded in the simulation. To capture the reduced utility associated with major hypoglycemic events, a disutility value of ⫺0.0118 was applied. Hypoglycemia disutility values were derived from a published source [23].
Treatment effects The treatment effects were derived from the Korean cohort of the PREDICTIVE study [15]. In the research, subjects who required changing treatment, as assessed by their treating physicians, were transferred to insulin detemir ⫾ oral antidiabetic drugs (OADs) from insulin glargine ⫾ OADs. The major reasons why the treating physicians transferred a patient to insulin detemir included the following: 1) improve glycemic control (89%); 2) reduce plasma variability (53%); 3) patient’s dissatisfaction with current therapy (33%); and 4) improve weight control (29%). Because PREDICTIVE is an observational study based on everyday clinical practice, the treating physicians could adjust the patient’s OADs and insulin detemir dose and frequency for the duration of the study according to the patient’s situation. The subgroup research of the Korean cohort of the PREDICTIVE study included 875 patients with T2DM who transferred to insulin detemir ⫾ OADs from insulin glargine ⫾ OADs. Reported changes in the Hb A1c level, body mass index (BMI), and hypoglycemic event rate were applied in the simulation as shown. After 3 months of treatment, subjects were associated with a reduction of 0.20% and 0.12 kg/m2 in the Hb A1c level and BMI, respectively. The hypoglycemic event rates were lower in patients using insulin detemir.
Simulation cohort A hypothetical simulation cohort was generated on the basis of patient demographics, risk factors, and preexisting complications. The patient demographics corresponded with the values from the Korean arm of the PREDICTIVE study [15]. Other data were extracted from published country-specific articles [16 –20] (see Table 1 [Inputs for Chinese simulation cohort with T2DM] in Supplemental Materials found at doi:10.1016/j.jval.2011.11.018). We defined this cohort including 1000 persons.
Costs and perspective Direct medical costs were accounted from a society perspective in 2010 US dollars. Direct costs were taken to be the sum of patients’ management costs and DM-related complications costs. The costs of lost productivity due to illness and death were excluded.
Discounting and time horizon All costs and clinical benefits were discounted at an annual rate of 3.0% in accordance with recommendations by the World Health Organization [24]. Furthermore, to reflect the chronic nature of T2DM and capture both mortality and the incidence of DM-related complications over patient lifetime, the clinical and economic impact of treatment with insulin detemir was prolonged to 30 years.
Sensitivity analysis Several one-way sensitivity analyses were performed to examine the influence of key input parameters on the outcomes projected by the model. The varying parameters included time horizon (10, 20, or 30 years), discounting rate (0% or 10%), hypoglycemic event rate (assume same), BMI (assume same), and the reduction in the Hb A1c level (assume Hb A1c benefits reduced by 50%).
Statistical methodology A simulated cohort of 1000 patients was run through the model 1000 times for each simulation (base-case and sensitivity analysis). The mean and standard deviation values were generated by using a nonparametric bootstrapping approach. One thousand mean values of incremental costs and incremental effectiveness in terms of qualityadjusted life expectancy were plotted on a cost-effectiveness plan. For treatment that was not dominant, it was planned to generate an acceptability curve by calculating the proportion of points below a range of willingness-to-pay threshold.
Willingness to pay There is no specific threshold to judge the cost-effectiveness of an intervention in China. According to World Health Organization recommendations [24], if the incremental costs per effectiveness are not more than three times the gross domestic product per capital, the intervention could be regarded as cost-effective. With gross domestic product per capital for China in 2010 at US$3750, we used a willingness-to-pay threshold of US$11,250 in this analysis to determine the likelihood of insulin detemir being considered cost-effective in China.
S58
VALUE IN HEALTH 15 (2012) S56 –S59
Results Clinical outcomes In the base-case analysis, transferring from insulin glargine ⫾ OADs to insulin detemir ⫾ OADs was associated with an improvement of 0.061 year (11.461 ⫾ 0.176 years vs. 11.522 ⫾ 0.177 years) in life expectancy per patient. Maybe it can be attributed to the reduction in the Hb A1c level. Capturing the quality of life similarly showed the benefits of converting to detemir treatment. After switching to detemir ⫾ OADs, the QALYs were increased from 7.457 to 6.974. It was associated with an increment of 0.484 QALYs per patient (see Table 3 [Summary of base-case results] in Supplemental Materials found at doi:10.1016/j.jval.2011.11.018). Therapy conversion to insulin detemir ⫾ OADs from insulin glargine ⫾ OADs was projected to reduce the cumulative incidence of eye complications, renal complications, foot complications, neuropathic complications, and most of cardiovascular complications. The most remarkable reduction was found in renal complications, with 6.37% reduction in end-stage renal complications (see Fig. 1 [Relative changes in the cumulative residence of complications over 30 years] in Supplemental Materials found at doi: 10.1016/j.jval.2011.11.018).
