Cost-Effectiveness of Insulin Degludec Versus Insulin Glargine U100 in Patients with Type 1 and Type 2 Diabetes Mellitus in Serbia

Cost-Effectiveness of Insulin Degludec Versus Insulin Glargine U100 in Patients with Type 1 and... Diabetes Ther (2018) 9:1201–1216 https://doi.org/10.1007/s13300-018-0426-0 ORIGINAL RESEARCH Cost-Effectiveness of Insulin Degludec Versus Insulin Glargine U100 in Patients with Type 1 and Type 2 Diabetes Mellitus in Serbia . . . Nebojsˇa Lalic´ Monika Russel-Szymczyk Marina Culic Christian Klyver Tikkanen Barrie Chubb Received: February 23, 2018 / Published online: April 26, 2018 The Author(s) 2018 effectiveness was analysed in a 1-year setting, ABSTRACT based on data from clinical trials. Costs were estimated from the healthcare payer perspec- Introduction: This study investigates the cost- tive, the Serbian Health Insurance Fund (RFZO). effectiveness of insulin degludec versus insulin The outcome measure was the incremental cost- glargine U100 in patients with type 1 and type 2 effectiveness ratio (ICER) or cost per quality- diabetes mellitus in Serbia. adjusted life-year (QALY) gained. Methods: A cost-utility analysis, implementing Results: Degludec is highly cost-effective com- a simple short-term model, was used to compare pared with glargine U100 for people with T1DM treatment costs and outcomes with degludec and T2DM in Serbia. The ICERs are estimated at versus glargine U100 in patients with type 1 417,586 RSD/QALY gained in T1DM, 558,811 (T1DM) and type 2 diabetes (T2DM). Cost- RSD/QALY gained in T2DM on basal oral ther- Enhanced digital features To view enhanced digital apy (T2DM ) and 1,200,141 RSD/QALY BOT features for this article go to https://doi.org/10.6084/ gained in T2DM on basal-bolus therapy m9.figshare.6106889. (T2DM ). All ICERs fall below the commonly B/B Electronic supplementary material The online accepted thresholds for cost-effectiveness in version of this article (https://doi.org/10.1007/s13300- Serbia (1,785,642 RSD/QALY gained). In all 018-0426-0) contains supplementary material, which is three patient groups, insulin costs are higher available to authorized users. with degludec than with glargine U100, but these costs are partially offset by savings from a N. Lalic´ Faculty of Medicine, Clinic for Endocrinology, lower daily insulin dose in T1DM and Diabetes and Metabolic Diseases, Clinical Center of T2DM , a reduction in hypoglycaemic events BOT Serbia, University of Belgrade, Belgrade, Serbia in all three patient groups and reduced costs of M. Russel-Szymczyk (&) SMBG testing in the T2DM groups with deglu- Novo Nordisk Pharma Sp. z o.o., Warsaw, Poland dec versus glargine U100. e-mail: zmns@novonordisk.com Conclusion: Degludec is a cost-effective alter- native to glargine U100 for patients with T1DM M. Culic Novo Nordisk Pharma d.o.o., Belgrade, Serbia and T2DM in Serbia. Degludec may particularly benefit those suffering from hypoglycaemia or C. K. Tikkanen where the patient would benefit from the Novo Nordisk Scandinavia AB, Copenhagen, Denmark option of flexible dosing. Funding: Novo Nordisk. B. Chubb Novo Nordisk Ltd, Gatwick, UK 1202 Diabetes Ther (2018) 9:1201–1216 Keywords: Cost-effectiveness; Insulin degludec; continuous subcutaneous insulin infusion Insulin glargine; Type 1 diabetes mellitus; Type (CSII) by means of an external portable insulin 2 diabetes mellitus; Serbia pump [8]. Type 2 diabetes (T2DM) is a progres- sive disease and treatment is intensified with disease progression. A significant proportion of INTRODUCTION patients with T2DM will eventually need insu- lin therapy to achieve optimal blood glucose Globally an estimated 415 million people have targets [8]. Insulin is the most effective method diabetes, and this is projected to increase to 642 of reducing blood glucose concentrations; million by 2040 [1]. According to the estimates however, despite clear guidance, glycaemic of the Serbian Institute of Public Health, dia- control remains suboptimal (HbA [ 7.0%) in a 1c betes affects 12.4% of the adult population in substantial number of patients [9, 10]. Key bar- Serbia [2]. Diabetes is a chronic metabolic dis- riers to insulin therapy include fear and risk of order characterised by high blood sugar levels hypoglycaemia, weight gain and restrictive (hyperglycaemia) and is a leading cause of treatment regimens [11, 12]. blindness, end-stage renal failure, heart attack/ Insulin degludec (degludec) is a basal insulin stroke, non-traumatic amputation and depres- analogue with a long duration of action and a sion [1]. Diabetes and related complications are distinct, slow absorption mechanism, which chief causes of death in most countries, and results in a flat and stable action profile [13, 14]. according to the International Diabetes Federa- It has four times less day-to-day variability in tion (IDF), there were an estimated 10,616 dia- glucose lowering effect than insulin glargine betes-related deaths in the 20–79 age group in U100 (glargine U100) [15]. Due to its long and Serbia in 2015 [1]. stable action profile, degludec allows for flexi- The treatment of diabetes has a substantial bility of dosing time. On occasions when economic impact on national healthcare sys- administration at the same time of day is not tems. The treatment of diabetes-related possible, patients can dose their insulin at any chronic complications accounts for the largest time of the day as long as there is a minimum of proportion of diabetes direct medical costs 8 h between doses [16–18]. [3, 4]. Indirect costs associated with diabetes, The efficacy and safety of degludec have such as absenteeism from work and reduced been evaluated in a large-scale clinical devel- productivity while at work, are also consider- opment programme. The degludec phase 3 able [3]. The most recent published cost data clinical trial programme (BEGIN) included more for Serbia are from 2007 [5] and estimate that than 9000 patients with T1DM and T2DM [19]. annual diabetes expenditures were 53,412.96 Meta-analyses of the phase 3a clinical trials RSD (or 444.42 EUR ) per patient. Indirect costs showed that degludec achieved equivalent represented 10.39% of total costs, while anti- reductions in HbA with a lower risk of hypo- 1c diabetes drugs and medical devices represented glycaemia versus glargine U100 and at a signif- 44% (38.10% and 5.94%, respectively) and icantly lower total daily insulin dose compared hospital and ambulatory costs the remaining with glargine U100 in T1DM (12% lower) and 45%. T2DM basal oral therapy (10% lower) [20, 21]. The aim of diabetes therapy is to keep blood Real-world studies confirm the observations glucose levels within recommended targets and from the clinical trials and support the effec- ultimately limit the development of diabetes- tiveness of degludec in clinical practice [22, 23]. related complications [6, 7]. All patients with Due to increasing constraints on healthcare type 1 diabetes (T1DM) require insulin. Current budgets, it is important that new therapies clinical guidance in Serbia recommends a basal- represent good value for the money. Decision- bolus regimen or for selected individuals making based on both clinical and economic evidence helps healthcare providers optimise 1 EUR = 120.185 RSD: Rulebook on the conditions, resource use and care for patients with diabetes. criteria, method and procedure for placing the medicinal Cost-effectiveness models assess the value of products on the reimbursement list. Diabetes Ther (2018) 9:1201–1216 1203 interventions by comparing the relative costs considered where a basal insulin analogue is and outcomes. An economic analysis typically indicated. estimates the difference in cost between one The model was a simple, transparent, short- healthcare intervention and an alternative, term (1-year time horizon) model developed in divided by the difference in health effects, Microsoft Excel 2010 (Microsoft Corp., Red- which is termed an incremental cost-effective- mond, WA, US) and has been previously pub- ness ratio (ICER). A commonly used effective- lished [26]. The short-term model focuses on ness measure in cost-effectiveness analyses is the impact of important aspects of insulin the quality-adjusted life year (QALY), which therapy, such as hypoglycaemia and dosing, measures health as a combination of both and accommodates the treat-to-target trials duration and health-related quality of life [24]. required by the Food and Drug Administration The incremental cost per additional QALY (FDA) [27]. In treat-to-target trials patients are gained (cost/QALY) allows decision makers to treated to the same glycaemic target and no compare across different disease areas to allo- differences in HbA are expected; thus, there is 1c cate healthcare resources for maximal economic no rationale for long-term modelling based on and clinical benefits. A financial threshold is HbA differences. 1c often set at which cost-effectiveness is accepted. Although cost-effectiveness was analysed in In Serbia, there is no officially indicated a 1-year setting and is based on data from 1-year threshold; however, the level of 3 9 GDP per clinical trials, the model can be replicated for capita suggested by the World Health Organi- subsequent years, and the outcomes represent zation (WHO) [25] is accepted by the Serbian the average annual cost-effectiveness in steady authorities. state. As the time horizon was 1-year, no dis- The objective of this study was to evaluate the counting was applied. cost-effectiveness of degludec versus glargine The model calculated the costs associated U100 from the healthcare payer perspective in with treatment (insulin, needles and SMBG test Serbia [Serbian Health Insurance Fund (RFZO )]. strips) and hypoglycaemic events (the cost of treating the event and the use of additional SMBG test strips associated with the event). METHODS Health-related quality of life (in the form of QALYs) was calculated by applying a disutility Model Overview per hypoglycaemic event incurred (Fig. 1). The type of cost-effectiveness analysis used was Perspective/Viewpoint a cost-utility analysis. Degludec was compared with glargine U100 in three patient All costs were estimated from the healthcare populations: payer perspective, RFZO. The threshold was • T1DM using basal-bolus therapy (T1DM) setat3 9 GDP per capita, as although there • T2DM using basal oral therapy (T2DM ) BOT are alternative approaches to thresholds for • T2DM using basal-bolus therapy (T2DM ) B/B cost-effectiveness of interventions [28], this is Glargine U100 was identified as the most an internationally recognised level [25]and is appropriate comparator for the cost-utility accepted by the Serbian authorities. According analysis. Treatment guidelines in Serbia recom- to the World Bank statistics for 2016, the GDP mend NPH insulin as the first-line insulin per capita in Serbia was 5348.30 USD. Using treatment for T1DM and T2DM, with a basal the exchange rate of 1 USD = 111.29 RSD, the insulin analogue considered if after 6 months threshold was set at the level of 1,785,642 on NPH there is persisting hypoglycaemia RSD. (\ 3.5 mmol/l). Thus, degludec should be Republicˇki fond za zdravstveno osiguranje. 1204 Diabetes Ther (2018) 9:1201–1216 Fig. 1 Overview of cost-effectiveness model. HCP health- ratio, IDeg insulin degludec, IGlar U100 insulin glargine, care professional, HRQoL health-related quality of life, QALY quality-adjusted life year, SMBG self-monitored hypo hypoglycaemia, ICER incremental cost-effectiveness blood glucose Data Used in the Model severe events data were obtained from the seri- ous adverse events case reports [29]. The use of additional SMBG tests in the week Cost Data following a non-severe event was also based on All costs were estimated from the healthcare the hypoglycaemia safety questionnaire. Data payer perspective, RFZO. were not collected on the testing pattern fol- lowing a severe event; therefore, it was conser- Direct Treatment Costs vatively assumed to be similar to a non-severe Costs of insulin, needles and SMGB tests were hypoglycaemic event. based on official RFZO prices elicited in For patients experiencing severe events it November 2017. Other costs of treatment (e.g. was assumed that all patients who were hospi- use of concomitant medication) or other costs talised used the ambulance service and that all resulting from treatment (e.g. long-term out- patients experiencing severe events used gluca- comes) were assumed to be equivalent in both gon to recover. treatment groups and were therefore not Patients suffering a hypoglycaemic event included. received the same treatment regardless of whe- ther they were on degludec or glargine U100. Therefore, any difference in costs of hypogly- Cost of Hypoglycaemic Events caemia between treatments was due to differ- Resource use associated with non-severe and ences in rates of hypoglycaemia and not the severe hypoglycaemic events was derived from cost per event. the clinical trial data. For non-severe events The average costs of hypoglycaemic events data were obtained from the patient-completed are calculated by multiplying the unit cost of hypoglycaemia safety questionnaire, and for the services by the share of patients using that Diabetes Ther (2018) 9:1201–1216 1205 Table 1 Total cost of an average severe/non-severe hypoglycaemic event Unit cost (RSD) Utilisation per hypoglycaemic event T1DM T2DM T2DM B/B BOT B/B Severe Non-severe Severe Non-severe Severe Non-severe Glucagon 1695.70 1.0 0.0 1.0 0.0 1.0 0.0 Ambulance 3430.00 0.14 0.0 0.88 0.0 0.28 0.0 bc Hospital 3980.00 0.14 0.0 0.88 0.0 0.28 0.0 SMBG 46.00 1.46 1.46 1.91 1.91 1.98 1.98 TOTAL (RSD) 3357.46 67.16 