The Economic Burden of Insulin-Related Hypoglycemia in Adults with Diabetes: An Analysis from the Perspective of the Italian Healthcare System

The Economic Burden of Insulin-Related Hypoglycemia in Adults with Diabetes: An Analysis from the... Diabetes Ther (2018) 9:1037–1047 https://doi.org/10.1007/s13300-018-0418-0 ORIGINAL RESEARCH The Economic Burden of Insulin-Related Hypoglycemia in Adults with Diabetes: An Analysis from the Perspective of the Italian Healthcare System . . . . Witesh Parekh Sophie E. Streeton James Baker-Knight Roberta Montagnoli Paolo Nicoziani Giulio Marchesini Received: February 23, 2018 / Published online: March 29, 2018 The Author(s) 2018 rates (0.49 severe and 53.3 non-severe episodes ABSTRACT per year for T1DM, and 0.09 severe and 9.3 non- severe episodes per year for T2DM). Uncertainty Introduction: The aim of this analysis was to around model inputs was explored through estimate the cost of insulin-related hypo- sensitivity and scenario analyses. glycemia in adult patients with diabetes in Italy Results: The direct cost of insulin-related using the Local Impact of Hypoglycemia Tool hypoglycemia in Italy is estimated at €144.7 (LIHT), and to explore the effect of different million per year, with €65 million hypoglycemia rates on budget impact. attributable to severe episodes and €79.6 mil- Methods: Direct costs and healthcare resource lion due to non-severe episodes. The total cost utilization were estimated for severe and non- of hypoglycemia is approximately 1.7-fold severe hypoglycemic episodes in Italy and higher for T2DM (€91.7 million) than for T1DM applied to the population of adults with type 1 (€53 million). The cost of a hypoglycemic epi- diabetes (T1DM) and type 2 diabetes (T2DM) sode ranges from €4.59 for a non-severe event and their corresponding hypoglycemia episode where additional self-monitoring of blood glu- cose (SMBG) testing is the only cost incurred, to Enhanced content To view enhanced content for this €5790.59 for a severe event that also requires an article go to https://doi.org/10.6084/m9.figshare. ambulance, A&E, hospitalization, and a visit to a diabetes specialist. A reduction in hypo- Electronic supplementary material The online glycemia event rates could result in substantial version of this article (https://doi.org/10.1007/s13300- cost savings; for example, a 20% reduction in 018-0418-0) contains supplementary material, which is available to authorized users. severe and non-severe hypoglycemia rates could result in a saving of €47,769 per general popu- W. Parekh lation of 100,000 people. Novo Nordisk, Crawley, UK Conclusions: The LIHT highlights the substan- tial economic burden of insulin-related hypo- S. E. Streeton  J. Baker-Knight glycemia in Italy, particularly with regards to DRG Abacus, Bicester, Oxfordshire, UK non-severe hypoglycemia, an aspect of hypo- R. Montagnoli (&)  P. Nicoziani glycemia that is often overlooked. This analysis Novo Nordisk, Rome, Italy may aid healthcare decision-making by allow- e-mail: romo@novonordisk.com ing the costs of insulin therapies or diabetes G. Marchesini self-management programs to be balanced with Unit of Metabolic Diseases and Clinical Dietetics, Alma Mater Studiorum University, Bologna, Italy 1038 Diabetes Ther (2018) 9:1037–1047 the savings provided by reductions in Hypoglycemia therefore represents a sub- hypoglycemia. stantial economic burden to healthcare sys- Funding: Novo Nordisk, UK. tems, with costs ranging from a few Euros for a non-severe episode, to very high costs for a severe episode requiring hospitalization of the Keywords: Diabetes mellitus; Economic burden; patient [5]. Due to greater event frequency, the Hypoglycemia; Insulin; Italy treatment of non-severe hypoglycemia may incur similar or greater annual costs than that of INTRODUCTION severe hypoglycemia [6, 7]. The aim of this study was to estimate the cost of insulin-related The prevalence of diabetes is increasing world- hypoglycemia in Italy using the Local Impact of Hypoglycemia Tool (LIHT), a model used pre- wide. In Italy, 3.18 million people aged 15 years and over were estimated to be suffering from viously for analyses in the UK [5], Denmark [6] and Spain [7]. In the current analysis we also diabetes in 2016 (6.1% of the population for this age group) [1]. Type 1 diabetes (T1DM) and explore the potential variation in budget impact arising from different hypoglycemia rates. type 2 diabetes (T2DM) represent 5.4 and 91.9% of all diabetes cases in Italy, respectively, with the remaining 2.7% of cases made up of sec- METHODS ondary diabetes (2.1%) and ‘other’ types (0.5%) [2]. The LIHT was developed for the UK and the Hypoglycemia, a common adverse effect of methods have previously been described [5]. insulin therapy, is defined by the American Briefly, the cost and utilization of specific Diabetes Association as blood glucose levels healthcare resources are applied to a population below an alert value of 70 mg/dL (with clinically of adults with T1DM and T2DM and their cor- significant hypoglycemia defined as a blood responding severe and non-severe hypo- glucose level of \ 54 mg/dL) [3]. Hypoglycemia glycemia episode rates. In the current analysis, is defined as non-severe when the episode can the LIHT was used to estimate the cost