Cost outcomes Compared with insulin glargine ⫾ OADs, treatment with insulin detemir ⫾ OADs was associated with an increase of US$369 in lifetime drug costs and US$31 in management costs. However, in the long run, therapy conversion to detemir ⫾ OADs would reduce the incidence of DM-related complications, which reduced US$820 in complication costs. Finally, the higher drug costs and management costs of detemir ⫾ OADs were more than offset by decreased complications costs, giving a net saving of US$420 per patient (see Table 3 [Summary of base-case results] in Supplemental Materials found at doi:10.1016/j.jval.2011.11.018).
Evaluation of cost-effectiveness For patients with T2DM, projection over 30 years indicated that therapy conversion to insulin detemir ⫾ OADs would lower total treatment costs and increase life expectancy by 0.061 years or 0.484 QALYs per patient. So, we came to the conclusion that switching patients to treatment with detemir ⫾ OADs was the dominant choice, being cost saving in comparison with glargine ⫾ OADs (see Table 3 [Summary of base-case results] in Supplemental Materials found at doi:10.1016/j.jval.2011.11.018). The scatter plot that described incremental costs versus incremental effectiveness (in terms of QALY) for insulin detemir ⫾ OADs versus insulin glargine ⫾ OADs from the 1000 patients simulated 1000 times each in the model showed that almost all the points fell on the right side of the vertical axis. Half of the points were distributed in the lower right corner quadrant (more effective and less costly). It meant that insulin detemir ⫾ OADs was cost saving or cost-effective (see Fig. 2 in Supplemental Materials found at doi:10.1016/j.jval.2011.11.018). A cost-effectiveness acceptability curve was generated to evaluate the likelihood that insulin detemir ⫾ OADs would represent good value for money in China. At a willingness-to-pay threshold of US$3750 per QALY gained, insulin detemir ⫾ OADs had a 100% probability of being acceptable compared with insulin glargine ⫾ OADs from a social perspective. At a willingness-to-pay threshold of US$3750 per life-year gained, insulin detemir ⫾ OADs had 65% probability at least. Even at a willingness-to-pay threshold of zero per QALY, insulin detemir ⫾ OADs still had a 65% likelihood of being considered cost-effective (see Fig. 3 in Supplemental Materials found at doi:10.1016/j.jval.2011.11.018).
Sensitivity analysis A series of sensitivity analysis revealed that the findings in patients with T2DM were most sensitive to changes in the Hb A1c level and incidence of all hypoglycemia. Varying time horizons, discount rate, and BMI had no substantial impact on final results. Nevertheless, in all the sensitivity analyses undertaken, the incremental cost-effectiveness ratio of insulin detemir ⫾ OADs was always within the limit of willingness to pay in China (see Table 4 in Supplemental Materials found at doi:10.1016/j.jval.2011.11.018). In all scenarios of the sensitivity analysis, QALY in patients treated with insulin detemir ⫾ OADs was higher than in patients treated with insulin glargine ⫾ OADs all the time. So, insulin detemir ⫾ OADs was a cost-saving treatment.
Discussion Despite the availability of numerous treatment options to reduce blood glucose levels, the condition of most patients with T2DM in China was not adequately controlled. Insulin detemir has been introduced in the Chinese market successfully, which has increased the choice for patients. In consideration of the disease burden and the current situation of diabetes, however, we should select the most cost-efficient treatment strategies to control the increasing costs in this growing population. Although both insulin detemir and insulin glargine are longacting basal insulin, they differ at the molecular level, which may lead to different treatment effects. The PREDICTIVE study results not only confirm the difference between them but also provide the proof of conversion to detemir treatment after failure of therapy with glargine. Based on the short-term, patient-based results of the PREDICTIVE study, long-term projections using the CORE diabetes model demonstrated that treatment with insulin detemir ⫾ OADs was associated not only with improvements in life expectancy and QALY in comparison with teatment with insulin glargine ⫾ OADs for patients with T2DM but also reduction in total costs. According to modeling analysis and sensitivity findings, switching poorly controlled patients to insulin detemir ⫾ OADs is likely to be cost saving in China. This study is the first to compare the cost-effectiveness of insulin detemir and insulin glargine by applying the CORE model to the Chinese setting. What is more, this study is not limited to short-term treatment effects but long-term health outcomes. So, it is very useful for decision makers. It should be noted, however, that there are some limitations in this research, which must be pointed out. First, the CORE model consists of 15 Markov submodels. Every model will use the probabilities for transitions among different states. The transition probabilities were mainly derived from clinical trials and epidemiological studies conducted in white cohort. But medical practice in China has a big gap in comparison with Western countries. Transition probabilities in China may be higher than those used in the CORE model. The CORE model is used under Chinese medical practice, which likely underestimates the incidence of DM-related complications. Second, the use of data from the Korean cohort of the PREDICTIVE study to simulate treatment with insulin detemir ⫾ OADs in China inherently carries the assumption that the reported treatment effects are applicable to Chinese patients who share similar baseline characteristics with those of the Korean cohort and who are receiving a similarly high level of diabetes care. To overcome this limitation, it is reassuring that the treatment effects were varied in the sensitivity analysis. Even though the effect of insulin detemir ⫾ OADs treatment on the Hb A1c level was reduced by 50%, insulin detemir ⫾ OADs treatment was still cost-effective. Finally, this study considered only the direct costs. The indirect costs associated with lost productivity were not included. How-
VALUE IN HEALTH 15 (2012) S56 –S59
ever, the indirect costs may account for a large percentage of total costs. The World Health Organization has noted that most people with DM in developing countries are middle aged, whereas in developed nations diabetes predominantly affects patients who have either left the workforce or are close to retirement [22]. Chan et al. [23] suggested that the main increase in diabetes prevalence in China will occur in the 45- to 54-year-old group, during the most productive working years. Consequently, the overall benefit of detemir may be underestimated in this analysis. Source of financial support: These findings are the result of work supported by the Novo Nordisk (China) Pharmaceuticals Co., Ltd. The views expressed in this article are those of the authors.