11,806.76 87.86 4975.98 91.08 TOTAL RSD values are given in bold B/B basal-bolus, BOT basal oral therapy, SMBG self-measured blood glucose, T1DM type 1 diabetes mellitus, T2DM type 2 diabetes mellitus, RSD Serbian dinar RFZO listed price (November 2017) HNF Rulebook on cost of services on the secondary/tertiary level indicators (July 2016) Hospitalisation cost includes 2 days of hospital stay = 7960.00 RSD (http://www.vma.mod.gov.rs/cenovnik-vma.pdf) treatment/service (Table 1). It was assumed that the end of trial doses captured from the clinical a non-severe event would be self-managed and trial data. The meta-analysis of insulin dose not require any additional resource use; thus, from the clinical trials was the source of the costs relate only to additional SMBG testing. glargine U100 dose and degludec/glargine U100 dose ratio [21]. The degludec dose was calcu- lated using the dose ratio to allow for adjust- Clinical Data ment of covariate factors such as trial, treatment, antidiabetic therapy at screening, Insulin Dose age, sex, region and baseline dose (Table 2). The daily insulin dose for the degludec and glargine U100 treatment groups was based on Table 2 Basal and bolus insulin use Treatment group Glargine U100 (units/day) Dose ratio (degludec/glargine U100) Degludec (units/day) T1DM , total dose 0.88* B/B Basal insulin 33.10 0.87* 28.80 Bolus insulin 35.00 0.88* 30.80 T2DM , total dose 0.90* BOT Basal insulin 51.70 0.90* 46.53 Bolus insulin Not relevant Not relevant Not relevant T2DM , total dose NS B/B Basal insulin 66.60 1.08* 71.93 Bolus insulin 72.70 NS 72.70 B/B basal-bolus, BOT basal oral therapy, T1DM type 1 diabetes mellitus, T2DM type 2 diabetes mellitus *p \ 0.05; NS non-significant; in the case of non-significant results, a relative rate of one was used in the calculation 1206 Diabetes Ther (2018) 9:1201–1216 Table 3 Calculation of hypoglycaemic event rates T1DM T2DM T2DM B/B BOT B/B Non-severe Severe Non-severe Severe Non-severe Severe Baseline hypoglycaemia 91.0 0.7 20.3 0.10 35.4 0.2 rate for glargine U100 Daytime/nocturnal split Daytime Nocturnal – Daytime Nocturnal – Daytime Nocturnal – 78% 22% 68% 32% 78% 22% Total events per patient 70.98 20.02 0.7 13.80 6.50 0.10 27.61 7.79 0.20 per year for glargine U100 Degludec/glargine U100 NS 0.83* NS NS 0.64* 0.14* 0.83* 0.75* NS hypoglycaemic event rate ratio Calculated degludec 70.98 16.62 0.7 13.80 4.10 0.01 22.92 5.84 0.20 hypoglycaemic event rate B/B basal-bolus, BOT basal oral therapy, T1DM type 1 diabetes mellitus, T2DM type 2 diabetes mellitus *p \ 0.05; NS, non-significant; in the case of non-significant results, a relative rate of one was used in the calculation Taken from Ostenson et al. [33] Proportion of daytime nocturnal events for glargine U100 taken from Ostenson et al. [33] SMBG Testing Hypoglycaemia Event Rates It was assumed that one SMBG test was con- Real-world hypoglycaemic event rates from a ducted with every main meal for regimens large-scale questionnaire-based study con- including bolus insulin (T1DM , T2DM ). ducted in seven European countries [33] were B/B B/B Three meals per day were assumed. used as the baseline values for severe and non- For SMBG tests related to basal insulin severe hypoglycaemic events. These rates pro- injections, the titration schedule recommended vide a better estimation of real-life event rates for use with glargine U100 [30] was used to than those from clinical trials [34], which can estimate SMBG utilisation for glargine U100. be biased in both the selection of patients and The algorithm recommends seven SMBG tests the treatment setting. For example, the clinical per week (the dose is adjusted every 3rd day trials excluded patients with a history of severe based on the mean of the FPG results over three hypoglycaemic episodes and anyone considered consecutive days) [31]. hypoglycaemia unaware [34–39]. The real-world Degludec enables patients to titrate, predict rates were used as the base case event rates in and monitor their blood glucose more effi- the glargine U100 group (as this is the current ciently [13, 14]. The recommended titration treatment on the market); see Table 3. Event algorithm for degludec is once-weekly adjust- rates for the degludec group (Table 3) were cal- ment of dose based on the average of two culated using the relative event rates taken from SMBG measurements from the 2 preceding the meta-analysis of hypoglycaemia [20, 21], days for patients with T2DM [32]. Thus, the which were adjusted for trial, type of diabetes, utilisation of SMBG tests related to degludec treatment, anti-diabetic therapy at screening, was assumed to be two tests per week for sex and region as fixed factors and age as a patients with T2DM. continuous covariate. Only rate ratios with a Diabetes Ther (2018) 9:1201–1216 1207 documented statistically significant difference glargine U100 group, with the majority repre- between the treatment arms were used. senting pharmacy costs. The total cost per patient per year is 11,990.32 RSD higher in the degludec group than in the glargine U100 group Compliance with Ethics Guidelines (6.9% higher costs), largely because of the This article does not contain any studies with increase in insulin costs. The costs of hypogly- human participants or animals performed by caemic events are similar for degludec and any of the authors. glargine U100 since only the non-severe noc- turnal events showed a statistically significant Utility Data difference between the two treatments, and the severe and daytime events were unchanged. QALYs were calculated by subtracting a disutil- In T2DM , the total cost per patient per BOT ity [40] per hypoglycaemic event experienced year in the degludec group (101,270.97 RSD) is from the baseline health utility. Disutilities 15,042.55 RSD (17.4%) higher than in the associated with hypoglycaemic events were glargine U100 group (86,228.42 RSD per patient obtained from a large-scale TTO study [40], per year). The difference in cost is due to the which reported a disutility of 0.0565 for a severe increased cost for insulin and the lower costs of event (with no significant difference between SMBG testing and non-severe nocturnal and daytime and nocturnal severe events) and severe hypoglycaemic events in the degludec disutilities of 0.0041 and 0.0067 for non-severe group. Lower costs of hypoglycaemia are driven daytime and non-severe nocturnal events, by the significant reduction in the number of respectively (a significant difference in utility severe hypoglycaemic events with degludec vs. was demonstrated for nocturnal vs. daytime glargine U100 in this patient group. non-severe events) [40]. The disutility per In T2DM , the total cost per patient per B/B hypoglycaemic event was multiplied by the year in the degludec group (285,584.00 RSD) is number of events observed in each treatment 45,912.88 (19.2%) higher than in the glargine group. U100 group (239,671.12 RSD per patient per In addition, the analysis included an esti- year), caused primarily by a higher spend for mate of the utility benefit for the option of insulin partly offset against a lower spend on flexible dosing time with degludec. Boye et al. SMBG and lower costs of non-severe daytime [41] reported a utility benefit of 0.006 associated and nocturnal hypoglycaemia. The higher with dosing flexibility; therefore, for degludec incremental cost for this group is driven mainly an extra utility gain of ? 0.006 was applied to by the slightly higher dose of basal insulin the QALY benefit. required in the degludec arm of the clinical trial. In the T2DM clinical trial [14], high B/B Sensitivity Analyses insulin doses were observed in both the deglu- dec and glargine U100 treatment arms. The One-way and probabilistic sensitivity analyses high doses observed in the trial are not expected were conducted to assess the impact of varying to be representative of a real-world setting for key assumptions and outcomes used in the base patients initiating a basal-bolus regimen as the case analysis (Table 4). trial mainly recruited patients who were already uncontrolled on an intensive basal-bolus regi- men or uncontrolled on a pre-mixed insulin RESULTS regimen. In a recent real-world study of deglu- dec, the daily basal and bolus insulin doses in Costs patients with T2DM (75% on a basal-bolus reg- imen) were considerably lower than those in In T1DM, total costs are estimated at 185,627.84 the clinical trial [23]. RSD per patient per year in the degludec group Average total costs for the three treatment and 173,637.52 per patient per year in the groups are presented in Table 5. 1208 Diabetes Ther (2018) 9:1201–1216 Table 4 Description of sensitivity analyses conducted Time horizon In the base case analysis the time horizon was 1 year. Sensitivity analysis increased the time horizon to 5 years Hypoglycaemia rates The base case rate of hypoglycaemia was taken from the published literature as it is believed that this provides a more realistic event rate. Additional published event rates, and the actual reported rates from the clinical trial programme, were investigated in sensitivity analyses SMBG costs SMBG costs were varied ± 20% Hypoglycaemia costs A sensitivity analysis was conducted where the cost of a non-severe hypoglycaemic event included the cost of one visit to the GP following the event (192.11 RSD) For severe hypoglycaemic events costs were varied ± 20% Dosing A sensitivity analysis was conducted with an assumption of equal doses for degludec and glargine U100 Injection frequency For some patients, current basal insulins need to be taken twice daily to ensure optimal control. The effect of using twice as many needles for the basal injections in the glargine U100 group was explored SMBG tests per week For degludec, the long duration of action and stable action profile [13, 14] mean that patients are able to titrate, predict and monitor their blood glucose more efficiently. The recommended titration algorithm for degludec is once-weekly adjustment of dose based on the average of two SMBG measurements from the 2 preceding days for patients with T2DM (98). The number of SMBG tests for basal titration with degludec was varied between 2 and 7 (T2DM ) and between 23 and 28 BOT (T2DM ) B/B Additional SMBG tests after In the base case analysis the number of additional SMBG tests following a non-severe non-severe events event was taken from the clinical trial data. A sensitivity analysis was conducted, which used a published source that reported 6.2 additional tests following a non- severe event [42] and another analysis that assumed no additional tests Flexible dosing utility A number of sensitivity analyses were conducted around this assumption: one where only 50% of patients were assumed to receive the benefit of flexible dosing, one that used an alternative published utility value (0.013 [43]) and one that assumed no utility from flexible dosing Probabilistic sensitivity analysis PSA used the standard errors and appropriate distributions of the parameters. The (PSA) distributions were assumed to be either normal or lognormal and each individual parameter was selected independently. The probabilistic sensitivity analyses were run with 1000 iterations. In the primary analysis for each of the groups—the standard error was only applied to differences that were statistically significant (i.e. if there was no statistical significance proven—then the rate ratio was set to 1 (assumed equivalent) and the SE was set to 0 (so as not to introduce random uncertainty) Diabetes Ther (2018) 9:1201–1216 1209 Table 5 Results: average total costs (RSD) per patient per year and incremental cost-effectiveness T1DM T2DM T2DM BOT B/B Cost, degludec (RSD) 185,627.84 101,270.97 285,584.00 Cost, glargine U100 (RSD) 173,637.52 86,228.42 239,671.12 D costs 11,990.32 15,042.55 45,912.88 D QALYs 0.0287 0.0269 0.0383 ICER (RSD/QALY) 417,586.27 558,811.00 1,200,140.55 Incremental Cost-Effectiveness Key parameters that influenced the ICER values were hypoglycaemia rates (T1DM, T2DM , T2DM ), insulin doses (T1DM, BOT B/B Degludec is a highly cost-effective treatment T2DM ) and the number of SMBG tests used option versus glargine U100 in patients with BOT weekly by patients (T2DM , T2DM ). BOT B/B T1DM as part of a basal-bolus insulin regimen For T1DM, the ICER remains below the cost- with an estimated ICER of 417,586.27 RSD/ effectiveness threshold in all scenarios tested in QALY gained. The result is primarily driven by one-way sensitivity analyses. ICERs range the significantly lower dose required, QALY between 297,452 RSD/QALY gained and gains due to significantly fewer non-severe 1,179,023 RSD/QALY gained. Rates of non-sev- nocturnal hypoglycaemic events versus glargine ere hypoglycaemia are influential on the ICER U100 and the opportunity to use flexible dosing in this group. When the annual number of non- when needed. This result falls well beneath the severe daytime events is reduced to 36.16 using cost-effectiveness threshold in Serbia of data from Donnelly et al. [34], the ICER 1,785,642 RSD/QALY gained. increases to 997,799 RSD/QALY gained. Con- In T2DM degludec is also highly cost-ef- BOT versely, when the annual number of non-severe fective (558,811.00 RSD/QALY gained), driven nocturnal events is increased to 20.80 using by cost savings from a significantly lower dose, a data from Ericsson et al. [44], the ICER is significant reduction in severe hypoglycaemia reduced to 404,428 RSD/QALY gained. The and the potential for once-weekly SMBG test- insulin dose ratio also has an impact on the ing. Due to the significantly fewer nocturnal ICER, but even under an extreme assumption of and severe