of be self-treated with glucose or a sweet drink or hypoglycemia in Italy and the potential cost snack (15 g oral glucose according to Italian savings from a reduction in severe and non- treatment guidelines for diabetes) [4]. Severe severe hypoglycemia rates. hypoglycemia requires third-party assistance and treatment with oral or intravenous glucose Patients and Hypoglycemia Rates or intramuscular/subcutaneous injections of glucagon [4]. A severe episode may be treated by family or friends at home or work or may The population of Italy, number of adults, and require assistance from emergency healthcare number of adults with diabetes were obtained professionals (HCPs), in which case an ambu- from Italian National Institute of Statistics lance, accident and emergency (A&E) depart- (ISTAT) data for 2016 [1, 8, 9]. At that time the ment and hospital treatment may be required. total population of Italy was estimated to be The Associazione Medici Diabetologi and Soci- 60,579,000, with 86.3% aged C 15 years. Of eta` Italiana di Diabetologia recommend self- those aged C 15 years, 3,182,000 people suffered monitoring of blood glucose (SMBG) testing from diabetes, corresponding to a prevalence of every 15 min after the onset of hypoglycemia 6.1%. until at least two normal values are measured in All patients with T1DM (171,828) were the absence of further treatment between the assumed to be treated with insulin in this two measurements [4]. Patients may also con- analysis. For patients with T2DM, the HYPOS-1 sult their general practitioner (GP) or a diabetes study, an Italian, questionnaire-based, retro- specialist after a severe or non-severe hypo- spective study [10], was used to inform the glycemic episode. proportion of insulin-treated patients. In HYPOS-1, 30.1% of patients with T2DM were Diabetes Ther (2018) 9:1037–1047 1039 reported to be treated with insulin (15.1% with countries found that HCPs treated 33.3 and oral antidiabetic drugs [OADs] plus insulin and 43.2% of severe episodes experienced by people 15.0% with insulin alone); the remaining with T1DM and T2DM, respectively. For severe patients were treated with OADs (59.9%) and episodes treated by HCPs, it was assumed that lifestyle intervention (10.0%). Therefore, the cost of glucose and/or glucagon was inclu- 880,202 patients with T2DM were assumed to ded in the ambulance, A&E or hospitalization be treated with insulin in the current analysis tariff to avoid double counting. Due to a lack of (Table 1). available utilization data, the cost of glucose Hypoglycemia rates were also taken from and/or glucagon was assumed to be zero for HYPOS-1; patients with T1DM were reported to severe episodes treated at work/home, and for experience an average of 0.49 severe and 53.3 non-severe episodes. non-severe episodes per year [11, 12], while Patients use extra SMBG tests following a patients with T2DM experienced an average of hypoglycemic episode to monitor their blood 0.09 severe and 9.3 non-severe episodes per year glucose more closely until it is stabilized. There [10, 12]. In line with Italian guidelines, severe was a lack of data for Italy on the number of hypoglycemia in the HYPOS-1 study was extra SMBG tests used, so this information was defined as an episode of hypoglycemia that led taken from a survey-based study of 1404 to unconsciousness or required intervention of patients with diabetes in the US, the UK, Ger- a third person. Non-severe episodes were many, and France [14]. It was found that, on defined as the onset of one or more of the fol- average, patients use 5.6 extra SMBG tests in the lowing symptoms, which resolved with the week following a non-severe episode. As no ingestion of food or a sugary drink: palpitations, equivalent data were identified for severe epi- tremors, sweating, difficulty concentrating, sodes, the value for non-severe episodes was dizziness, hunger, blurred vision, confusion, or assumed to also apply to severe episodes. This is difficulty moving [10, 11]. a conservative assumption, since patients are likely to monitor their blood glucose levels more closely following a severe episode than Resource Utilization following a non-severe episode. The treatment pathway and resource utilization values applied The HYPOS-1 study informed utilization values in the model are shown in Fig. 1. for ambulance, A&E, hospitalization, diabetes specialist visits, and GP visits [10–12]. As no Cost Inputs data were identified on the percentages of sev- ere episodes treated in different settings in Italy, these values were taken from a survey of people Costs for ambulance and hospitalization were with diabetes in Germany, Spain, and the UK taken from a study estimating the costs associ- [13]. A weighted average across the three ated with emergency care and hospitalization Table 1 Population in the model Population Number of people Total Italian population 60,579,000 Adult Italian population (aged C 15 years) 52,279,677 Adult Italian population with diabetes 3,182,000 T1DM (5.4% of diabetic population; all assumed to be insulin-treated) 171,828 T2DM (91.9% of diabetic population) 2,924,258 T2DM treated with insulin (30.1% of T2DM population) 880,202 T1DM type 1 diabetes mellitus, T2DM type 2 diabetes mellitus 1040 Diabetes Ther (2018) 9:1037–1047 Fig. 1 