[9]
[10]
[11]
[12]
Conclusion The findings of this evaluation indicate that therapy conversion to insulin detemir ⫾ OADs in patients with T2DM failing insulin glargine ⫾ OADs would be cost saving in China. So, the widespread use of detemir will help to reduce the economic burden of the nation and patients.
[13]
[14]
[15]
Supplemental Materials Supplemental material accompanying this article can be found in the online version as a hyperlink at doi:10.1016/j.jval.2011.11.018 or, if a hard copy of article, at www.valueinhealthjournal.com/ issues (select volume, issue, and article).
[16]
[17]
REFERENCES [18] [1] Yang WY, Lu JM, Weng JP, et al. Prevalence of diabetes among men and women in China. New Engl J Med 2010;362:1090 –101. [2] Hu SL, Liu G, Xu ZR, et al. Current status of epidemic and economic burden of diabetes mellitus in China. Chin Health Econ 2008:8:5– 8. [3] International Diabetes Federation. Diabetes Atlas (3rd ed.)[R]. 2007. Available from: http://www.diabetesatlas.org/map?id⫽13. [Accessed May 22, 2011]. [4] Xiang HD. The epidemiological state and preventive strategy of diabetes or chronic complications in China. Bull Med Res 2005;6:6 – 8. [5] Chen XB, Tang L, Chen HY, et al. Assessing the impact of complications on the costs of type 2 diabetes in urban China. Chin J Diabetes 2003;11:238 –91. [6] Stratton IM, Adler AI, Neil HAW, et al. Association of glycaemia with macrovascular and microvascular complications of type 2 diabetes (UKPDS 35): prospective observational study. BMJ 2000;321:405–12. [7] Pan CY. Control situation of diabetes— difference between guideline and practice (the outcome of diabetes control in 1998, 2001, 2003). Section Endocrinol Foreign Med Sci 2005;3:174 – 8. [8] Valentine WJ, Goodall G, Aagren M, et al. Evaluating the costeffectiveness of therapy conversion to insulin detemir in patients with
[19]
[20]
[21]
[22]
[23]
[24]
S59
type 2 diabetes in Korea: a modelling study of long-term clinical and cost outcomes. Adv Ther 2008;25:567– 84. Valentine WJ, Erny-Albrecht KM, Ray JA, et al. Therapy conversion to Insulin detemir among patients with type 2 diabetes treated with oral agents: a modeling study of cost-effectiveness in the United States. Adv Ther 2007;24:273–90. Palmer AJ, Valentine WJ, Ray JA, et al. An economic assessment of analogue basal– bolus insulin versus human basal– bolus insulin in subjects with type 1 diabetes in the UK. Curr Res Med Opin 2007;23: 895–901. Tunis SL, Minshall ME, Conner C, et al. Cost-effectiveness of insulin detemir compared to NPH insulin for type 1 and type 2 diabetes mellitus in the Canadian payer setting: modeling analysis. Curr Res Med Opin 2009;25:1273– 84. Gschwend MH, Aagren M, Valentine WJ, et al. Cost-effectiveness of insulin detemir compared with neutral protamine Hagedorn insulin in patients with type 1 diabetes using a basal-bolus regimen in five European countries. J Med Econ 2009;12:114 –23. Palmer AJ, Roze S, Valentine WJ, et al. The CORE Diabetes Model: projecting long-term clinical outcomes, costs and cost-effectiveness 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. Palmer AJ, Roze S, Valentine WJ, et al. Validation of the CORE diabetes model against epidemiological and clinical studies. Curr Med Res Opin 2004;20(Suppl. 1):S27– 40. Srishyla MV. Observational study to evaluate the safety of Levemir® in diabetes. Available from: http://clinicaltrials.gov/ct2/show/ NCT00706017?term⫽PREDICTIVE&rank⫽3. [Accessed May 22, 2011]. Jia WP, Pang C, Chen L, et al. Epidemiological characteristics of diabetes mellitus and impaired glucose regulation in a Chinese adult population: the Shanghai Diabetes Studies, a cross-sectional 3-year follow-up study in Shanghai urban communities. Diabetologia 2007; 50:286 –92. Koenen C. Observational study of safety and effectiveness of NovoMix® 30 for the treatment of diabetes (IMPROVE™). Available from: http:// clinicaltrials.gov/ct2/show/NCT00659282?term⫽novomix⫹30⫹therapy &rank⫽15. [Accessed May 22, 2011]. Yang GH, Ma JM, Liu N, et al. Smoking and passive smoking in Chinese, 2002. Chin J Epidemiol 2005;2:77– 83. China Health Care Association. The investigation report over smoking and drinking including Chinese twenty-five provinces in 2007. Available from: http://www1.chc.org.cn/news/detail.php?id⫽51296. [Accessed May 22, 2011]. Sharma SK, Joshi SR, Kumar A, et al. Efficacy, safety and acceptability of biphasic insulin aspart 30 in Indian patients with type 2 diabetes: results