hypoglycaemic events in this group equal dosing of basal insulin, degludec remains and the opportunity to use a flexible dosing highly cost-effective (720,507 RSD/QALY time when needed, degludec is associated with a gained). If the utility gain from flexibility is QALY gain versus glargine U100. excluded, the ICER is still within commonly In the T2DM setting, degludec is cost-ef- B/B accepted thresholds (527,896 RSD/QALY fective with an ICER of 1,200,140.55 RSD/QALY gained). gained, which falls below commonly accepted Similarly for T2DM , the favourable cost- BOT cost-effectiveness threshold. Degludec is asso- effectiveness results are invariant to changes in ciated with significantly fewer non-severe day- most of the input parameters. ICERs range time and nocturnal events, which results in a between 294,932 and 1,955,556 RSD/QALY QALY gain versus glargine U100. gained. The number of SMBG tests per week is influential on the ICER in this setting, but Sensitivity Analysis degludec is still cost-effective (1,004,600 RSD/ QALY gained) even under the strict assumption Extensive sensitivity analyses demonstrate that of no reductions in SMBG testing in the deglu- the results are robust and largely insensitive to dec group. Insulin dose ratio and utility gain changes in input parameters (results presented from flexibility also have an impact on the in supplementary table). results, but ICERs fall below the commonly 1210 Diabetes Ther (2018) 9:1201–1216 accepted threshold even with extreme assump- available, economic as well as clinical evidence tions (919,928 and 719,091 RSD/QALY gained is essential for healthcare providers to maximise respectively). Varying the rates of hypogly- healthcare with constrained budgets. caemia has an impact on the ICER in this group This simple, short-term economic evaluation reflecting the large benefits in terms of reduced demonstrates that degludec is likely to be con- hypoglycaemia with degludec versus glargine sidered highly cost-effective compared with U100. Drawing on higher event rates for non- glargine U100 for people with T1DM and T2DM severe events generates highly cost-effective in Serbia. All ICERs fall below the commonly results for degludec versus glargine U100; con- accepted thresholds for cost-effectiveness in versely, using lower event rates for severe Serbia. hypoglycaemic events, e.g. from Ratner [20], increases the ICER to 1,955,556.24 RSD/QALY gained, which is just above the commonly accepted threshold. For T2DM the parameter that has the most B/B influence on the ICER is the rate of hypogly- caemia. Varying the rate of non-severe hypo- glycaemia has an impact on the ICER in this group due to the significant reduction of non- severe events with degludec versus glargine U100. ICERs range between 980,305 and 3,636,384 RSD/QALY gained, with low rates of non-severe hypoglycaemia generating the higher ICERs. In all other scenarios analysed the ICERs are below the commonly accepted threshold (range 938,549–1,513,491 RSD/QALY gained), indicating that degludec is a cost-ef- fective therapeutic option in this group. Probabilistic Sensitivity Analysis To capture the uncertainty of the results caused by statistical uncertainty with respect to the parameter inputs, probabilistic sensitivity anal- ysis has been conducted. The probabilistic sen- sitivity analyses have each been run with 1000 iterations. The cost-effectiveness acceptability curves show that at a willingness-to-pay threshold of 2,048,112 RSD/QALY gained, the probability that degludec is cost-effective relative to glar- gine U100 is 77.5, 97.3 and 85.7% for T1DM, T2DM and T2DM , respectively (Fig. 2). BOT B/B DISCUSSION There are limited data on the cost-effectiveness of treatments for diabetes in Serbia. With an Fig. 2 Probabilistic sensitivity analysis—cost-effectiveness increasing number of insulin formulations acceptability curves Diabetes Ther (2018) 9:1201–1216 1211 In all three patient groups, insulin costs are flexible once-daily injection was associated with higher with degludec than with glargine U100, a 0.016 higher utility versus a fixed time of but these costs are partially offset by savings injection [43]. from a lower daily insulin dose in T1DM and Extensive sensitivity analyses show that the T2DM , a reduction in hypoglycaemic events results are robust and invariant to changes in BOT in all three patient groups and reduced costs of most of the input parameters. In all three SMBG testing in the T2DM groups, with deglu- patients groups the ICER remained below the dec versus glargine U100. For degludec, the threshold of 1,785,642 RSD in all scenarios tes- longer half-life and the corresponding long flat ted, except when low rates of non-severe insulin profile in steady state [15] and lower hypoglycaemia were applied in the T2DM B/B variability over the day compared with glargine setting and low rates of severe hypoglycaemia U100 [45] mean that patients are able to titrate, applied in the T2DM setting. PSA shows that BOT predict and monitor their blood glucose more there is a high probability that degludec will be efficiently. Consequently, degludec can be cost-effective versus glargine U100 in all three monitored with less frequent SMBG testing in patient groups. patients with T2DM. The titration algorithm for Hypoglycaemia can have a major impact on degludec is once-weekly adjustment of dose patients’ lives affecting employment, driving, based on the average of two SMBG measure- relationships, travel and leisure activities. Sev- ments from the 2 preceding days for patients ere episodes can result in serious clinical with T2DM [46]. The concept of fewer SMBG sequelae such as physical (e.g. fractures, head tests per week is supported by a phase 3b trial, injuries), neurological (e.g. convul- which examined different scenarios of SMBG sions/seizures, paralysis, coma) and cardiac (e.g. testing in patients with T2DM. The trial arrhythmia, cardiac failure) morbidity, as well demonstrated that degludec was effective and as death in some extreme cases [49, 50]. Fur- well tolerated using only one SMBG test per thermore, recurrent episodes, whether severe or week to titrate the insulin dose compared with non-severe, can lead to the development of titration based on three consecutive pre-break- impaired awareness of hypoglycaemia; this can fast SMBG values in patients with T2DM [47]. be problematic as it means that patients are less No statistically significant differences in HbA , able to recognise their symptoms and take 1c body weight, hypoglycaemia or any other safety remedial action. A recent non-interventional, parameters were observed between the two prospective-cohort survey conducted across titration algorithms [47]. 2004 sites in 24 countries, including Serbia, An important aspect to consider, especially investigated the prevalence of hypoglycaemia for chronic diseases such as diabetes mellitus, is [51]. The study included 27,585 patients with the quality of life of patients. With degludec insulin-treated diabetes and reported a higher higher gains in health-related quality of life frequency of hypoglycaemia than previously (HRQoL) were observed [48]. Incremental QALY reported, with marked variation across geo- gains with degludec versus glargine U100 in all graphic regions. Overall rates of hypoglycaemia three groups are associated with the reduction in Eastern Europe were 66.9 events/person/year in hypoglycaemic events and the opportunity for T1DM and 23.7 events/person/year for to use flexible dosing when needed. The current T2DM, with the highest rates in T2DM observed analysis included an estimate of the utility in Russia and Eastern Europe. Data from this benefit of a flexible dosing time with degludec. study also show that hypoglycaemia has a major This additional utility was not captured in the impact on patients and their behaviour [52]. In clinical studies and therefore a utility gain of Eastern Europe 51.7% T1DM and 40.6% T2DM 0.006 for degludec related to flexible dosing was patients reduced their insulin dose in response assumed based on the study by Boye et al. [41]. to hypoglycaemia, which could compromise This can be seen as a conservative estimate as a glycaemic control, and high proportions large time trade-off study demonstrated that increased blood glucose monitoring (75.5% 1212 Diabetes Ther (2018) 9:1201–1216 T1DM and 71.9% T2DM). clinical practice because of non-adherence or Hypoglycaemia and the fear of hypogly- poor clinic attendance. However, the sensitivity caemia act as barriers to timely initiation and analyses indicate that the conclusions are intensification of insulin and remain the major robust. Furthermore, real-world studies confirm limiting factors for achieving target levels of the observations from the clinical trials and glucose control in insulin-treated patients with support the effectiveness of degludec in clinical diabetes [53, 54]. Poor glycaemic control is practice. The EUropean TREsiba AudiT associated with serious long-term complica- (EU_TREAT) is a multinational observational tions, including cardiovascular disease, renal study in the general diabetes population in disease, retinopathy, amputation, depression routine care to assess the effectiveness of and neuropathy [55, 56]. degludec [23]. A total of 2550 patients were Due to its distinct pharmacological profile included in the study (T1DM = 1717, T2DM = degludec is associated with lower rates of 833), and results demonstrated that switching hypoglycaemia. In phase 3a clinical trials to degludec from other basal insulins signifi- degludec showed equivalent reductions in cantly improved glycaemic control and reduced HbA with a significantly lower rate of hypo- the risk of overall, nocturnal and severe hypo- 1c glycaemia versus glargine U100 [20, 21]. glycaemia [23]. Degludec was associated with a 17% lower rate The real-world study used for baseline rates of nocturnal non-severe hypoglycaemia in of hypoglycaemia [33] was based upon patient T1DM [21], an 86% lower rate of severe hypo- recall of hypoglycaemic events and the inter- glycaemia and a 36% lower rate of nocturnal pretation of symptoms may be open to bias. The non-severe hypoglycaemia in T2DM study was designed to maximise the optimum BOT [20, 21], as well as a 25% lower rate of nocturnal recall period; however, symptoms of hypogly- non-severe hypoglycaemia and 17% lower rate caemia can be incorrectly interpreted. The of daytime non-severe hypoglycaemia in uncertainty around hypoglycaemia rates was T2DM [21], compared with glargine U100. explored in a number of sensitivity analyses, B/B Two phase 3b studies, SWITCH 1 and SWITCH and whilst hypoglycaemia rates are a key driver 2, investigated the efficacy and safety of deglu- of cost-effectiveness outcomes the conclusion dec in patients with T1DM and T2DM with an that degludec is cost-effective is robust to increased risk of hypoglycaemia [57, 58]. In applying alternative values. these studies equivalent reductions in HbA An advantage of the modelling approach 1c were achieved with degludec with a lower total used for this economic analysis is its simplicity. daily insulin dose at end of trial and with a This short-term approach focuses on the impact lower risk of hypoglycaemic episodes versus of important aspects of insulin therapy such as glargine U100 [57, 58]. The hypoglycaemia hypoglycaemia and dosing and enables eco- benefits of degludec have also been reported in nomic analysis of new insulins that have been real-world clinical practice, with reductions of evaluated in treat-to-target trials. This model up to 90% observed in patients switching to has been previously used to evaluate the cost- degludec because of problems with hypogly- effectiveness of degludec versus glargine U100 caemia on glargine U100 or insulin detemir in patients with T1DM and T2DM [44, 59–61]. [22]. An analysis from the UK perspective found that A limitation of the current economic analy- degludec was dominant to glargine U100 in the sis, common to modelling studies, is the gen- T1DM and T2DM populations and highly BOT eralisability of clinical trial data to routine cost-effective in the T2DM population [26]. B/B clinical practice. The clinical trials included in Similarly, in a Swedish setting degludec was the meta-analyses used a treat-to-target cost-effective versus glargine U100 in all three approach, and insulin was titrated to a prede- patient groups [44]. In both cases, results are termined glycaemic target. In contrast, optimal consistent with those observed in this study. glycaemic control may not be achieved in Diabetes Ther (2018) 9:1201–1216 1213 Data Availability. All data generated or CONCLUSION analysed during this study are included in this published article/as supplementary information This short-term economic evaluation indicates files. that degludec is a cost-effective alternative to glargine U100 for patients with T1DM and Open Access. This article is distributed T2DM in Serbia. Degludec may particularly under the terms of the Creative Commons benefit those suffering from hypoglycaemia or Attribution-NonCommercial 4.0 International where the patient would benefit from the License (http://creativecommons.org/licenses/ additional flexibility. by-nc/4.0/), which permits any noncommercial use, distribution, and reproduction in any medium, provided you give appropriate credit ACKNOWLEDGEMENTS to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Funding. The study and article processing charges were funded by Novo Nordisk. All authors had full access to all of the data in this study and take complete responsibility for the REFERENCES integrity of the data and accuracy of the data analysis. 