Treatment pathway and resource utilization. A&E diabetes mellitus. Dagger () Patients treated by HCPs accident and emergency, GP general practitioner, HCP received treatment in the community (e.g., from a healthcare professional, SMBG self-monitoring of blood paramedic or medical practitioner) or in hospital. Double glucose, T1DM type 1 diabetes mellitus, T2DM type 2 dagger () Assumed because no data were identified for severe hypoglycemia in Italy [15]; hospital- combination, to explore the potential cost sav- ization is a weighted average reimbursed cost ing/increase associated with a reduction/in- per hospitalized patient with diabetes that is crease in hypoglycemia rates. All model inputs based on Diagnosis Related Group reimburse- were also varied by ± 20% to assess the level of ment methodology. Costs for A&E and GP visits uncertainty. Finally, the diabetes prevalence were taken from a cost analysis of the HYPOS-1 was varied in order to further explore the study [12]. The cost of a diabetes specialist visit impact of hypoglycemia on the Italian health- was obtained from a study of the cost of man- care system. agement of severe hypoglycemia in Italy [15] This article does not contain any studies and is in line with the cost of visits to other with human participants or animals performed medical specialists published in Italian national by any of the authors. tariffs [16]. The cost of an extra SMBG test was taken from a study of the cost-effectiveness of RESULTS SMBG in patients with T2DM in France, Ger- many, Italy, and Spain [17] and was inflated to The LIHT highlights the cost burden of hypo- 2017 Euros using health price indices for Italy glycemia in Italy for insulin-treated adults with [18]. Cost inputs are shown in Table 2. diabetes. The direct cost of a hypoglycemic episode ranges from €4.59 to €5,790.59, where Sensitivity Analyses the minimum cost corresponds to a non-severe episode for which the only cost incurred is Severe and non-severe hypoglycemia rates were additional SMBG testing and the maximum cost varied by ± 10–50%, both separately and in Diabetes Ther (2018) 9:1037–1047 1041 Table 2 Minimum and maximum cost of a hypoglycemic episode for type 1 diabetes mellitus and type 2 diabetes mellitus in Italy Resource Cost inputs (unit cost [reference]) Minimum cost of episode Maximum cost of episode Ambulance €205.00 [15]– €205.00 A&E €241.00 [12]– €241.00 Hospitalization €5317.00 [15]– €5317.00 GP visit €25.82 [12]– – Diabetes specialist visit €23.00 [15]– €23.00 Extra SMBG tests €0.82 [17, 18] €4.59 €4.59 Total n/a €4.59 €5790.59 A&E Accident and emergency, GP general practitioner, n/a not applicable, SMBG self-monitoring of blood glucose is for a severe episode that is treated in hospital episodes per year (79,218 vs. 84,196 for those and requires an ambulance, the A&E service, with T1DM). and a diabetes specialist (Table 2). The variation A reduction in hypoglycemia may be in cost per hypoglycemic episode arises due to achieved, for example, by switching a patient’s differences in healthcare resource use for indi- current insulin treatment to an insulin with a vidual patients. The estimated average cost of a better hypoglycemic profile. To explore the severe episode is €683.14 for T2DM and €129.77 potential variation in budget impact arising for T1DM, while the average cost of a non-sev- from differences in hypoglycemia event rate, ere episode is €4.59 for both T1DM and T2DM. the rates of severe and non-severe episodes were A breakdown of average costs by healthcare varied by ± 10–50%, both separately in a one- resource is shown in Fig. 2. way sensitivity analysis (Electronic Supplemen- The total direct cost of insulin-related tary Material [ESM] Table 1) and in combination hypoglycemia in Italy is estimated to be €144.7 in a two-way sensitivity analysis (ESM Table 2). million per year, of which €65 million is The results show that reductions in the rates of attributable to severe episodes and €79.6 mil- hypoglycemia events result in substantial cost lion is attributable to non-severe episodes. The savings. For example, in a sample general pop- total cost for a hypothetical general population ulation of 100,000, a 30% reduction leads to an of 100,000 people was also calculated to facili- annual cost saving of €32,211 for severe epi- tate the assessment of cost burden in a specific sodes, €39,442 for non-severe episodes and region or hospital; this is estimated to be €71,653 for both severe and non-severe episodes €238,843, with €107,370 and €131,473 due to combined. severe and non-severe episodes, respectively To assess the uncertainty surrounding the (Table 3). The total cost of hypoglycemia asso- model inputs, we performed one-way sensitivity ciated with T2DM is approximately €91.7 mil- analyses in which all unit costs and utilization lion, 1.7-fold higher than the cost associated values were varied by ± 20%. The most influ- with T1DM (€53 million; Table 3). This differ- ential variables were associated with treatment ence can be largely attributed to the approxi- setting for patients with T2DM (percentage of mately five-fold higher total cost of treating patients treated by HCPs and percentage of severe episodes in patients with T2DM than in patients hospitalized for a severe episode) and those with T1DM (approximately €54.1 million with additional SMBG testing after non-severe vs. €10.9 million), which is incurred despite episodes. The main drivers of the model are patients with T2DM experiencing fewer severe shown in Fig. 3. 