from the PRESENT study. J Assoc Phys India 2008;56:859 – 63. Palmer JL, Gibbs M, Scheijbeler HW, et al. Cost-effectiveness of switching to biphasic insulin aspart in poorly-controlled type 2 diabetes patients in China. Adv Ther 2008;25:752–74. National Bureau of Statistics of China. Statistical Communiqué of the People’s Republic of China on the 2008 National Economic and Social Development. Available from: http://www.stats.gov.cn/tjgb/. [Accessed May 22, 2011]. Currie CJ, Morgan CL, Poole CD, et al. Multivariate models of healthrelated utility and the fear of hypoglycaemia in people with diabetes. Curr Med Res Opin 2006;22:1523–34. World Health Organization. Making Choices in Health: WHO Guide to Cost-Effectiveness Analysis. Geneva: World Health Organization, 2003.
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/jval
Cost-Effectiveness of Switching Patients with Type 2 Diabetes from Insulin Glargine to Insulin Detemir in Chinese Setting: A Health Economic Model Based on the PREDICTIVE Study Li Yang, PhD1,*, Torsten Christensen, MSc2, Fengyu Sun, MD3, Jinghua Chang, MSc3 1 Department of Health Policy and Management, School of Public Health, Peking University, China; 2Novo Nordisk A/S, Virum, Denmark; 3Novo Nordisk (China) Pharmaceuticals Co., Ltd., Beijing, China
A B S T R A C T
Objectives: To evaluate the long-term cost-effectiveness of switching from insulin glargine (IGla) to insulin detemir (IDet) in patients with type 2 diabetes in the setting of Chinese secondary and tertiary hospitals. Methods: A published and validated computer simulation model of diabetes (the Center for Outcomes Research model) was used to make the long-term (30 years) projection of health economic outcomes. Patient demographic information and clinical end points were derived from a subgroup analysis of the Predictable Results and Experience in Diabetes through Intensification and Control to Target: an International Variability Evaluation (PREDICTIVE) study. Baseline risk factors and racial characteristic data were obtained from Chinese cohort studies. The diabetes management and complications costs were obtained from Chinese published data and adjusted to 2010 values by using the Chinese consumer price index. An annual discounting rate of 3% was used for both health and cost outcomes, and one-way sensitivities analysis was performed, which illustrated that the results were
Introduction The effective prevention and management of diabetes mellitus (DM) is an increasing serious challenge to China. The China National Diabetes and Metabolic Disorders Study estimated that the age-standardized prevalence of diabetes and prediabetes was 9.7% and 15.5%, respectively, accounting for 92.4 million adults with diabetes and 148.2 million adults with prediabetes in China [1]. According to another survey conducted by the Ministry of Health [2], type 2 DM (T2DM) constitutes about 95% of all patients and the total costs of DM in China were estimated to be $20.5 billion for 2007, which accounted for 14.2% of national health-care expenditures [3]. In China, DM-related complications are the key driver of medical costs. The results of the first national DM-related complications survey based on a sample of 24,496 patients with diabetes in China showed that 73.2% of patients with diabetes had one or more complications [4]. The research conducted by Chen et al. [5] revealed that the annual direct medical cost of patients with complications was increased by 3.71 times compared with that of those without complications.
robust. Results: Conversion to IDet from IGla was projected to improve patient life expectancy by 0.06 year and 0.48 quality-adjusted life-years. Drug costs and management costs of diabetes mellitus were increased by US$368 (US$17,466 vs. US$17,097) and US$31 (US$5464 vs. US$5433), respectively. However, the costs of complications, including cerebrovascular disease, renal complications, ulcer/ amputation/neuropathy, eye complications, and hypoglycemia events, were reduced by US$819 (US$21,294 vs. US$22,114), resulting in a total direct medical cost saving of US$420 when converting to IDet. Conclusions: Conversion to IDet from an IGla regimen improved life expectancy and was a cost-saving treatment approach in a Chinese setting. Keywords: Chinese setting, cost-effectiveness, cost-utility, insulin detemir. Copyright © 2012, International Society for Pharmacoeconomics and Outcomes Research (ISPOR). Published by Elsevier Inc.