1. International Diabetes Federation (IDF). Diabetes atlas, 7th edition, 2015. Available at: http://www. Medical Writing and Editorial Assis- idf.org/idf-diabetes-atlas-seventh-edition. tance. The authors acknowledge writing and 2. Institute of Public Health of Serbia. Incidence and editorial support from Carrie Fidler of DRG mortality from diabetes in Serbia 2013. Institute of Abacus (sponsored by Novo Nordisk). Public Health of Serbia ‘‘Dr. Milan Jovanovic´ Batut’’, Belgrade: 2014. http://www.batut.org.rs/. Authorship. All named authors meet the 3. American Diabetes Association. Economic costs of International Committee of Medical Journal diabetes in the US in 2012. Diabetes Care. Editors (ICMJE) criteria for authorship for this 2013;36:1033–46. manuscript, take responsibility for the integrity of the work as a whole and have given final 4. Jonsson B. Revealing the cost of type II diabetes in Europe. Diabetologia. 2002;45:S5–12. approval for the version to be published. 5. Biorac N, Jakovljevic M, Stefanovic D, et al. Assess- Authorship Contributions. All authors con- ment of diabetes mellitus type 2 treatment costs in tributed to study design, data collection and the Republic of Serbia. Vojnosanit Pregl. analysis, and interpretation of results, as well as 2009;66:271–6. to the writing and final approval of this 6. Holman RR, Paul SK, Bethel MA, et al. 10-year fol- manuscript. low-up of intensive glucose control in type 2 dia- betes. N Engl J Med. 2008;359:1577–89. ˇ ´ Disclosures. Nebojsa Lalic has nothing to 7. Nathan DM, Cleary PA, Backlund JY, et al. Intensive disclose. Monika Russel-Szymczyk is an diabetes treatment and cardiovascular disease in employee of Novo Nordisk. Marina Culic is an patients with type 1 diabetes. N Engl J Med. employee of Novo Nordisk. Christian Klyver 2005;353:2643–53. Tikkanen is an employee of Novo Nordisk. 8. Ministry of Health of the Republic of Serbia. Barrie Chubb is an employee of Novo Nordisk. National clinical guidelines for management of diabetes mellitus. The Republic Expert Comittee for Compliance with Ethics Guidelines. This development and implementation of a good clini- article does not contain any studies with cal practice guide. human participants or animals performed by any of the authors. 1214 Diabetes Ther (2018) 9:1201–1216 9. de Pablos-Velasco P, Parhofer KG, Bradley C, et al. 19. Wangnoo SK, Chowdhury S, Rao PV. Treating to Current level of glycaemic control and its associ- target in type 2 diabetes: the BEGIN trial pro- ated factors in patients with type 2 diabetes across gramme. J Assoc Phys India. 2014;62:21–6. Europe: data from the PANORAMA study. Clin Endocrinol (Oxf). 2014;80:47–56. 20. Ratner R, Gough SC, Mathieu C, et al. Hypogly- caemia risk with insulin degludec compared with 10. Giugliano D, Maiorino MI, Bellastella G, et al. Effi- insulin glargine in type 2 and type 1 diabetes: a pre- cacy of insulin analogs in achieving the hemoglo- planned meta-analysis of phase 3 trials. Diabetes bin A1c target of \ 7% in type 2 diabetes: meta- Obes Metab. 2013;15:175–84. analysis of randomized controlled trials. Diabetes Care. 2011;34:510–7. 21. Vora J, Christensen T, Rana A, et al. Insulin deglu- dec versus insulin glargine in type 1 and type 2 11. Davies M, Gagliardino J, Gray L, et al. Real-world diabetes mellitus: a meta-analysis of endpoints in factors affecting adherence to insulin therapy in phase 3a trials. Diabetes Ther. 2014;5:435–46. patients with type 1 or type 2 diabetes mellitus: a systematic review. Diabet Med. 2013;30:512–24. 22. Evans M, McEwan P, Foos V. Insulin degludec early clinical experience: does the promise from the 12. Peyrot M, Rubin RR, Lauritzen T, et al. Resistance to clinical trials translate into clinical practice—a case- insulin therapy among patients and providers: based evaluation. J Med Econ. 2015;18:96–105. results of the cross-national diabetes attitudes, wishes, and needs (DAWN) study. Diabetes Care. 23. Siegmund T, Tentolouris N, Knudsen ST, et al. 2005;28:2673–9. A European, multicentre, retrospective, non-inter- ventional study (EU-TREAT) of the effectiveness of 13. Heise T, Nosek L, Bottcher SG, et al. Ultra-long- insulin degludec after switching basal insulin in a acting insulin degludec has a flat and stable glu- population with type 1 or type 2 diabetes. Diabetes cose-lowering effect in type 2 diabetes. Diabetes Obes Metab. 2017. https://doi.org/10.1111/dom. Obes Metab. 2012;14:944–50. 13149. 14. Haahr H, Heise T. A review of the pharmacological 24. Sassi F. Calculating QALYs, comparing QALY and properties of insulin degludec and their clinical DALY calculations. Health Policy Plan. relevance. Clin Pharmacokinet. 2014;53:787–800. 2006;21:402–8. 15. Heise T, Hermanski L, Nosek L, et al. Insulin 25. Macroeconomics and health: investing in health degludec: four times lower pharmacodynamic for economic development. Report of the Com- variability than insulin glargine under steady-state mission on Macroeconomics and Health. Geneva: conditions in type 1 diabetes. Diabetes Obes Metab. World Health Organization; 2001. Available from: 2012;14:859–64. http://apps.who.int/iris/bitstream/10665/42435/1/ 924154550X.pdf. 16. Meneghini L, Atkin SL, Gough SC, et al. The effi- cacy and safety of insulin degludec given in variable 26. Evans M, Chubb B, Gundgaard J. Cost-effectiveness once-daily dosing intervals compared with insulin of insulin degludec versus insulin glargine in adults glargine and insulin degludec dosed at the same with type 1 and type 2 diabetes mellitus. Diabetes time daily: a 26-week, randomized, open-label, Ther. 2017;8:275–91. parallel-group, treat-to-target trial in individuals with type 2 diabetes. Diabetes Care. 27. Food and Drug Administration. Guidance for 2013;36:858–64. Industry. Diabetes mellitus: developing drugs and therapeutic biologics for treatment and preven- 17. Mathieu C, Hollander P, Miranda-Palma B, et al. tion—Draft Guidance. February 2008. Available at: Efficacy and safety of insulin degludec in a flexible http://www.fda.gov/downloads/Drugs/Guidance dosing regimen vs insulin glargine in patients with ComplianceRegulatoryInformation/Guidances/ucm type 1 diabetes (BEGIN: flex T1): a 26-week ran- 071624.pdf. domized, treat-to-target trial with a 26-week extension. J Clin Endocrinol Metab. 28. Marseille E, Larson B, Kazi DS, et al. Thresholds for 2013;98:1154–62. the cost-effectiveness of interventions: alternative approaches. Bull World Health Organ. 18. European Medicines Agency. Tresiba (insulin 2015;93:118–24. degludec) Summary of Product Characteristics. Available at: http://www.ema.europa.eu/docs/en_ 29. Heller SR, Frier BM, Herslov ML, et al. Severe GB/document_library/EPAR_-_Product_Information/ hypoglycaemia in adults with insulin-treated dia- human/002498/WC500138940.pdf. betes: impact on healthcare resources. Diabet Med. 2016;33:471–7. Diabetes Ther (2018) 9:1201–1216 1215 30. Sanofi-Aventis. Lantus (insulin glargine): Structure 41. Boye KS, Matza LS, Walter KN, et al. Utilities and titration to target fasting glucose. Available at: http:// disutilities for attributes of injectable treatments for www.lantus.com/hcp/dosing-titration/titration- type 2 diabetes. Eur J Health Econ. 2011;12:219–30. guide.aspx. 42. Brod M, Christensen T, Thomsen TL, et al. The 31. Davies M, Storms F, Shutler S, et al. Improvement of impact of non-severe hypoglycemic events on work glycemic control in subjects with poorly controlled productivity and diabetes management. Value type 2 diabetes: comparison of two treatment Health. 2011;14:665–71. algorithms using insulin glargine. Diabetes Care. 2005;28:1282–8. 43. Evans M, Jensen HH, Bogelund M, et al. Flexible insulin dosing improves health-related quality-of- 32. Endocrinologic and Metabolic Drug Advisory life (HRQoL): a time trade-off survey. J Med Econ. Committee. Insulin degludec and insulin degludec/ 2013;16:1357–65. insulin aspart treatment to improve glycemic con- trol in patients with diabetes mellitus: NDAs 44. Ericsson A, Pollock RF, Hunt B, et al. Evaluation of 203314 and 203313 briefing document. 2012. the cost-utility of insulin degludec vs insulin glar- Available at http://www.fda.gov/downloads/Adviso gine in Sweden. J Med Econ. 2013;16:1442–52. ryCommittees/CommitteesMeetingMaterials/Drugs/ EndocrinologicandMetabolicDrugsAdvisoryCommit 45. Meneghini L, Miranda-Palma B. Insulin degludec: a tee/UCM327017.pdf. Accessed June 2014. novel ultra-long-acting basal insulin for use in type 1 and type 2 diabetes. Expert Rev Endocrinol Metab. 33. Ostenson CG, Geelhoed-Duijvestijn P, Lahtela J, 2012;7:9–14. et al. Self-reported non-severe hypoglycaemic events in Europe. Diabet Med. 2014;31:92–101. 46. Food and Drug Administration. Endocrinologic and Metabolic Drug Advisory Committee. Insulin 34. Donnelly LA, Morris AD, Frier BM, et al. Frequency degludec and insulin degludec/insulin aspart treat- and predictors of hypoglycaemia in type 1 and ment to improve glycemic control in patients with insulin-treated type 2 diabetes: a population-based diabetes mellitus: NDAs 203314 and 203313 study. Diabet Med. 2005;22:749–55. briefing document. Available at: http://www.fda. gov/downloads/AdvisoryCommittees/Committees 35. Henderson JN, Allen KV, Deary IJ, et al. Hypogly- MeetingMaterials/Drugs/EndocrinologicandMeta caemia in insulin-treated type 2 diabetes: fre- bolicDrugsAdvisoryCommittee/UCM327017.pdf quency, symptoms and impaired awareness. Diabet (last accessed 13 Feb 2014). 2012. Med. 2003;20:1016–21. 47. Philis-Tsimikas A, Brod M, Niemeyer M, et al. 36. Akram K, Pedersen-Bjergaard U, Borch-Johnsen K, Insulin degludec once-daily in type 2 diabetes: et al. Frequency and risk factors of severe hypo- simple or step-wise titration (BEGIN: once simple glycemia in insulin-treated type 2 diabetes: a use). Adv Ther. 2013;30:607–22. literature survey. J Diabetes Complicat. 2006;20:402–8. 48. Freemantle N, Meneghini L, Christensen T, et al. Insulin degludec improves health-related quality of 37. Gold AE, MacLeod KM, Frier BM. Frequency of life (SF-36 ) compared with insulin glargine in severe hypoglycemia in patients with type I dia- people with type 2 diabetes starting on basal insu- betes with impaired awareness of hypoglycemia. lin: a meta-analysis of phase 3a trials. Diabet Med. Diabetes Care. 1994;17:697–703. 2013;30:226–32. 38. Smith CB, Choudhary P, Pernet A, et al. Hypo- 49. Graveling AJ, Frier BM. Hypoglycaemia: an over- glycemia unawareness is associated with reduced view. Prim Care Diabetes. 2009;3:131–9. adherence to therapeutic decisions in patients with type 1 diabetes: evidence from a clinical audit. 50. Cryer P. Hypoglycemia in diabetes. In: Holt RIG, Diabetes Care. 2009;32:1196–8. Cockram CS, Flyvbjerg A, et al., editors. Textbook of diabetes. Hoboken: Blackwell Publishing; 2010. 39. Heller SR. Hypoglycaemia: its pathophysiology in insulin treated diabetes and hypoglycaemia 51. Khunti K, Alsifri S, Aronson R, et al. Rates and unawareness. Br J Diabetes Vasc Dis. 2011;11:6–9. predictors of hypoglycaemia in 27 585 people from 24 countries with insulin-treated type 1 and type 2 40. Evans M, Khunti K, Mamdani M, et al. Health-re- diabetes: the global HAT study. Diabetes Obes lated quality of life associated with daytime and Metab. 2016;18:907–15. nocturnal hypoglycaemic events: a time trade-off survey in five countries. Health Qual Life Out- 52. Khunti K, Alsifri S, Aronson R, et al. Impact of comes. 2013;11:90. hypoglycaemia on patient-reported outcomes from a global, 24-country study of 27,585 people with 1216 Diabetes Ther (2018) 9:1201–1216 type 1 and insulin-treated type 2 diabetes. Diabetes 58. Wysham C, Bhargava A, Chaykin L, et al. Effect of Res Clin Pract. 2017;130:121–9. insulin degludec vs insulin glargine U100 on hypoglycemia in patients with type 2 diabetes: the 53. Cryer PE. Hypoglycaemia: the limiting factor in the SWITCH 2 randomized clinical trial. JAMA. glycaemic management of type I and type II dia- 2017;318:45–56. betes. Diabetologia. 2002;45:937–48. 59. Evans M, Wolden M, Gundgaard J, et al. Cost-ef- 54. Meneghini L. Why and how to use insulin therapy fectiveness of insulin degludec compared with earlier in the management of type 2 diabetes. South insulin glargine for patients with type 2 diabetes Med J. 2007;100:164–74. treated with basal insulin—from the UK health care cost perspective. Diabetes Obes Metab. 55. Diabetes UK. Facts and stats 2016. Available at: 2014;16:366–75. https://www.diabetes.org.uk/Documents/Position% 20statements/DiabetesUK_Facts_Stats_Oct16.pdf. 60. Evans M, Wolden M, Gundgaard J, et al. Cost-ef- fectiveness of insulin degludec compared with 56. Inzucchi SE, Bergenstal RM, Buse JB, et al. Man- insulin glargine in a basal-bolus regimen in patients with type 1 diabetes mellitus in the UK. J Med Econ. agement of hyperglycemia in type 2 diabetes, 2015: 2015;18:56–68. a patient-centered approach: update to a position statement of the American diabetes association and 61. Pollock RF, Tikkanen CK. A short-term cost-utility the European association for the study of diabetes. analysis of insulin degludec versus insulin glargine Diabetes Care. 2015;38:140–9. U100 in patients with type 1 or type 2 diabetes in Denmark. J Med Econ. 2017;20:213–20. 57. Lane W, Bailey TS, Gerety G, et al. Effect of insulin degludec vs insulin glargine U100 on hypoglycemia in patients with type 1 diabetes: the SWITCH 1 randomized clinical trial. JAMA. 2017;318:33–44. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Diabetes Therapy Springer Journals

Cost-Effectiveness of Insulin Degludec Versus Insulin Glargine U100 in Patients with Type 1 and Type 2 Diabetes Mellitus in Serbia

Free
16 pages

Loading next page...