1042 Diabetes Ther (2018) 9:1037–1047 Fig. 2 Cost breakdown by resource for an average severe and non-severe hypoglycemic episode. Hospital costs is the cost of A&E and hospitalization; HCP consultations is the cost of GP and diabetes specialist visits. HCP healthcare professional Table 3 Cost of insulin-related hypoglycemia in Italy Population (n) Severity of episode Total Severe Non-severe Italian general population (60,579,000) T1DM (171,828) €10,926,180 €42,055,522 €52,981,702 T2DM (2,924,258, of whom 880,202 are receiving insulin) €54,117,474 €37,589,540 €91,707,014 Total €65,043,654 €79,645,062 €144,688,715 Sample general population (100,000) T1DM (284) €18,036 €69,423 €87,459 T2DM (4827, of whom 1453 are receiving insulin) €89,334 €62,050 €151,384 Total €107,370 €131,473 €238,843 n number of patients, T1DM type 1 diabetes mellitus, T2DM type 2 diabetes mellitus ISTAT data for 2016 show that the overall associated with a total cost of €313,932, which prevalence of diabetes in Italy varies between is an increase of €75,089 compared with the regions, ranging from 3.2% in the Autonomous default model prevalence of 6.1%. A prevalence Province of Trento in the north, to 7.9% in the of 3.0% is associated with a cost of €117,724 southern region of Calabria [19]. A sensitivity (reduction of €121,119 versus the default analysis exploring the impact of diabetes model). prevalence on the cost of insulin-related hypo- Finally, alternative model inputs for glucose glycemia was therefore also performed (ESM and/or glucagon utilization, and SMBG unit Table 3). A disease prevalence of 8.0% in a cost, were tested in scenario analyses (ESM sample general population of 100,000 is Table 4). In the base case analysis the cost of Diabetes Ther (2018) 9:1037–1047 1043 Fig. 3 Tornado diagram showing the main drivers of the SMBG self-monitoring of blood glucose, T1DM type 1 model for a sample general population of 100,000. A&E diabetes mellitus, T2DM type 2 diabetes mellitus accident and emergency, HCP healthcare professional, glucose and/or glucagon was conservatively The higher average cost of a severe episode assumed to be zero for severe hypoglycemic for patients with T2DM compared with that for episodes treated at home or work as no data on those with T1DM likely reflects the fact that, on the utilization of treatment were identified. The average, people with T2DM have a shorter potential impact of this assumption was inves- duration of diabetes and are therefore less tigated by applying the average utilization and familiar with hypoglycemia and the treatment costs of the UK and Danish LIHT analyses [5, 6]; required. These individuals may be more likely this had a small effect on the overall results, than those with T1DM to require medical increasing the total cost of T1DM and T2DM by assistance for a severe episode instead of €22,724 and €18,207, respectively, from the receiving treatment from family or friends. This, default values. Applying alternative SMBG unit along with the association of T2DM with older costs had a larger effect on the results, with the age (and therefore increased frailty) and the greatest impact observed when this cost was presence of comorbidities, is likely to increase taken from a study in patients with T2DM that the need for hospital admission. Indeed, for used data on SMBG unit cost from two regions patients with diabetes experiencing severe in northern Italy [20]. hypoglycemia, T2DM was recently found to be associated with more frequent hospitalizations and longer length of hospital stay compared DISCUSSION with T1DM [12]. In the sensitivity analyses, the proportion of In this study we highlight the cost to the Italian patients with T2DM who received treatment healthcare system of insulin-related hypo- from a HCP for a hypoglycemic episode, the glycemia in adults with diabetes. The results proportion of patients with T2DM who were show that, although frequently overlooked, treated in hospital for a severe episode, and the non-severe hypoglycemia incurs the greatest number of extra SMBG tests used following cost, with an estimated total of €79.6 million non-severe episodes, had the largest effect on per year (55% of the total annual cost). The total the overall results; these are therefore the inputs cost of hypoglycemia associated with T2DM is associated with the greatest uncertainty. approximately 1.7-fold higher than that associ- Utilization of HCPs, A&E, and hospital ated with T1DM. 1044 Diabetes Ther (2018) 9:1037–1047 treatment were also identified as key drivers of insulin-related hypoglycemia; however, while the model. SMBG testing and the utilization of hypoglycemia is most commonly associated services such as HCPs and hospitals were also with insulin treatment, it is also a complication key model drivers in analyses performed using of other glucose-lowering medications, such as the LIHT in the UK [5], Denmark, [6] and Spain sulfonylureas and glinides [28]. [7]. Secondly, the hypoglycemia rates used in the Hypoglycemia is a common adverse effect of analysis are