In the treatment of DM, insulin plays an important role. Earlier use of insulin in patients with T2DM will help to better control the glucose level and reduce the risk of complications. Detemir and glargine are both long-acting human insulin analogues used for maintaining the basal level of insulin. They have been proved to more closely reflect physiological basal insulin and are associated with a significantly less risk of hypoglycemia and weight gain [6,7]. Two pharmacoeconomic studies conducted in Germany [8] and the United States [9] evaluated the cost-effectiveness of detemir based on the clinical results of the German PREDICTIVE study. Moreover, other researches from the United Kingdom [10], Canada [11], and Europe [12] also systematically evaluated the long-term health benefits of detemir caused by the reduction in the hemoglobin A1c (Hb A1c) level. Insulin detemir came to the Chinese market in 2010 and is being used in secondary and tertiary hospitals in China. To better inform the insurance coverage decision makers and the doctors whether it is value for money in the long run, it is necessary to assess the long-term cost-effectiveness of insulin detemir com-
Conflicts of interest: The authors have indicated that they have no conflicts of interest with regard to the content of this article. * Address correspondence to: Li Yang, PhD, Department of Health Policy and Management, School of Public Health, Peking University, No. 38 Xueyuan Road, Haidian District, Beijing 100191, China. E-mail: [email protected] or [email protected]. 1098-3015/$36.00 – see front matter Copyright © 2012, International Society for Pharmacoeconomics and Outcomes Research (ISPOR). Published by Elsevier Inc. doi:10.1016/j.jval.2011.11.018
VALUE IN HEALTH 15 (2012) S56 –S59
pared with that of the existing long-acting human insulin analogues such as insulin glargine. This study was conducted to quantify the long-term cost-effectiveness of insulin detemir compared with that of insulin glargine for the treatment of T2DM in China from a societal perspective, and to assess the sensitivity of results to disutility values.
Methods Model We used a validated computer simulation model of diabetes (the Center for Outcomes Research [CORE] model) specifically for diabetes to convert the short-term clinical effects to long-term economic outcomes [13]. The model is based on a series of submodels that simulate the important complications of DM. Each submodel is a Markov model using Monte Carlo simulation. The CORE model takes into account baseline cohort characteristics, history of complications, current and future diabetes management and concomitant medications, screening strategies, and changes in physiological parameters over time. The development of complications, life expectancy, quality-adjusted life-year (QALY), and total costs within populations can be calculated. Furthermore, the reliability of simulated outcomes has been tested [13,14].
S57
The management costs and DM-related complications costs were referred to the results of the survey conducted by the IMS Health in six tertiary and three secondary hospitals in China [21]. Complication costs refer to medical expenditure that resulted from diabetes-related complications. Management costs refer to administration devices and blood glucose monitoring and took into account OAD usage and variations in insulin dosage as noted in the PREDICTIVE study. The outpatient visit services and others including screening services were included. All the costs were inflated to 2010 values by using the consumer price index of health care, which was 3.2% in 2009, 2.9% in 2008, and 2.1% in 2007, respectively [22]. The annual medication costs were calculated and are shown in Table 2 (Cost data for management and diabetesrelated complications in China for 2010) in Supplemental Materials found at doi:10.1016/j.jval.2011.11.018.
Quality of life and utilities Quality-adjusted life expectancy was incorporated into the analysis by using diabetes-related health state utility and event disutility values published by Palmer et al. [13]. A disutility value of ⫺0.0035 was applied for each minor hypoglycemic event (an event not requiring hospitalization) recorded in the simulation. To capture the reduced utility associated with major hypoglycemic events, a disutility value of ⫺0.0118 was applied. Hypoglycemia disutility values were derived from a published source [23].
Treatment effects The treatment effects were derived from the Korean cohort of the PREDICTIVE study [15]. In the research, subjects who required changing treatment, as assessed by their treating physicians, were transferred to insulin detemir ⫾ oral antidiabetic drugs (OADs) from insulin glargine ⫾ OADs. The major reasons why the treating physicians transferred a patient to insulin detemir included the following: 1) improve glycemic control (89%); 2) reduce plasma variability (53%); 3) patient’s dissatisfaction with current therapy (33%); and 4) improve weight control (29%). Because PREDICTIVE is an observational study based on everyday clinical practice, the treating physicians could adjust the patient’s OADs and insulin detemir dose and frequency for the duration of the study according to the patient’s situation. The subgroup research of the Korean cohort of the PREDICTIVE study included 875 patients with T2DM who transferred to insulin detemir ⫾ OADs from insulin glargine ⫾ OADs. Reported changes in the Hb A1c level, body mass index (BMI), and hypoglycemic event rate were applied in the simulation as shown. After 3 months of treatment, subjects were associated with a reduction of 0.20% and 0.12 kg/m2 in the Hb A1c level and BMI, respectively. The hypoglycemic event rates were lower in patients using insulin detemir.
Simulation cohort A hypothetical simulation cohort was generated on the basis of patient demographics, risk factors, and preexisting complications. The patient demographics corresponded with the values from the Korean arm of the PREDICTIVE study [15]. Other data were extracted from published country-specific articles [16 –20] (see Table 1 [Inputs for Chinese simulation cohort with T2DM] in Supplemental Materials found at doi:10.1016/j.jval.2011.11.018). We defined this cohort including 1000 persons.