1
 
/lp/springer_journal/cost-effectiveness-of-insulin-degludec-versus-insulin-glargine-u100-in-Okd3fryNZV
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s)
Subject
Medicine & Public Health; Internal Medicine; Diabetes; Cardiology; Endocrinology
ISSN
1869-6953
eISSN
1869-6961
D.O.I.
10.1007/s13300-018-0426-0
Publisher site
See Article on Publisher Site

Abstract

Diabetes Ther (2018) 9:1201–1216 https://doi.org/10.1007/s13300-018-0426-0 ORIGINAL RESEARCH Cost-Effectiveness of Insulin Degludec Versus Insulin Glargine U100 in Patients with Type 1 and Type 2 Diabetes Mellitus in Serbia . . . Nebojsˇa Lalic´ Monika Russel-Szymczyk Marina Culic Christian Klyver Tikkanen Barrie Chubb Received: February 23, 2018 / Published online: April 26, 2018 The Author(s) 2018 effectiveness was analysed in a 1-year setting, ABSTRACT based on data from clinical trials. Costs were estimated from the healthcare payer perspec- Introduction: This study investigates the cost- tive, the Serbian Health Insurance Fund (RFZO). effectiveness of insulin degludec versus insulin The outcome measure was the incremental cost- glargine U100 in patients with type 1 and type 2 effectiveness ratio (ICER) or cost per quality- diabetes mellitus in Serbia. adjusted life-year (QALY) gained. Methods: A cost-utility analysis, implementing Results: Degludec is highly cost-effective com- a simple short-term model, was used to compare pared with glargine U100 for people with T1DM treatment costs and outcomes with degludec and T2DM in Serbia. The ICERs are estimated at versus glargine U100 in patients with type 1 417,586 RSD/QALY gained in T1DM, 558,811 (T1DM) and type 2 diabetes (T2DM). Cost- RSD/QALY gained in T2DM on basal oral ther- Enhanced digital features To view enhanced digital apy (T2DM ) and 1,200,141 RSD/QALY BOT features for this article go to https://doi.org/10.6084/ gained in T2DM on basal-bolus therapy m9.figshare.6106889. (T2DM ). All ICERs fall below the commonly B/B Electronic supplementary material The online accepted thresholds for cost-effectiveness in version of this article (https://doi.org/10.1007/s13300- Serbia (1,785,642 RSD/QALY gained). In all 018-0426-0) contains supplementary material, which is three patient groups, insulin costs are higher available to authorized users. with degludec than with glargine U100, but these costs are partially offset by savings from a N. Lalic´ Faculty of Medicine, Clinic for Endocrinology, lower daily insulin dose in T1DM and Diabetes and Metabolic Diseases, Clinical Center of T2DM , a reduction in hypoglycaemic events BOT Serbia, University of Belgrade, Belgrade, Serbia in all three patient groups and reduced costs of M. Russel-Szymczyk (&) SMBG testing in the T2DM groups with deglu- Novo Nordisk Pharma Sp. z o.o., Warsaw, Poland dec versus glargine U100. e-mail: zmns@novonordisk.com Conclusion: Degludec is a cost-effective alter- native to glargine U100 for patients with T1DM M. Culic Novo Nordisk Pharma d.o.o., Belgrade, Serbia and T2DM in Serbia. Degludec may particularly benefit those suffering from hypoglycaemia or C. K. Tikkanen where the patient would benefit from the Novo Nordisk Scandinavia AB, Copenhagen, Denmark option of flexible dosing. Funding: Novo Nordisk. B. Chubb Novo Nordisk Ltd, Gatwick, UK 1202 Diabetes Ther (2018) 9:1201–1216 Keywords: Cost-effectiveness; Insulin degludec; continuous subcutaneous insulin infusion Insulin glargine; Type 1 diabetes mellitus; Type (CSII) by means of an external portable insulin 2 diabetes mellitus; Serbia pump [8]. Type 2 diabetes (T2DM) is a progres- sive disease and treatment is intensified with disease progression. A significant proportion of INTRODUCTION patients with T2DM will eventually need insu- lin therapy to achieve optimal blood glucose Globally an estimated 415 million people have targets [8]. Insulin is the most effective method diabetes, and this is projected to increase to 642 of reducing blood glucose concentrations; million by 2040 [1]. According to the estimates however, despite clear guidance, glycaemic of the Serbian Institute of Public Health, dia- control remains suboptimal (HbA [ 7.0%) in a 1c betes affects 12.4% of the adult population in substantial number of patients [9, 10]. Key bar- Serbia [2]. Diabetes is a chronic metabolic dis- riers to insulin therapy include fear and risk of order characterised by high blood sugar levels hypoglycaemia, weight gain and restrictive (hyperglycaemia) and is a leading cause of treatment regimens [11, 12]. blindness, end-stage renal failure, heart attack/ Insulin degludec (degludec) is a basal insulin stroke, non-traumatic amputation and depres- analogue with a long duration of action and a sion [1]. Diabetes and related complications are distinct, slow absorption mechanism, which chief causes of death in most countries, and results in a flat and stable action profile [13, 14]. according to the International Diabetes Federa- It has four times less day-to-day variability in tion (IDF), there were an estimated 10,616 dia- glucose lowering effect than insulin glargine betes-related deaths in the 20–79 age group in U100 (glargine U100) [15]. Due to its long and Serbia in 2015 [1]. stable action profile, degludec allows for flexi- The treatment of diabetes has a substantial bility of dosing time. On occasions when economic impact on national healthcare sys- administration at the same time of day is not tems. The treatment of diabetes-related possible, patients can dose their insulin at any chronic complications accounts for the largest time of the day as long as there is a minimum of proportion of diabetes direct medical costs 8 h between doses [16–18]. [3, 4]. Indirect costs associated with diabetes, The efficacy and safety of degludec have such as absenteeism from work and reduced been evaluated in a large-scale clinical devel- productivity while at work, are also consider- opment programme. The degludec phase 3 able [3]. The most recent published cost data clinical trial programme (BEGIN) included more for Serbia are from 2007 [5] and estimate that than 9000 patients with T1DM and T2DM [19]. annual diabetes expenditures were 53,412.96 Meta-analyses of the phase 3a clinical trials RSD (or 444.42 EUR ) per patient. Indirect costs showed that degludec achieved equivalent represented 10.39% of total costs, while anti- reductions in HbA with a lower risk of hypo- 1c diabetes drugs and medical devices represented glycaemia versus glargine U100 and at a signif- 44% (38.10% and 5.94%, respectively) and icantly lower total daily insulin dose compared hospital and ambulatory costs the remaining with glargine U100 in T1DM (12% lower) and 45%. T2DM basal oral therapy (10% lower) [20, 21]. The aim of diabetes therapy is to keep blood Real-world studies confirm the observations glucose levels within recommended targets and from the clinical trials and support the effec- ultimately limit the development of diabetes- tiveness of degludec in clinical practice [22, 23]. related complications [6, 7]. All patients with Due to increasing constraints on healthcare type 1 diabetes (T1DM) require insulin. Current budgets, it is important that new therapies clinical guidance in Serbia recommends a basal- represent good value for the money. Decision- bolus regimen or for selected individuals making based on both clinical and economic evidence helps healthcare providers optimise 1 EUR = 120.185 RSD: Rulebook on the conditions, resource use and care for patients with diabetes. criteria, method and procedure for placing the medicinal Cost-effectiveness models assess the value of products on the reimbursement list. Diabetes Ther (2018) 9:1201–1216 1203 interventions by comparing the relative costs considered where a basal insulin analogue is and outcomes. An economic analysis typically indicated. estimates the difference in cost between one The model was a simple, transparent, short- healthcare intervention and an alternative, term (1-year time horizon) model developed in divided by the difference in health effects, Microsoft Excel 2010 (Microsoft Corp., Red- which is termed an incremental cost-effective- mond, WA, US) and has been previously pub- ness ratio (ICER). A commonly used effective- lished [26]. The short-term model focuses on ness measure in cost-effectiveness analyses is the impact of important aspects of insulin the quality-adjusted life year (QALY), which therapy, such as hypoglycaemia and dosing, measures health as a combination of both and accommodates the treat-to-target trials duration and health-related quality of life [24]. required by the Food and Drug Administration The incremental cost per additional QALY (FDA) [27]. In treat-to-target trials patients are gained (cost/QALY) allows decision makers to treated to the same glycaemic target and no compare across different disease areas to allo- differences in HbA are expected; thus, there is 1c cate healthcare resources for maximal economic no rationale for long-term modelling based on and clinical benefits. A financial threshold is HbA differences. 1c often set at which cost-effectiveness is accepted. Although cost-effectiveness was analysed in In Serbia, there is no officially indicated a 1-year setting and is based on data from 1-year threshold; however, the level of 3 9 GDP per clinical trials, the model can be replicated for capita suggested by the World Health Organi- subsequent years, and the outcomes represent zation (WHO) [25] is accepted by the Serbian the average annual cost-effectiveness in steady authorities. state. As the time horizon was 1-year, no dis- The objective of this study was to evaluate the counting was applied. cost-effectiveness of degludec versus glargine The model calculated the costs associated U100 from the healthcare payer perspective in with treatment (insulin, needles and SMBG test Serbia [Serbian Health Insurance Fund (RFZO )]. strips) and hypoglycaemic events (the cost of treating the event and the use of additional SMBG test strips associated with the event). METHODS Health-related quality of life (in the form of QALYs) was calculated by applying a disutility Model Overview per hypoglycaemic event incurred (Fig. 1). The type of cost-effectiveness analysis used was Perspective/Viewpoint a cost-utility analysis. Degludec was compared with glargine U100 in three patient All costs were estimated from the healthcare populations: payer perspective, RFZO. The threshold was • T1DM using basal-bolus therapy (T1DM) setat3 9 GDP per capita, as although there • T2DM using basal oral therapy (T2DM ) BOT are alternative approaches to thresholds for • T2DM using basal-bolus therapy (T2DM ) B/B cost-effectiveness of interventions [28], this is Glargine U100 was identified as the most an internationally recognised level [25]and is appropriate comparator for the cost-utility accepted by the Serbian authorities. According analysis. Treatment guidelines in Serbia recom- to the World Bank statistics for 2016, the GDP mend NPH insulin as the first-line insulin per capita in Serbia was 5348.30 USD. Using treatment for T1DM and T2DM, with a basal the exchange rate of 1 USD = 111.29 RSD, the insulin analogue considered if after 6 months threshold was set at the level of 1,785,642 on NPH there is persisting hypoglycaemia RSD. (\ 3.5 mmol/l). Thus, degludec should be Republicˇki fond za zdravstveno osiguranje. 1204 Diabetes Ther (2018) 9:1201–1216 Fig. 1 Overview of cost-effectiveness model. HCP health- ratio, IDeg insulin degludec, IGlar U100 insulin glargine, care professional, HRQoL health-related quality of life, QALY quality-adjusted life year, SMBG self-monitored hypo hypoglycaemia, ICER incremental cost-effectiveness blood glucose Data Used in the Model severe events data were obtained from the seri- ous adverse events case reports [29]. The use of additional SMBG tests in the week Cost Data following a non-severe event was also based on All costs were estimated from the healthcare the hypoglycaemia safety questionnaire. Data payer perspective, RFZO. were not collected on the testing pattern fol- lowing a severe event; therefore, it was conser- Direct Treatment Costs vatively assumed to be similar to a non-severe Costs of insulin, needles and SMGB tests were hypoglycaemic event. based on official RFZO prices elicited in For patients experiencing severe events it November 2017. Other costs of treatment (e.g. was assumed that all patients who were hospi- use of concomitant medication) or other costs talised used the ambulance service and that all resulting from treatment (e.g. long-term out- patients experiencing severe events used gluca- comes) were assumed to be equivalent in both gon to recover. treatment groups and were therefore not Patients suffering a hypoglycaemic event included. received the same treatment regardless of whe- ther they were on degludec or glargine U100. Therefore, any difference in costs of hypogly- Cost of Hypoglycaemic Events caemia between treatments was due to differ- Resource use associated with non-severe and ences in rates of hypoglycaemia and not the severe hypoglycaemic events was derived from cost per event. the clinical trial data. For non-severe events The average costs of hypoglycaemic events data were obtained from the patient-completed are calculated by multiplying the unit cost of hypoglycaemia safety questionnaire, and for the services by the share of patients using that Diabetes Ther (2018) 9:1201–1216 1205 Table 1 Total cost of an average severe/non-severe hypoglycaemic event Unit cost (RSD) Utilisation per hypoglycaemic event T1DM T2DM T2DM B/B BOT B/B Severe Non-severe Severe Non-severe Severe Non-severe Glucagon 1695.70 1.0 0.0 1.0 0.0 1.0 0.0 Ambulance 3430.00 