likely to be conservative. Rates of insulin therapy and despite pharmacological severe and non-severe hypoglycemia were taken and technological advances, event rates in from HYPOS-1 [10–12], a retrospective, ques- clinical practice remain high [21]. Hypo- tionnaire-based study that relied on patient glycemia is a key barrier to the intensification of recall. Hypoglycemic episodes are frequently insulin therapy; physicians acknowledge that under-reported by patients in real-world prac- many insulin-treated patients do not have ade- tice; indeed, a study of insulin-treated diabetes quate glycemic control, and 70% would treat across seven European countries found that more aggressively were hypoglycemia not a 65% of patients with T1DM and 50–59% of concern [22]. Hypoglycemia risk can be reduced patients with T2DM rarely or never inform their by switching a patient’s current insulin treat- GP or diabetes specialist about non-severe ment to insulin therapy with a better hypo- hypoglycemic events [29]. Patients may be glycemic profile; indeed, reductions of up to reluctant to report hypoglycemia for several 50% have been observed upon switching from reasons, including implications for employ- neutral protamine Hagedorn (NPH) insulin to ment, the risk of losing their driving license, insulin analogues [23, 24], or from older to and the fear that they will be perceived to have newer insulin analogues [25, 26]. However, poor disease control. The model also does not other factors also contribute to hypoglycemia account for asymptomatic hypoglycemia, an rates. In particular, structured education and aspect of diabetes associated with impaired training programs in self-management of dia- awareness of hypoglycemia. Approximately betes are now recognized as an important 50% of patients with T1DM [29, 30] and 45% strategy for minimizing hypoglycemia risk [27]. with T2DM [29, 31] were reported to experience The LIHT allows healthcare decision-makers to impaired awareness of hypoglycemia in ques- explore how introducing new insulin therapies tionnaire-based studies. or self-management programs could impact Thirdly, indirect costs incurred when healthcare budgets by enabling them to balance absence from work due to hypoglycemia results these costs with the savings that can be realized in lost productivity are not considered in the when hypoglycemia rates are reduced. In the model, as they do not directly affect the clini- current study, sensitivity analyses exploring the cian or budget holder. The financial impact on impact of varying hypoglycemia rates show that the patient and their family and friends is reductions in severe and non-severe events therefore an additional cost to society that is result in substantial cost savings. not accounted for in the current study. The costs of hypoglycemia in Italy are unli- Finally, while hypoglycemia is an acute kely to be overestimated in this study because of complication of diabetes, it may also impose a several conservative assumptions. Firstly, only long-term cost burden, for example when fear an adult, insulin-treated population is consid- of hypoglycemia leads patients to reduce or ered in the model. There are far fewer children omit an insulin dose, leading to sub-optimal than adults with diabetes in Italy, and it is dif- glucose control and increased risk of long-term ficult to speculate about the cost of hypo- complications [32, 33]. Such long-term cost glycemia in these patients compared with implications of hypoglycemia are not consid- adults. However, there will inevitably be addi- ered in the current analysis. tional costs of treating hypoglycemia for the There are some limitations associated with 21,000 children aged 0–14 years with diabetes the LIHT analysis due to a lack of resource uti- in Italy [1]. The model also only considers lization data for Italy. Firstly, data on the Diabetes Ther (2018) 9:1037–1047 1045 percentages of severe episodes treated in differ- decision-makers in their treatment choices by ent settings (by family and friends at home/- allowing the costs of insulin therapies or dia- work or by HCPs in the community/hospital) betes self-management programs to be balanced were taken from a survey of people with severe with the savings provided by reductions in hypoglycemia in Germany, Spain, and the UK hypoglycemia rates. [13]. Secondly, due to a lack of utilization data, glucose and/or glucagon costs are conserva- tively assumed to be zero for non-severe epi- ACKNOWLEDGEMENTS sodes and for severe episodes treated at home/work. When tested in a scenario analysis by applying the average utilization and costs Funding. This study, article processing from the UK and Danish LIHT analyses [5, 6], charges, and editorial assistance were funded by this assumption was found to have only a small Novo Nordisk, Crawley, UK. impact on the overall results. For severe epi- sodes treated in the community/hospital, glu- Editorial Assistance. Editorial support was cose and/or glucagon costs are assumed to be provided by DRG Abacus (funded by Novo included in the cost of ambulance, A&E, or Nordisk). hospitalization tariff to avoid double counting. Finally, as Italian data on the number of extra Authorship. All authors had full access to all SMBG tests used following hypoglycemic