Costs and perspective Direct medical costs were accounted from a society perspective in 2010 US dollars. Direct costs were taken to be the sum of patients’ management costs and DM-related complications costs. The costs of lost productivity due to illness and death were excluded.
Discounting and time horizon All costs and clinical benefits were discounted at an annual rate of 3.0% in accordance with recommendations by the World Health Organization [24]. Furthermore, to reflect the chronic nature of T2DM and capture both mortality and the incidence of DM-related complications over patient lifetime, the clinical and economic impact of treatment with insulin detemir was prolonged to 30 years.
Sensitivity analysis Several one-way sensitivity analyses were performed to examine the influence of key input parameters on the outcomes projected by the model. The varying parameters included time horizon (10, 20, or 30 years), discounting rate (0% or 10%), hypoglycemic event rate (assume same), BMI (assume same), and the reduction in the Hb A1c level (assume Hb A1c benefits reduced by 50%).
Statistical methodology A simulated cohort of 1000 patients was run through the model 1000 times for each simulation (base-case and sensitivity analysis). The mean and standard deviation values were generated by using a nonparametric bootstrapping approach. One thousand mean values of incremental costs and incremental effectiveness in terms of qualityadjusted life expectancy were plotted on a cost-effectiveness plan. For treatment that was not dominant, it was planned to generate an acceptability curve by calculating the proportion of points below a range of willingness-to-pay threshold.
Willingness to pay There is no specific threshold to judge the cost-effectiveness of an intervention in China. According to World Health Organization recommendations [24], if the incremental costs per effectiveness are not more than three times the gross domestic product per capital, the intervention could be regarded as cost-effective. With gross domestic product per capital for China in 2010 at US$3750, we used a willingness-to-pay threshold of US$11,250 in this analysis to determine the likelihood of insulin detemir being considered cost-effective in China.
S58
VALUE IN HEALTH 15 (2012) S56 –S59
Results Clinical outcomes In the base-case analysis, transferring from insulin glargine ⫾ OADs to insulin detemir ⫾ OADs was associated with an improvement of 0.061 year (11.461 ⫾ 0.176 years vs. 11.522 ⫾ 0.177 years) in life expectancy per patient. Maybe it can be attributed to the reduction in the Hb A1c level. Capturing the quality of life similarly showed the benefits of converting to detemir treatment. After switching to detemir ⫾ OADs, the QALYs were increased from 7.457 to 6.974. It was associated with an increment of 0.484 QALYs per patient (see Table 3 [Summary of base-case results] in Supplemental Materials found at doi:10.1016/j.jval.2011.11.018). Therapy conversion to insulin detemir ⫾ OADs from insulin glargine ⫾ OADs was projected to reduce the cumulative incidence of eye complications, renal complications, foot complications, neuropathic complications, and most of cardiovascular complications. The most remarkable reduction was found in renal complications, with 6.37% reduction in end-stage renal complications (see Fig. 1 [Relative changes in the cumulative residence of complications over 30 years] in Supplemental Materials found at doi: 10.1016/j.jval.2011.11.018).
Cost outcomes Compared with insulin glargine ⫾ OADs, treatment with insulin detemir ⫾ OADs was associated with an increase of US$369 in lifetime drug costs and US$31 in management costs. However, in the long run, therapy conversion to detemir ⫾ OADs would reduce the incidence of DM-related complications, which reduced US$820 in complication costs. Finally, the higher drug costs and management costs of detemir ⫾ OADs were more than offset by decreased complications costs, giving a net saving of US$420 per patient (see Table 3 [Summary of base-case results] in Supplemental Materials found at doi:10.1016/j.jval.2011.11.018).
Evaluation of cost-effectiveness For patients with T2DM, projection over 30 years indicated that therapy conversion to insulin detemir ⫾ OADs would lower total treatment costs and increase life expectancy by 0.061 years or 0.484 QALYs per patient. So, we came to the conclusion that switching patients to treatment with detemir ⫾ OADs was the dominant choice, being cost saving in comparison with glargine ⫾ OADs (see Table 3 [Summary of base-case results] in Supplemental Materials found at doi:10.1016/j.jval.2011.11.018). The scatter plot that described incremental costs versus incremental effectiveness (in terms of QALY) for insulin detemir ⫾ OADs versus insulin glargine ⫾ OADs from the 1000 patients simulated 1000 times each in the model showed that almost all the points fell on the right side of the vertical axis. Half of the points were distributed in the lower right corner quadrant (more effective and less costly). It meant that insulin detemir ⫾ OADs was cost saving or cost-effective (see Fig. 2 in Supplemental Materials found at doi:10.1016/j.jval.2011.11.018). A cost-effectiveness acceptability curve was generated to evaluate the likelihood that insulin detemir ⫾ OADs would represent good value for money in China. At a willingness-to-pay threshold of US$3750 per QALY gained, insulin detemir ⫾ OADs had a 100% probability of being acceptable compared with insulin glargine ⫾ OADs from a social perspective. At a willingness-to-pay threshold of US$3750 per life-year gained, insulin detemir ⫾ OADs had 65% probability at least. Even at a willingness-to-pay threshold of zero per QALY, insulin detemir ⫾ OADs still had a 65% likelihood of being considered cost-effective (see Fig. 3 in Supplemental Materials found at doi:10.1016/j.jval.2011.11.018).