0.14 0.0 0.88 0.0 0.28 0.0 bc Hospital 3980.00 0.14 0.0 0.88 0.0 0.28 0.0 SMBG 46.00 1.46 1.46 1.91 1.91 1.98 1.98 TOTAL (RSD) 3357.46 67.16 11,806.76 87.86 4975.98 91.08 TOTAL RSD values are given in bold B/B basal-bolus, BOT basal oral therapy, SMBG self-measured blood glucose, T1DM type 1 diabetes mellitus, T2DM type 2 diabetes mellitus, RSD Serbian dinar RFZO listed price (November 2017) HNF Rulebook on cost of services on the secondary/tertiary level indicators (July 2016) Hospitalisation cost includes 2 days of hospital stay = 7960.00 RSD (http://www.vma.mod.gov.rs/cenovnik-vma.pdf) treatment/service (Table 1). It was assumed that the end of trial doses captured from the clinical a non-severe event would be self-managed and trial data. The meta-analysis of insulin dose not require any additional resource use; thus, from the clinical trials was the source of the costs relate only to additional SMBG testing. glargine U100 dose and degludec/glargine U100 dose ratio [21]. The degludec dose was calcu- lated using the dose ratio to allow for adjust- Clinical Data ment of covariate factors such as trial, treatment, antidiabetic therapy at screening, Insulin Dose age, sex, region and baseline dose (Table 2). The daily insulin dose for the degludec and glargine U100 treatment groups was based on Table 2 Basal and bolus insulin use Treatment group Glargine U100 (units/day) Dose ratio (degludec/glargine U100) Degludec (units/day) T1DM , total dose 0.88* B/B Basal insulin 33.10 0.87* 28.80 Bolus insulin 35.00 0.88* 30.80 T2DM , total dose 0.90* BOT Basal insulin 51.70 0.90* 46.53 Bolus insulin Not relevant Not relevant Not relevant T2DM , total dose NS B/B Basal insulin 66.60 1.08* 71.93 Bolus insulin 72.70 NS 72.70 B/B basal-bolus, BOT basal oral therapy, T1DM type 1 diabetes mellitus, T2DM type 2 diabetes mellitus *p \ 0.05; NS non-significant; in the case of non-significant results, a relative rate of one was used in the calculation 1206 Diabetes Ther (2018) 9:1201–1216 Table 3 Calculation of hypoglycaemic event rates T1DM T2DM T2DM B/B BOT B/B Non-severe Severe Non-severe Severe Non-severe Severe Baseline hypoglycaemia 91.0 0.7 20.3 0.10 35.4 0.2 rate for glargine U100 Daytime/nocturnal split Daytime Nocturnal – Daytime Nocturnal – Daytime Nocturnal – 78% 22% 68% 32% 78% 22% Total events per patient 70.98 20.02 0.7 13.80 6.50 0.10 27.61 7.79 0.20 per year for glargine U100 Degludec/glargine U100 NS 0.83* NS NS 0.64* 0.14* 0.83* 0.75* NS hypoglycaemic event rate ratio Calculated degludec 70.98 16.62 0.7 13.80 4.10 0.01 22.92 5.84 0.20 hypoglycaemic event rate B/B basal-bolus, BOT basal oral therapy, T1DM type 1 diabetes mellitus, T2DM type 2 diabetes mellitus *p \ 0.05; NS, non-significant; in the case of non-significant results, a relative rate of one was used in the calculation Taken from Ostenson et al. [33] Proportion of daytime nocturnal events for glargine U100 taken from Ostenson et al. [33] SMBG Testing Hypoglycaemia Event Rates It was assumed that one SMBG test was con- Real-world hypoglycaemic event rates from a ducted with every main meal for regimens large-scale questionnaire-based study con- including bolus insulin (T1DM , T2DM ). ducted in seven European countries [33] were B/B B/B Three meals per day were assumed. used as the baseline values for severe and non- For SMBG tests related to basal insulin severe hypoglycaemic events. These rates pro- injections, the titration schedule recommended vide a better estimation of real-life event rates for use with glargine U100 [30] was used to than those from clinical trials [34], which can estimate SMBG utilisation for glargine U100. be biased in both the selection of patients and The algorithm recommends seven SMBG tests the treatment setting. For example, the clinical per week (the dose is adjusted every 3rd day trials excluded patients with a history of severe based on the mean of the FPG results over three hypoglycaemic episodes and anyone considered consecutive days) [31]. hypoglycaemia unaware [34–39]. The real-world Degludec enables patients to titrate, predict rates were used as the base case event rates in and monitor their blood glucose more effi- the glargine U100 group (as this is the current ciently [13, 14]. The recommended titration treatment on the market); see Table 3. Event algorithm for degludec is once-weekly adjust- rates for the degludec group (Table 3) were cal- ment of dose based on the average of two culated using the relative event rates taken from SMBG measurements from the 2 preceding the meta-analysis of hypoglycaemia [20, 21], days for patients with T2DM [32]. Thus, the which were adjusted for trial, type of diabetes, utilisation of SMBG tests related to degludec treatment, anti-diabetic therapy at screening, was assumed to be two tests per week for sex and region as fixed factors and age as a patients with T2DM. continuous covariate. Only rate ratios with a Diabetes Ther (2018) 9:1201–1216 1207 documented statistically significant difference glargine U100 group, with the majority repre- between the treatment arms were used. senting pharmacy costs. The total cost per patient per year is 11,990.32 RSD higher in the degludec group than in the glargine U100 group Compliance with Ethics Guidelines (6.9% higher costs), largely because of the This article does not contain any studies with increase in insulin costs. The costs of hypogly- human participants or animals performed by caemic events are similar for degludec and any of the authors. glargine U100 since only the non-severe noc- turnal events showed a statistically significant Utility Data difference between the two treatments, and the severe and daytime events were unchanged. QALYs were calculated by subtracting a disutil- In T2DM , the total cost per patient per BOT ity [40] per hypoglycaemic event experienced year in the degludec group (101,270.97 RSD) is from the baseline health utility. Disutilities 15,042.55 RSD (17.4%) higher than in the associated with hypoglycaemic events were glargine U100 group (86,228.42 RSD per patient obtained from a large-scale TTO study [40], per year). The difference in cost is due to the which reported a disutility of 0.0565 for a severe increased cost for insulin and the lower costs of event (with no significant difference between SMBG testing and non-severe nocturnal and daytime and nocturnal severe events) and severe hypoglycaemic events in the degludec disutilities of 0.0041 and 0.0067 for non-severe group. Lower costs of hypoglycaemia are driven daytime and non-severe nocturnal events, by the significant reduction in the number of respectively (a significant difference in utility severe hypoglycaemic events with degludec vs. was demonstrated for nocturnal vs. daytime glargine U100 in this patient group. non-severe events) [40]. The disutility per In T2DM , the total cost per patient per B/B hypoglycaemic event was multiplied by the year in the degludec group (285,584.00 RSD) is number of events observed in each treatment 45,912.88 (19.2%) higher than in the glargine group. U100 group (239,671.12 RSD per patient per In addition, the analysis included an esti- year), caused primarily by a higher spend for mate of the utility benefit for the option of insulin partly offset against a lower spend on flexible dosing time with degludec. Boye et al. SMBG and lower costs of non-severe daytime [41] reported a utility benefit of 0.006 associated and nocturnal hypoglycaemia. The higher with dosing flexibility; therefore, for degludec incremental cost for this group is driven mainly an extra utility gain of ? 0.006 was applied to by the slightly higher dose of basal insulin the QALY benefit. required in the degludec arm of the clinical trial. In the T2DM clinical trial [14], high B/B Sensitivity Analyses insulin doses were observed in both the deglu- dec and glargine U100 treatment arms. The One-way and probabilistic sensitivity analyses high doses observed in the trial are not expected were conducted to assess the impact of varying to be representative of a real-world setting for key assumptions and outcomes used in the base patients initiating a basal-bolus regimen as the case analysis (Table 4). trial mainly recruited patients who were already uncontrolled on an intensive basal-bolus regi- men or uncontrolled on a pre-mixed insulin RESULTS regimen. In a recent real-world study of deglu- dec, the daily basal and bolus insulin doses in Costs patients with T2DM (75% on a basal-bolus reg- imen) were considerably lower than those in In T1DM, total costs are estimated at 185,627.84 the clinical trial [23]. RSD per patient per year in the degludec group Average total costs for the three treatment and 173,637.52 per patient per year in the groups are presented in Table 5. 1208 Diabetes Ther (2018) 9:1201–1216 Table 4 Description of sensitivity analyses conducted Time horizon In the base case analysis the time horizon was 1 year. Sensitivity analysis increased the time horizon to 5 years Hypoglycaemia rates The base case rate of hypoglycaemia was taken from the published literature as it is believed that this provides a more realistic event rate. Additional published event rates, and the actual reported rates from the clinical trial programme, were investigated in sensitivity analyses SMBG costs SMBG costs were varied ± 20% Hypoglycaemia costs A sensitivity analysis was conducted where the cost of a non-severe hypoglycaemic event included the cost of one visit to the GP following the event (192.11 RSD) For severe hypoglycaemic events costs were varied ± 20% Dosing A sensitivity analysis was conducted with an assumption of equal doses for degludec and glargine U100 Injection frequency For some patients, current basal insulins need to be taken twice daily to ensure optimal control. The effect of using twice as many needles for the basal injections in the glargine U100 group was explored SMBG tests per week For degludec, the long duration of action and stable action profile [13, 14] mean that patients are able to titrate, predict and monitor their blood glucose more efficiently. The recommended titration algorithm for degludec is once-weekly adjustment of dose based on the average of two SMBG measurements from the 2 preceding days for patients with T2DM (98). The number of SMBG tests for basal titration with degludec was varied between 2 and 7 (T2DM ) and between 23 and 28 BOT (T2DM ) B/B Additional SMBG tests after In the base case analysis the number of additional SMBG tests following a non-severe non-severe events event was taken from the clinical trial data. A sensitivity analysis was conducted, which used a published source that reported 6.2 additional tests following a non- severe event [42] and another analysis that assumed no additional tests Flexible dosing utility A number of sensitivity analyses were conducted around this assumption: one where only 50% of patients were assumed to receive the benefit of flexible dosing, one that used an alternative published utility value (0.013 [43]) and one that assumed no utility from flexible dosing Probabilistic sensitivity analysis PSA used the standard errors and appropriate distributions of the parameters. The (PSA) distributions were assumed to be either normal or lognormal and each individual parameter was selected independently. The probabilistic sensitivity analyses were run with 1000 iterations. In the primary analysis for each of the groups—the standard error was only applied to differences that were statistically significant (i.e. if there was no statistical significance proven—then the rate ratio was set to 1 (assumed equivalent) and the SE was set to 0 (so as not to introduce random uncertainty) Diabetes Ther (2018) 9:1201–1216 1209 Table 5 Results: average total costs (RSD) per patient per year and incremental cost-effectiveness T1DM T2DM T2DM BOT B/B Cost, degludec (RSD) 185,627.84 101,270.97 285,584.00 Cost, glargine U100 (RSD) 173,637.52 86,228.42 239,671.12 D costs 11,990.32 15,042.55 45,912.88 D QALYs 0.0287 0.0269 0.0383 ICER (RSD/QALY) 417,586.27 558,811.00 1,200,140.55 Incremental Cost-Effectiveness Key parameters that influenced the ICER values were hypoglycaemia rates (T1DM, T2DM , T2DM ), insulin doses (T1DM, BOT B/B Degludec is a highly cost-effective treatment T2DM ) and the number of SMBG tests used option versus glargine U100 in patients with BOT weekly by patients (T2DM , T2DM ). BOT B/B T1DM as part of a basal-bolus insulin regimen For T1DM, the ICER remains below the cost- with an estimated ICER of 417,586.27 RSD/ effectiveness threshold in all scenarios tested in QALY gained. The result is primarily driven by one-way sensitivity analyses. ICERs range the significantly lower dose required, QALY between 297,452 RSD/QALY gained and gains due to significantly fewer non-severe 1,179,023 RSD/QALY gained. Rates of non-sev- nocturnal hypoglycaemic events versus glargine ere hypoglycaemia are influential on the ICER U100 and the opportunity to use flexible dosing in this group. When the annual number of non- when needed. This result falls well beneath the severe daytime events is reduced to 36.16 using cost-effectiveness threshold in Serbia of data from Donnelly et al. [34], the ICER 1,785,642 RSD/QALY gained. increases to 997,799 RSD/QALY gained. Con- In T2DM degludec is also highly cost-ef- BOT versely, when the annual number of non-severe fective (558,811.00 RSD/QALY gained), driven nocturnal events is increased to 20.80 using by cost savings from a significantly lower dose, a data from Ericsson et al. [44], the ICER is significant reduction in severe hypoglycaemia reduced to 404,428 RSD/QALY gained. The and the potential for once-weekly SMBG test- insulin dose