epi- of the data in this study and take complete sodes were not available, a survey of patients responsibility for the integrity of the data and with diabetes experiencing non-severe episodes accuracy of the data analysis. All named authors in the US, UK, Germany, and France [14] was meet the International Committee of Medical used. The values for non-severe episodes are Journal Editors (ICMJE) criteria for authorship assumed to also apply to severe episodes, a for this manuscript, take responsibility for the integrity of the work as a whole, and have given conservative assumption given that patients are likely to monitor their blood glucose levels their approval for this version to be published. more frequently following a severe episode than Disclosures. Witesh Parekh is an employee a non-severe episode. Given the potential for of Novo Nordisk, UK. Roberta Montagnoli is an regional variation in SMBG unit costs in Italy employee of Novo Nordisk, Italy. Paolo Nico- [34], and the observed impact of extra SMBG ziani is an employee of Novo Nordisk, Italy. testing in sensitivity analyses, alternative costs Giulio Marchesini participates in advisory of SMBG testing were also explored in scenario boards for Eli Lilly and Gilead, has received analyses. The greatest impact on the results was honoraria from Sanofi-Aventis, Merck Sharp & seen when the SMBG unit cost reported by Dohme and Novartis, and has been involved in Afonso et al. [20] was applied; this study was not clinical studies for Novo Nordisk, Boehringer selected for the base case analysis as the SMBG Ingelheim, Sanofi-Aventis, Eli Lilly, Gilead, unit cost reflects two northern Italian regions Genfit, Janssen, and AstraZeneca. Sophie Stree- only. ton is an employee of DRG Abacus. James Baker- Knight is an employee of DRG Abacus. CONCLUSIONS Compliance with Ethics Guidelines. This This study highlights the substantial economic article does not contain any studies with burden of insulin-related hypoglycemia for human participants or animals performed by adults with diabetes in Italy. The LIHT is a any of the authors. useful model for estimating the costs of severe Data Availability. The datasets generated and non-severe hypoglycemic episodes for Italy during and/or analyzed during the current as a whole, as well as for individual regions or hospitals. The tool may also aid healthcare 1046 Diabetes Ther (2018) 9:1037–1047 8. Italian National Institute of Statistics (ISTAT). study are available from the corresponding Demographic indicators 2016. http://www.istat.it/ author on reasonable request. en/archive/197555. Accessed November 2017. Open Access. This article is distributed 9. Italian National Institute of Statistics (ISTAT). Data under the terms of the Creative Commons table 2016. Population and households: popula- tion: demographic indicators. Percentage of popu- Attribution-NonCommercial 4.0 International lation aged 15 and over. http://dati.istat.it/?lang= License (http://creativecommons.org/licenses/ en. Accessed November 2017. by-nc/4.0/), which permits any non- commercial use, distribution, and reproduction 10. Giorda CB, Ozzello A, Gentile S, et al. Incidence and correlates of hypoglycemia in type 2 diabetes. The in any medium, provided you give appropriate Hypos-1 study. J Diabetes Metab. 2014;5:344. credit to the original author(s) and the source, provide a link to the Creative Commons license, 11. Giorda CB, Ozzello A, Gentile S, et al. Incidence and and indicate if changes were made. risk factors for severe and symptomatic hypo- glycemia in type 1 diabetes. Results of the HYPOS-1 study. Acta Diabetol. 2015;52:845–53. 12. Giorda CB, Rossi MC, Ozzello O, et al. Healthcare resource use, direct and indirect costs of hypo- REFERENCES glycemia in type 1 and type 2 diabetes, and nationwide projections. Results of the HYPOS-1 study. Nutr Metab Cardiovasc Dis. 2017;27:209–16. 1. Italian National Institute of Statistics (ISTAT). Data table 2016. Health statistics: health conditions: 13. Hammer M, Lammert M, Mejı´as SM, Kern W, Frier health status: age class. Number of people aged 15 BM. Costs of managing severe hypoglycaemia in and over suffering from diabetes. http://dati.istat.it/ three European countries. J Med Econ. ?lang=en. Accessed November 2017. 2009;12:281–90. 2. Associazione Medici Diabetologi (AMD). AMD 14. Brod M, Christensen T, Thomsen TL, Bushnell DM. annals 2010: quality indicators in diabetes care in The impact of non-severe hypoglycemic events on Italy. http://aemmedi.it/files/ANNALI-AMD/2010/ work productivity and diabetes management. Value Annali%202010%20inglese.pdf. Accessed Novem- Health. 2011;14:665–71. ber 2017. 15. Veronese G, Marchesini G, Forlani G, et al. Costs 3. American Diabetes Association. Standards of medi- associated with emergency care and hospitalization cal care in diabetes—2017. Diabetes Care. 40[Suppl for severe hypoglycemia. Nutr Metab Cardiovasc 1];S1–S135. Dis. 2016;26:345–51. 4. Associazione Medici Diabetologi (AMD) and Societa 16. Italian Ministry of Health. Tariff for specialist out- Italiana di Diabetologia (SID). Standard italiani per la patient care (Allegato 3). http://www.trovanorme. cura del diabete mellito 2016. http://www. salute.gov.it/norme/renderPdf.spring?seriegu=SG& standarditaliani.it/skin/www.standarditaliani.it/pdf/ datagu=28/01/2013&redaz=13A00528&artp=3&art= STANDARD_2016_June20.pdf 1&subart=1&subart1=10&vers=1&prog=001. Acces- sed November 2017. 