Sensitivity analysis A series of sensitivity analysis revealed that the findings in patients with T2DM were most sensitive to changes in the Hb A1c level and incidence of all hypoglycemia. Varying time horizons, discount rate, and BMI had no substantial impact on final results. Nevertheless, in all the sensitivity analyses undertaken, the incremental cost-effectiveness ratio of insulin detemir ⫾ OADs was always within the limit of willingness to pay in China (see Table 4 in Supplemental Materials found at doi:10.1016/j.jval.2011.11.018). In all scenarios of the sensitivity analysis, QALY in patients treated with insulin detemir ⫾ OADs was higher than in patients treated with insulin glargine ⫾ OADs all the time. So, insulin detemir ⫾ OADs was a cost-saving treatment.
Discussion Despite the availability of numerous treatment options to reduce blood glucose levels, the condition of most patients with T2DM in China was not adequately controlled. Insulin detemir has been introduced in the Chinese market successfully, which has increased the choice for patients. In consideration of the disease burden and the current situation of diabetes, however, we should select the most cost-efficient treatment strategies to control the increasing costs in this growing population. Although both insulin detemir and insulin glargine are longacting basal insulin, they differ at the molecular level, which may lead to different treatment effects. The PREDICTIVE study results not only confirm the difference between them but also provide the proof of conversion to detemir treatment after failure of therapy with glargine. Based on the short-term, patient-based results of the PREDICTIVE study, long-term projections using the CORE diabetes model demonstrated that treatment with insulin detemir ⫾ OADs was associated not only with improvements in life expectancy and QALY in comparison with teatment with insulin glargine ⫾ OADs for patients with T2DM but also reduction in total costs. According to modeling analysis and sensitivity findings, switching poorly controlled patients to insulin detemir ⫾ OADs is likely to be cost saving in China. This study is the first to compare the cost-effectiveness of insulin detemir and insulin glargine by applying the CORE model to the Chinese setting. What is more, this study is not limited to short-term treatment effects but long-term health outcomes. So, it is very useful for decision makers. It should be noted, however, that there are some limitations in this research, which must be pointed out. First, the CORE model consists of 15 Markov submodels. Every model will use the probabilities for transitions among different states. The transition probabilities were mainly derived from clinical trials and epidemiological studies conducted in white cohort. But medical practice in China has a big gap in comparison with Western countries. Transition probabilities in China may be higher than those used in the CORE model. The CORE model is used under Chinese medical practice, which likely underestimates the incidence of DM-related complications. Second, the use of data from the Korean cohort of the PREDICTIVE study to simulate treatment with insulin detemir ⫾ OADs in China inherently carries the assumption that the reported treatment effects are applicable to Chinese patients who share similar baseline characteristics with those of the Korean cohort and who are receiving a similarly high level of diabetes care. To overcome this limitation, it is reassuring that the treatment effects were varied in the sensitivity analysis. Even though the effect of insulin detemir ⫾ OADs treatment on the Hb A1c level was reduced by 50%, insulin detemir ⫾ OADs treatment was still cost-effective. Finally, this study considered only the direct costs. The indirect costs associated with lost productivity were not included. How-
VALUE IN HEALTH 15 (2012) S56 –S59
ever, the indirect costs may account for a large percentage of total costs. The World Health Organization has noted that most people with DM in developing countries are middle aged, whereas in developed nations diabetes predominantly affects patients who have either left the workforce or are close to retirement [22]. Chan et al. [23] suggested that the main increase in diabetes prevalence in China will occur in the 45- to 54-year-old group, during the most productive working years. Consequently, the overall benefit of detemir may be underestimated in this analysis. Source of financial support: These findings are the result of work supported by the Novo Nordisk (China) Pharmaceuticals Co., Ltd. The views expressed in this article are those of the authors.
[9]
[10]
[11]
[12]
Conclusion The findings of this evaluation indicate that therapy conversion to insulin detemir ⫾ OADs in patients with T2DM failing insulin glargine ⫾ OADs would be cost saving in China. So, the widespread use of detemir will help to reduce the economic burden of the nation and patients.
[13]
[14]
[15]
Supplemental Materials Supplemental material accompanying this article can be found in the online version as a hyperlink at doi:10.1016/j.jval.2011.11.018 or, if a hard copy of article, at www.valueinhealthjournal.com/ issues (select volume, issue, and article).