ratio also has an impact on the ing. Due to the significantly fewer nocturnal ICER, but even under an extreme assumption of and severe hypoglycaemic events in this group equal dosing of basal insulin, degludec remains and the opportunity to use a flexible dosing highly cost-effective (720,507 RSD/QALY time when needed, degludec is associated with a gained). If the utility gain from flexibility is QALY gain versus glargine U100. excluded, the ICER is still within commonly In the T2DM setting, degludec is cost-ef- B/B accepted thresholds (527,896 RSD/QALY fective with an ICER of 1,200,140.55 RSD/QALY gained). gained, which falls below commonly accepted Similarly for T2DM , the favourable cost- BOT cost-effectiveness threshold. Degludec is asso- effectiveness results are invariant to changes in ciated with significantly fewer non-severe day- most of the input parameters. ICERs range time and nocturnal events, which results in a between 294,932 and 1,955,556 RSD/QALY QALY gain versus glargine U100. gained. The number of SMBG tests per week is influential on the ICER in this setting, but Sensitivity Analysis degludec is still cost-effective (1,004,600 RSD/ QALY gained) even under the strict assumption Extensive sensitivity analyses demonstrate that of no reductions in SMBG testing in the deglu- the results are robust and largely insensitive to dec group. Insulin dose ratio and utility gain changes in input parameters (results presented from flexibility also have an impact on the in supplementary table). results, but ICERs fall below the commonly 1210 Diabetes Ther (2018) 9:1201–1216 accepted threshold even with extreme assump- available, economic as well as clinical evidence tions (919,928 and 719,091 RSD/QALY gained is essential for healthcare providers to maximise respectively). Varying the rates of hypogly- healthcare with constrained budgets. caemia has an impact on the ICER in this group This simple, short-term economic evaluation reflecting the large benefits in terms of reduced demonstrates that degludec is likely to be con- hypoglycaemia with degludec versus glargine sidered highly cost-effective compared with U100. Drawing on higher event rates for non- glargine U100 for people with T1DM and T2DM severe events generates highly cost-effective in Serbia. All ICERs fall below the commonly results for degludec versus glargine U100; con- accepted thresholds for cost-effectiveness in versely, using lower event rates for severe Serbia. hypoglycaemic events, e.g. from Ratner [20], increases the ICER to 1,955,556.24 RSD/QALY gained, which is just above the commonly accepted threshold. For T2DM the parameter that has the most B/B influence on the ICER is the rate of hypogly- caemia. Varying the rate of non-severe hypo- glycaemia has an impact on the ICER in this group due to the significant reduction of non- severe events with degludec versus glargine U100. ICERs range between 980,305 and 3,636,384 RSD/QALY gained, with low rates of non-severe hypoglycaemia generating the higher ICERs. In all other scenarios analysed the ICERs are below the commonly accepted threshold (range 938,549–1,513,491 RSD/QALY gained), indicating that degludec is a cost-ef- fective therapeutic option in this group. Probabilistic Sensitivity Analysis To capture the uncertainty of the results caused by statistical uncertainty with respect to the parameter inputs, probabilistic sensitivity anal- ysis has been conducted. The probabilistic sen- sitivity analyses have each been run with 1000 iterations. The cost-effectiveness acceptability curves show that at a willingness-to-pay threshold of 2,048,112 RSD/QALY gained, the probability that degludec is cost-effective relative to glar- gine U100 is 77.5, 97.3 and 85.7% for T1DM, T2DM and T2DM , respectively (Fig. 2). BOT B/B DISCUSSION There are limited data on the cost-effectiveness of treatments for diabetes in Serbia. With an Fig. 2 Probabilistic sensitivity analysis—cost-effectiveness increasing number of insulin formulations acceptability curves Diabetes Ther (2018) 9:1201–1216 1211 In all three patient groups, insulin costs are flexible once-daily injection was associated with higher with degludec than with glargine U100, a 0.016 higher utility versus a fixed time of but these costs are partially offset by savings injection [43]. from a lower daily insulin dose in T1DM and Extensive sensitivity analyses show that the T2DM , a reduction in hypoglycaemic events results are robust and invariant to changes in BOT in all three patient groups and reduced costs of most of the input parameters. In all three SMBG testing in the T2DM groups, with deglu- patients groups the ICER remained below the dec versus glargine U100. For degludec, the threshold of 1,785,642 RSD in all scenarios tes- longer half-life and the corresponding long flat ted, except when low rates of non-severe insulin profile in steady state [15] and lower hypoglycaemia were applied in the T2DM B/B variability over the day compared with glargine setting and low rates of severe hypoglycaemia U100 [45] mean that patients are able to titrate, applied in the T2DM setting. PSA shows that BOT predict and monitor their blood glucose more there is a high probability that degludec will be efficiently. Consequently, degludec can be cost-effective versus glargine U100 in all three monitored with less frequent SMBG testing in patient groups. patients with T2DM. The titration algorithm for Hypoglycaemia can have a major impact on degludec is once-weekly adjustment of dose patients’ lives affecting employment, driving, based on the average of two SMBG measure- relationships, travel and leisure activities. Sev- ments from the 2 preceding days for patients ere episodes can result in serious clinical with T2DM [46]. The concept of fewer SMBG sequelae such as physical (e.g. fractures, head tests per week is supported by a phase 3b trial, injuries), neurological (e.g. convul- which examined different scenarios of SMBG sions/seizures, paralysis, coma) and cardiac (e.g. testing in patients with T2DM. The trial arrhythmia, cardiac failure) morbidity, as well demonstrated that degludec was effective and as death in some extreme cases [49, 50]. Fur- well tolerated using only one SMBG test per thermore, recurrent episodes, whether severe or week to titrate the insulin dose compared with non-severe, can lead to the development of titration based on three consecutive pre-break- impaired awareness of hypoglycaemia; this can fast SMBG values in patients with T2DM [47]. be problematic as it means that patients are less No statistically significant differences in HbA , able to recognise their symptoms and take 1c body weight, hypoglycaemia or any other safety remedial action. A recent non-interventional, parameters were observed between the two prospective-cohort survey conducted across titration algorithms [47]. 2004 sites in 24 countries, including Serbia, An important aspect to consider, especially investigated the prevalence of hypoglycaemia for chronic diseases such as diabetes mellitus, is [51]. The study included 27,585 patients with the quality of life of patients. With degludec insulin-treated diabetes and reported a higher higher gains in health-related quality of life frequency of hypoglycaemia than previously (HRQoL) were observed [48]. Incremental QALY reported, with marked variation across geo- gains with degludec versus glargine U100 in all graphic regions. Overall rates of hypoglycaemia three groups are associated with the reduction in Eastern Europe were 66.9 events/person/year in hypoglycaemic events and the opportunity for T1DM and 23.7 events/person/year for to use flexible dosing when needed. The current T2DM, with the highest rates in T2DM observed analysis included an estimate of the utility in Russia and Eastern Europe. Data from this benefit of a flexible dosing time with degludec. study also show that hypoglycaemia has a major This additional utility was not captured in the impact on patients and their behaviour [52]. In clinical studies and therefore a utility gain of Eastern Europe 51.7% T1DM and 40.6% T2DM 0.006 for degludec related to flexible dosing was patients reduced their insulin dose in response assumed based on the study by Boye et al. [41]. to hypoglycaemia, which could compromise This can be seen as a conservative estimate as a glycaemic control, and high proportions large time trade-off study demonstrated that increased blood glucose monitoring (75.5% 1212 Diabetes Ther (2018) 9:1201–1216 T1DM and 71.9% T2DM). clinical practice because of non-adherence or Hypoglycaemia and the fear of hypogly- poor clinic attendance. However, the sensitivity caemia act as barriers to timely initiation and analyses indicate that the conclusions are intensification of insulin and remain the major robust. Furthermore, real-world studies confirm limiting factors for achieving target levels of the observations from the clinical trials and glucose control in insulin-treated patients with support the effectiveness of degludec in clinical diabetes [53, 54]. Poor glycaemic control is practice. The EUropean TREsiba AudiT associated with serious long-term complica- (EU_TREAT) is a multinational observational tions, including cardiovascular disease, renal study in the general diabetes population in disease, retinopathy, amputation, depression routine care to assess the effectiveness of and neuropathy [55, 56]. degludec [23]. A total of 2550 patients were Due to its distinct pharmacological profile included in the study (T1DM = 1717, T2DM = degludec is associated with lower rates of 833), and results demonstrated that switching hypoglycaemia. In phase 3a clinical trials to degludec from other basal insulins signifi- degludec showed equivalent reductions in cantly improved glycaemic control and reduced HbA with a significantly lower rate of hypo- the risk of overall, nocturnal and severe hypo- 1c glycaemia versus glargine U100 [20, 21]. glycaemia [23]. Degludec was associated with a 17% lower rate The real-world study used for baseline rates of nocturnal non-severe hypoglycaemia in of hypoglycaemia [33] was based upon patient T1DM [21], an 86% lower rate of severe hypo- recall of hypoglycaemic events and the inter- glycaemia and a 36% lower rate of nocturnal pretation of symptoms may be open to bias. The non-severe hypoglycaemia in T2DM study was designed to maximise the optimum BOT [20, 21], as well as a 25% lower rate of nocturnal recall period; however, symptoms of hypogly- non-severe hypoglycaemia and 17% lower rate caemia can be incorrectly interpreted. The of daytime non-severe hypoglycaemia in uncertainty around hypoglycaemia rates was T2DM [21], compared with glargine U100. explored in a number of sensitivity analyses, B/B Two phase 3b studies, SWITCH 1 and SWITCH and whilst hypoglycaemia rates are a key driver 2, investigated the efficacy and safety of deglu- of cost-effectiveness outcomes the conclusion dec in patients with T1DM and T2DM with an that degludec is cost-effective is robust to increased risk of hypoglycaemia [57, 58]. In applying alternative values. these studies equivalent reductions in HbA An advantage of the modelling approach 1c were achieved with degludec with a lower total used for this economic analysis is its simplicity. daily insulin dose at end of trial and with a This short-term approach focuses on the impact lower risk of hypoglycaemic episodes versus of important aspects of insulin therapy such as glargine U100 [57, 58]. The hypoglycaemia hypoglycaemia and dosing and enables eco- benefits of degludec have also been reported in nomic analysis of new insulins that have been real-world clinical practice, with reductions of evaluated in treat-to-target trials. This model up to 90% observed in patients switching to has been previously used to evaluate the cost- degludec because of problems with hypogly- effectiveness of degludec versus glargine U100 caemia on glargine U100 or insulin detemir in patients with T1DM and T2DM [44, 59–61]. [22]. An analysis from the UK perspective found that A limitation of the current economic analy- degludec was dominant to glargine U100 in the sis, common to modelling studies, is the gen- T1DM and T2DM populations and highly BOT eralisability of clinical trial data to routine cost-effective in the T2DM population [26]. B/B clinical practice. The clinical trials included in Similarly, in a Swedish setting degludec was the meta-analyses used a treat-to-target cost-effective versus glargine U100 in all three approach, and insulin was titrated to a prede- patient groups [44]. In both cases, results are termined glycaemic target. In contrast, optimal consistent with those observed in this study. glycaemic control may not be achieved in Diabetes Ther (2018) 9:1201–1216 1213 Data Availability. All data generated or CONCLUSION analysed during this study are included in this published article/as supplementary information This short-term economic evaluation indicates files. that degludec is a cost-effective alternative to glargine U100 for patients with T1DM and Open Access. This article is distributed T2DM in Serbia. Degludec may particularly under the terms of the Creative Commons benefit those suffering from hypoglycaemia or Attribution-NonCommercial 4.0 International where the patient would benefit from the License (http://creativecommons.org/licenses/ additional flexibility. by-nc/4.0/), which permits any noncommercial use, distribution, and reproduction in any medium, provided you give appropriate credit ACKNOWLEDGEMENTS to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Funding. The study and article processing charges were funded by Novo Nordisk. All authors had full access to all of the data in this study and take complete responsibility for the REFERENCES integrity of the data and accuracy of the data analysis. 