5. Parekh WA, Ashley D, Chubb B, Gillies H, Evans M. Approach to assessing the economic impact of 17. Tunis SL, Willis WD, Foos V. Self-monitoring of insulin-related hypoglycaemia using the novel local blood glucose (SMBG) in patients with type 2 dia- impact of hypoglycaemia tool. Diabetic Med. betes on oral anti-diabetes drugs: cost-effectiveness 2015;32:1156–66. in France, Germany, Italy, and Spain. Curr Med Res Opin. 2010;26:163–75. 6. Hoskins N, Tikkanen CK, Pedersen-Bjergaard U. The economic impact of insulin-related hypoglycemia 18. European Central Bank. Health price indices for in Denmark: an analysis using the Local Impact of Italy, October 2017. https://www.ecb.europa.eu/ Hypoglycemia Tool. J Med Econ. 2017:20(4): stats/ecb_statistics/escb/html/table.en.html?id=JDF_ 363–370. ICP_COICOP_INX. Accessed November 2017. 7. Parekh W, Hoskins N, Baker-Knight J, de Arellano 19. Italian National Institute of Statistics (ISTAT). Data AR, Raya PM. The economic burden of insulin-re- table 2016. Health statistics: health conditions: lated hypoglycemia in Spain. Diabetes Ther. health status: regions and type of municipality. 2017;8(4):899–913. Diabetes Ther (2018) 9:1037–1047 1047 http://dati.istat.it/?lang=en. Accessed November presented at the American Diabetes Association, 2017. 76th annual scientific sessions, 10–14 June 2016. New Orleans, LA, USA. 20. Afonso M, Ryan F, Pitcher A, Lew E. Evaluating drug cost per responder and number needed to treat 27. Iqbal A, Heller SR. The role of structured education associated with lixisenatide on top of glargine when in the management of hypoglycaemia. Diabetolo- compared to rapid-acting insulin intensification gia. 2017:61(4):751–760. regimens on top of glargine, in patients with type 2 diabetes in the UK, Italy, and Spain. J Med Econ. 28. Inzucchi SE, Bergenstal RM, Buse JB, et al. Man- 2017;20:633–9. agement of hyperglycemia in type 2 diabetes, 2015: a patient-centered approach: update to a position 21. Khunti K, Alsifri S, Aronson R, et al. Rates and statement of the American Diabetes Association predictors of hypoglycaemia in 27 585 people from and the European Association for the Study of 24 countries with insulin-treated type 1 and type 2 Diabetes. Diabetes Care. 2015;38:140–9. diabetes: the global HAT study. Diabetes Obes Metab. 2016;18:907–15. 29. Ostenson CG, Geelhoed-Duijvestijn P, Lahtela J, Weitgasser R, Markert Jensen M, Pedersen-Bjergaard 22. Peyrot M, Barnett AH, Meneghini LF, Schumm- U. Self-reported non-severe hypoglycaemic events Draeger PM. Insulin adherence behaviours and in Europe. Diabetic Med. 2014;31:92–101. barriers in the multinational global attitudes of patients and physicians in insulin therapy study. 30. Pedersen-Bjergaard U, Pramming S, Thorsteinsson Diabetic Med. 2012;29:682–9. B. Recall of severe hypoglycaemia and self-esti- mated state of awareness in type 1 diabetes. Dia- 23. Horvath K, Jeitler K, Berghold A, et al. Long-acting betes Metab Res Rev. 2003;19:232–40. insulin analogues versus NPH insulin (human iso- phane insulin) for type 2 diabetes mellitus. 31. Akram K, Pedersen-Bjergaard U, Carstensen B, Cochrane Database Syst Rev. 2007:CD005613. Borch-Johnsen K, Thorsteinsson B. Frequency and risk factors of severe hypoglycaemia in insulin- 24. Monami M, Marchionni N, Mannucci E. Long-act- treated Type 2 diabetes: a cross-sectional survey. ing insulin analogues vs. NPH human insulin in Diabetic Med. 2006;23:750–6. type 1 diabetes. A meta-analysis. Diabetes Obes Metab. 2009;11:372–8. 32. Leiter LA, Yale JF, Chiasson JL, Harris S, Kleinstiver P, Sauriol L. Assessment of the impact of fear of 25. Lane W, Gerety G; Gumprecht J, Philis-Tsimikas A, hypoglycemic episodes on glycemic and hyp- Hansen CT, Nielsen TSS, Warren M. SWITCH 1: ogycemia management. Can J Diabetes. reduced hypoglycemia with insulin degludec (IDeg) 2005;29:186–92. versus insulin glargine (IGlar), both U100, in patients with T1D at high risk of hypoglycemia: a 33. Wild D, von Maltzahn R, Brohan E, Christensen T, randomized, double-blind, crossover trial. Poster Clauson P, Gonder-Frederick L. A critical review of presented at the American Diabetes Association, the literature on fear of hypoglycemia in diabetes: 76th annual scientific sessions, 10–14 June 2016. implications for diabetes management and patient New Orleans, LA, USA. education. Patient Educ Couns. 2007;68:10–5. 26. Wysham CB, Chaykin L, de la Rosa R, Handelsman 34. Italian National Agency for Regional Healthcare. Y, Troelsen LN, Kvist K, Norwood P. SWITCH 2: Outpatient services: comparison of rates. http://www. Reduced hypoglycemia with insulin degludec agenas.it/prestazioni-specialistiche-ambulatoriali- (IDeg) versus insulin glargine (IGlar), both U100, in confronto-tra-tariffe. Accessed November 2017. patients with T2D at high risk of hypoglycemia: a randomized, double-blind, crossover trial. Poster http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Diabetes Therapy Springer Journals

The Economic Burden of Insulin-Related Hypoglycemia in Adults with Diabetes: An Analysis from the Perspective of the Italian Healthcare System

Free
11 pages
Loading next page...