[16]
[17]
REFERENCES [18] [1] Yang WY, Lu JM, Weng JP, et al. Prevalence of diabetes among men and women in China. New Engl J Med 2010;362:1090 –101. [2] Hu SL, Liu G, Xu ZR, et al. Current status of epidemic and economic burden of diabetes mellitus in China. Chin Health Econ 2008:8:5– 8. [3] International Diabetes Federation. Diabetes Atlas (3rd ed.)[R]. 2007. Available from: http://www.diabetesatlas.org/map?id⫽13. [Accessed May 22, 2011]. [4] Xiang HD. The epidemiological state and preventive strategy of diabetes or chronic complications in China. Bull Med Res 2005;6:6 – 8. [5] Chen XB, Tang L, Chen HY, et al. Assessing the impact of complications on the costs of type 2 diabetes in urban China. Chin J Diabetes 2003;11:238 –91. [6] Stratton IM, Adler AI, Neil HAW, et al. Association of glycaemia with macrovascular and microvascular complications of type 2 diabetes (UKPDS 35): prospective observational study. BMJ 2000;321:405–12. [7] Pan CY. Control situation of diabetes— difference between guideline and practice (the outcome of diabetes control in 1998, 2001, 2003). Section Endocrinol Foreign Med Sci 2005;3:174 – 8. [8] Valentine WJ, Goodall G, Aagren M, et al. Evaluating the costeffectiveness of therapy conversion to insulin detemir in patients with
[19]
[20]
[21]
[22]
[23]
[24]
S59
type 2 diabetes in Korea: a modelling study of long-term clinical and cost outcomes. Adv Ther 2008;25:567– 84. Valentine WJ, Erny-Albrecht KM, Ray JA, et al. Therapy conversion to Insulin detemir among patients with type 2 diabetes treated with oral agents: a modeling study of cost-effectiveness in the United States. Adv Ther 2007;24:273–90. Palmer AJ, Valentine WJ, Ray JA, et al. An economic assessment of analogue basal– bolus insulin versus human basal– bolus insulin in subjects with type 1 diabetes in the UK. Curr Res Med Opin 2007;23: 895–901. Tunis SL, Minshall ME, Conner C, et al. Cost-effectiveness of insulin detemir compared to NPH insulin for type 1 and type 2 diabetes mellitus in the Canadian payer setting: modeling analysis. Curr Res Med Opin 2009;25:1273– 84. Gschwend MH, Aagren M, Valentine WJ, et al. Cost-effectiveness of insulin detemir compared with neutral protamine Hagedorn insulin in patients with type 1 diabetes using a basal-bolus regimen in five European countries. J Med Econ 2009;12:114 –23. Palmer AJ, Roze S, Valentine WJ, et al. The CORE Diabetes Model: projecting long-term clinical outcomes, costs and cost-effectiveness 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. Palmer AJ, Roze S, Valentine WJ, et al. Validation of the CORE diabetes model against epidemiological and clinical studies. Curr Med Res Opin 2004;20(Suppl. 1):S27– 40. Srishyla MV. Observational study to evaluate the safety of Levemir® in diabetes. Available from: http://clinicaltrials.gov/ct2/show/ NCT00706017?term⫽PREDICTIVE&rank⫽3. [Accessed May 22, 2011]. Jia WP, Pang C, Chen L, et al. Epidemiological characteristics of diabetes mellitus and impaired glucose regulation in a Chinese adult population: the Shanghai Diabetes Studies, a cross-sectional 3-year follow-up study in Shanghai urban communities. Diabetologia 2007; 50:286 –92. Koenen C. Observational study of safety and effectiveness of NovoMix® 30 for the treatment of diabetes (IMPROVE™). Available from: http:// clinicaltrials.gov/ct2/show/NCT00659282?term⫽novomix⫹30⫹therapy &rank⫽15. [Accessed May 22, 2011]. Yang GH, Ma JM, Liu N, et al. Smoking and passive smoking in Chinese, 2002. Chin J Epidemiol 2005;2:77– 83. China Health Care Association. The investigation report over smoking and drinking including Chinese twenty-five provinces in 2007. Available from: http://www1.chc.org.cn/news/detail.php?id⫽51296. [Accessed May 22, 2011]. Sharma SK, Joshi SR, Kumar A, et al. Efficacy, safety and acceptability of biphasic insulin aspart 30 in Indian patients with type 2 diabetes: results from the PRESENT study. J Assoc Phys India 2008;56:859 – 63. Palmer JL, Gibbs M, Scheijbeler HW, et al. Cost-effectiveness of switching to biphasic insulin aspart in poorly-controlled type 2 diabetes patients in China. Adv Ther 2008;25:752–74. National Bureau of Statistics of China. Statistical Communiqué of the People’s Republic of China on the 2008 National Economic and Social Development. Available from: http://www.stats.gov.cn/tjgb/. [Accessed May 22, 2011]. Currie CJ, Morgan CL, Poole CD, et al. Multivariate models of healthrelated utility and the fear of hypoglycaemia in people with diabetes. Curr Med Res Opin 2006;22:1523–34. World Health Organization. Making Choices in Health: WHO Guide to Cost-Effectiveness Analysis. Geneva: World Health Organization, 2003.