1. International Diabetes Federation (IDF). Diabetes atlas, 7th edition, 2015. Available at: http://www. Medical Writing and Editorial Assis- idf.org/idf-diabetes-atlas-seventh-edition. tance. The authors acknowledge writing and 2. Institute of Public Health of Serbia. Incidence and editorial support from Carrie Fidler of DRG mortality from diabetes in Serbia 2013. Institute of Abacus (sponsored by Novo Nordisk). Public Health of Serbia ‘‘Dr. Milan Jovanovic´ Batut’’, Belgrade: 2014. http://www.batut.org.rs/. Authorship. All named authors meet the 3. American Diabetes Association. Economic costs of International Committee of Medical Journal diabetes in the US in 2012. Diabetes Care. Editors (ICMJE) criteria for authorship for this 2013;36:1033–46. manuscript, take responsibility for the integrity of the work as a whole and have given final 4. Jonsson B. Revealing the cost of type II diabetes in Europe. Diabetologia. 2002;45:S5–12. approval for the version to be published. 5. Biorac N, Jakovljevic M, Stefanovic D, et al. Assess- Authorship Contributions. All authors con- ment of diabetes mellitus type 2 treatment costs in tributed to study design, data collection and the Republic of Serbia. Vojnosanit Pregl. analysis, and interpretation of results, as well as 2009;66:271–6. to the writing and final approval of this 6. Holman RR, Paul SK, Bethel MA, et al. 10-year fol- manuscript. low-up of intensive glucose control in type 2 dia- betes. N Engl J Med. 2008;359:1577–89. ˇ ´ Disclosures. Nebojsa Lalic has nothing to 7. Nathan DM, Cleary PA, Backlund JY, et al. Intensive disclose. Monika Russel-Szymczyk is an diabetes treatment and cardiovascular disease in employee of Novo Nordisk. Marina Culic is an patients with type 1 diabetes. N Engl J Med. employee of Novo Nordisk. Christian Klyver 2005;353:2643–53. Tikkanen is an employee of Novo Nordisk. 8. Ministry of Health of the Republic of Serbia. Barrie Chubb is an employee of Novo Nordisk. National clinical guidelines for management of diabetes mellitus. The Republic Expert Comittee for Compliance with Ethics Guidelines. This development and implementation of a good clini- article does not contain any studies with cal practice guide. human participants or animals performed by any of the authors. 1214 Diabetes Ther (2018) 9:1201–1216 9. de Pablos-Velasco P, Parhofer KG, Bradley C, et al. 19. Wangnoo SK, Chowdhury S, Rao PV. Treating to Current level of glycaemic control and its associ- target in type 2 diabetes: the BEGIN trial pro- ated factors in patients with type 2 diabetes across gramme. J Assoc Phys India. 2014;62:21–6. Europe: data from the PANORAMA study. Clin Endocrinol (Oxf). 2014;80:47–56. 20. Ratner R, Gough SC, Mathieu C, et al. Hypogly- caemia risk with insulin degludec compared with 10. Giugliano D, Maiorino MI, Bellastella G, et al. Effi- insulin glargine in type 2 and type 1 diabetes: a pre- cacy of insulin analogs in achieving the hemoglo- planned meta-analysis of phase 3 trials. Diabetes bin A1c target of \ 7% in type 2 diabetes: meta- Obes Metab. 2013;15:175–84. analysis of randomized controlled trials. Diabetes Care. 2011;34:510–7. 21. Vora J, Christensen T, Rana A, et al. Insulin deglu- dec versus insulin glargine in type 1 and type 2 11. Davies M, Gagliardino J, Gray L, et al. Real-world diabetes mellitus: a meta-analysis of endpoints in factors affecting adherence to insulin therapy in phase 3a trials. Diabetes Ther. 2014;5:435–46. patients with type 1 or type 2 diabetes mellitus: a systematic review. Diabet Med. 2013;30:512–24. 22. Evans M, McEwan P, Foos V. Insulin degludec early clinical experience: does the promise from the 12. Peyrot M, Rubin RR, Lauritzen T, et al. Resistance to clinical trials translate into clinical practice—a case- insulin therapy among patients and providers: based evaluation. J Med Econ. 2015;18:96–105. results of the cross-national diabetes attitudes, wishes, and needs (DAWN) study. Diabetes Care. 23. Siegmund T, Tentolouris N, Knudsen ST, et al. 2005;28:2673–9. A European, multicentre, retrospective, non-inter- ventional study (EU-TREAT) of the effectiveness of 13. Heise T, Nosek L, Bottcher SG, et al. Ultra-long- insulin degludec after switching basal insulin in a acting insulin degludec has a flat and stable glu- population with type 1 or type 2 diabetes. Diabetes cose-lowering effect in type 2 diabetes. Diabetes Obes Metab. 2017. https://doi.org/10.1111/dom. Obes Metab. 2012;14:944–50. 13149. 14. Haahr H, Heise T. A review of the pharmacological 24. Sassi F. Calculating QALYs, comparing QALY and properties of insulin degludec and their clinical DALY calculations. Health Policy Plan. relevance. Clin Pharmacokinet. 2014;53:787–800. 2006;21:402–8. 15. Heise T, Hermanski L, Nosek L, et al. Insulin 25. Macroeconomics and health: investing in health degludec: four times lower pharmacodynamic for economic development. Report of the Com- variability than insulin glargine under steady-state mission on Macroeconomics and Health. Geneva: conditions in type 1 diabetes. Diabetes Obes Metab. World Health Organization; 2001. Available from: 2012;14:859–64. http://apps.who.int/iris/bitstream/10665/42435/1/ 924154550X.pdf. 16. Meneghini L, Atkin SL, Gough SC, et al. The effi- cacy and safety of insulin degludec given in variable 26. Evans M, Chubb B, Gundgaard J. Cost-effectiveness once-daily dosing intervals compared with insulin of insulin degludec versus insulin glargine in adults glargine and insulin degludec dosed at the same with type 1 and type 2 diabetes mellitus. Diabetes time daily: a 26-week, randomized, open-label, Ther. 2017;8:275–91. parallel-group, treat-to-target trial in individuals with type 2 diabetes. Diabetes Care. 27. Food and Drug Administration. Guidance for 2013;36:858–64. Industry. Diabetes mellitus: developing drugs and therapeutic biologics for treatment and preven- 17. Mathieu C, Hollander P, Miranda-Palma B, et al. tion—Draft Guidance. February 2008. Available at: Efficacy and safety of insulin degludec in a flexible http://www.fda.gov/downloads/Drugs/Guidance dosing regimen vs insulin glargine in patients with ComplianceRegulatoryInformation/Guidances/ucm type 1 diabetes (BEGIN: flex T1): a 26-week ran- 071624.pdf. domized, treat-to-target trial with a 26-week extension. J Clin Endocrinol Metab. 28. Marseille E, Larson B, Kazi DS, et al. Thresholds for 2013;98:1154–62. the cost-effectiveness of interventions: alternative approaches. Bull World Health Organ. 18. European Medicines Agency. Tresiba (insulin 2015;93:118–24. degludec) Summary of Product Characteristics. Available at: http://www.ema.europa.eu/docs/en_ 29. Heller SR, Frier BM, Herslov ML, et al. Severe GB/document_library/EPAR_-_Product_Information/ hypoglycaemia in adults with insulin-treated dia- human/002498/WC500138940.pdf. betes: impact on healthcare resources. Diabet Med. 2016;33:471–7. Diabetes Ther (2018) 9:1201–1216 1215 30. Sanofi-Aventis. Lantus (insulin glargine): Structure 41. Boye KS, Matza LS, Walter KN, et al. Utilities and titration to target fasting glucose. Available at: http:// disutilities for attributes of injectable treatments for www.lantus.com/hcp/dosing-titration/titration- type 2 diabetes. Eur J Health Econ. 2011;12:219–30. guide.aspx. 42. Brod M, Christensen T, Thomsen TL, et al. The 31. Davies M, Storms F, Shutler S, et al. Improvement of impact of non-severe hypoglycemic events on work glycemic control in subjects with poorly controlled productivity and diabetes management. Value type 2 diabetes: comparison of two treatment Health. 2011;14:665–71. algorithms using insulin glargine. Diabetes Care. 2005;28:1282–8. 43. Evans M, Jensen HH, Bogelund M, et al. Flexible insulin dosing improves health-related quality-of- 32. Endocrinologic and Metabolic Drug Advisory life (HRQoL): a time trade-off survey. J Med Econ. Committee. Insulin degludec and insulin degludec/ 2013;16:1357–65. insulin aspart treatment to improve glycemic con- trol in patients with diabetes mellitus: NDAs 44. Ericsson A, Pollock RF, Hunt B, et al. Evaluation of 203314 and 203313 briefing document. 2012. the cost-utility of insulin degludec vs insulin glar- Available at http://www.fda.gov/downloads/Adviso gine in Sweden. J Med Econ. 2013;16:1442–52. ryCommittees/CommitteesMeetingMaterials/Drugs/ EndocrinologicandMetabolicDrugsAdvisoryCommit 45. Meneghini L, Miranda-Palma B. Insulin degludec: a tee/UCM327017.pdf. Accessed June 2014. novel ultra-long-acting basal insulin for use in type 1 and type 2 diabetes. Expert Rev Endocrinol Metab. 33. Ostenson CG, Geelhoed-Duijvestijn P, Lahtela J, 2012;7:9–14. et al. Self-reported non-severe hypoglycaemic events in Europe. Diabet Med. 2014;31:92–101. 46. Food and Drug Administration. Endocrinologic and Metabolic Drug Advisory Committee. Insulin 34. Donnelly LA, Morris AD, Frier BM, et al. Frequency degludec and insulin degludec/insulin aspart treat- and predictors of hypoglycaemia in type 1 and ment to improve glycemic control in patients with insulin-treated type 2 diabetes: a population-based diabetes mellitus: NDAs 203314 and 203313 study. Diabet Med. 2005;22:749–55. briefing document. Available at: http://www.fda. gov/downloads/AdvisoryCommittees/Committees 35. Henderson JN, Allen KV, Deary IJ, et al. Hypogly- MeetingMaterials/Drugs/EndocrinologicandMeta caemia in insulin-treated type 2 diabetes: fre- bolicDrugsAdvisoryCommittee/UCM327017.pdf quency, symptoms and impaired awareness. Diabet (last accessed 13 Feb 2014). 2012. Med. 2003;20:1016–21. 47. Philis-Tsimikas A, Brod M, Niemeyer M, et al. 36. Akram K, Pedersen-Bjergaard U, Borch-Johnsen K, Insulin degludec once-daily in type 2 diabetes: et al. Frequency and risk factors of severe hypo- simple or step-wise titration (BEGIN: once simple glycemia in insulin-treated type 2 diabetes: a use). Adv Ther. 2013;30:607–22. literature survey. J Diabetes Complicat. 2006;20:402–8. 48. Freemantle N, Meneghini L, Christensen T, et al. Insulin degludec improves health-related quality of 37. Gold AE, MacLeod KM, Frier BM. Frequency of life (SF-36 ) compared with insulin glargine in severe hypoglycemia in patients with type I dia- people with type 2 diabetes starting on basal insu- betes with impaired awareness of hypoglycemia. lin: a meta-analysis of phase 3a trials. Diabet Med. Diabetes Care. 1994;17:697–703. 2013;30:226–32. 38. Smith CB, Choudhary P, Pernet A, et al. Hypo- 49. Graveling AJ, Frier BM. Hypoglycaemia: an over- glycemia unawareness is associated with reduced view. Prim Care Diabetes. 2009;3:131–9. adherence to therapeutic decisions in patients with type 1 diabetes: evidence from a clinical audit. 50. Cryer P. Hypoglycemia in diabetes. In: Holt RIG, Diabetes Care. 2009;32:1196–8. Cockram CS, Flyvbjerg A, et al., editors. Textbook of diabetes. Hoboken: Blackwell Publishing; 2010. 39. Heller SR. Hypoglycaemia: its pathophysiology in insulin treated diabetes and hypoglycaemia 51. Khunti K, Alsifri S, Aronson R, et al. Rates and unawareness. Br J Diabetes Vasc Dis. 2011;11:6–9. predictors of hypoglycaemia in 27 585 people from 24 countries with insulin-treated type 1 and type 2 40. Evans M, Khunti K, Mamdani M, et al. Health-re- diabetes: the global HAT study. Diabetes Obes lated quality of life associated with daytime and Metab. 2016;18:907–15. nocturnal hypoglycaemic events: a time trade-off survey in five countries. Health Qual Life Out- 52. Khunti K, Alsifri S, Aronson R, et al. Impact of comes. 2013;11:90. hypoglycaemia on patient-reported outcomes from a global, 24-country study of 27,585 people with 1216 Diabetes Ther (2018) 9:1201–1216 type 1 and insulin-treated type 2 diabetes. Diabetes 58. Wysham C, Bhargava A, Chaykin L, et al. Effect of Res Clin Pract. 2017;130:121–9. insulin degludec vs insulin glargine U100 on hypoglycemia in patients with type 2 diabetes: the 53. Cryer PE. Hypoglycaemia: the limiting factor in the SWITCH 2 randomized clinical trial. JAMA. glycaemic management of type I and type II dia- 2017;318:45–56. betes. Diabetologia. 2002;45:937–48. 59. Evans M, Wolden M, Gundgaard J, et al. Cost-ef- 54. Meneghini L. Why and how to use insulin therapy fectiveness of insulin degludec compared with earlier in the management of type 2 diabetes. South insulin glargine for patients with type 2 diabetes Med J. 2007;100:164–74. treated with basal insulin—from the UK health care cost perspective. Diabetes Obes Metab. 55. Diabetes UK. Facts and stats 2016. Available at: 2014;16:366–75. https://www.diabetes.org.uk/Documents/Position% 20statements/DiabetesUK_Facts_Stats_Oct16.pdf. 60. Evans M, Wolden M, Gundgaard J, et al. Cost-ef- fectiveness of insulin degludec compared with 56. Inzucchi SE, Bergenstal RM, Buse JB, et al. Man- insulin glargine in a basal-bolus regimen in patients with type 1 diabetes mellitus in the UK. J Med Econ. agement of hyperglycemia in type 2 diabetes, 2015: 2015;18:56–68. a patient-centered approach: update to a position statement of the American diabetes association and 61. Pollock RF, Tikkanen CK. A short-term cost-utility the European association for the study of diabetes. analysis of insulin degludec versus insulin glargine Diabetes Care. 2015;38:140–9. U100 in patients with type 1 or type 2 diabetes in Denmark. J Med Econ. 2017;20:213–20. 57. Lane W, Bailey TS, Gerety G, et al. Effect of insulin degludec vs insulin glargine U100 on hypoglycemia in patients with type 1 diabetes: the SWITCH 1 randomized clinical trial. JAMA. 2017;318:33–44.

Journal

Diabetes TherapySpringer Journals

Published: Apr 26, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off