 
/lp/springer_journal/the-economic-burden-of-insulin-related-hypoglycemia-in-adults-with-T6H6fmJxS0
Publisher
Springer Healthcare
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/s13300-018-0418-0
Publisher site
See Article on Publisher Site

Abstract

Diabetes Ther (2018) 9:1037–1047 https://doi.org/10.1007/s13300-018-0418-0 ORIGINAL RESEARCH The Economic Burden of Insulin-Related Hypoglycemia in Adults with Diabetes: An Analysis from the Perspective of the Italian Healthcare System . . . . Witesh Parekh Sophie E. Streeton James Baker-Knight Roberta Montagnoli Paolo Nicoziani Giulio Marchesini Received: February 23, 2018 / Published online: March 29, 2018 The Author(s) 2018 rates (0.49 severe and 53.3 non-severe episodes ABSTRACT per year for T1DM, and 0.09 severe and 9.3 non- severe episodes per year for T2DM). Uncertainty Introduction: The aim of this analysis was to around model inputs was explored through estimate the cost of insulin-related hypo- sensitivity and scenario analyses. glycemia in adult patients with diabetes in Italy Results: The direct cost of insulin-related using the Local Impact of Hypoglycemia Tool hypoglycemia in Italy is estimated at €144.7 (LIHT), and to explore the effect of different million per year, with €65 million hypoglycemia rates on budget impact. attributable to severe episodes and €79.6 mil- Methods: Direct costs and healthcare resource lion due to non-severe episodes. The total cost utilization were estimated for severe and non- of hypoglycemia is approximately 1.7-fold severe hypoglycemic episodes in Italy and higher for T2DM (€91.7 million) than for T1DM applied to the population of adults with type 1 (€53 million). The cost of a hypoglycemic epi- diabetes (T1DM) and type 2 diabetes (T2DM) sode ranges from €4.59 for a non-severe event and their corresponding hypoglycemia episode where additional self-monitoring of blood glu- cose (SMBG) testing is the only cost incurred, to Enhanced content To view enhanced content for this €5790.59 for a severe event that also requires an article go to https://doi.org/10.6084/m9.figshare. ambulance, A&E, hospitalization, and a visit to a diabetes specialist. A reduction in hypo- Electronic supplementary material The online glycemia event rates could result in substantial version of this article (https://doi.org/10.1007/s13300- cost savings; for example, a 20% reduction in 018-0418-0) contains supplementary material, which is available to authorized users. severe and non-severe hypoglycemia rates could result in a saving of €47,769 per general popu- W. Parekh lation of 100,000 people. Novo Nordisk, Crawley, UK Conclusions: The LIHT highlights the substan- tial economic burden of insulin-related hypo- S. E. Streeton  J. Baker-Knight glycemia in Italy, particularly with regards to DRG Abacus, Bicester, Oxfordshire, UK non-severe hypoglycemia, an aspect of hypo- R. Montagnoli (&)  P. Nicoziani glycemia that is often overlooked. This analysis Novo Nordisk, Rome, Italy may aid healthcare decision-making by allow- e-mail: romo@novonordisk.com ing the costs of insulin therapies or diabetes G. Marchesini self-management programs to be balanced with Unit of Metabolic Diseases and Clinical Dietetics, Alma Mater Studiorum University, Bologna, Italy 1038 Diabetes Ther (2018) 9:1037–1047 the savings provided by reductions in Hypoglycemia therefore represents a sub- hypoglycemia. stantial economic burden to healthcare sys- Funding: Novo Nordisk, UK. tems, with costs ranging from a few Euros for a non-severe episode, to very high costs for a severe episode requiring hospitalization of the Keywords: Diabetes mellitus; Economic burden; patient [5]. Due to greater event frequency, the Hypoglycemia; Insulin; Italy treatment of non-severe hypoglycemia may incur similar or greater annual costs than that of INTRODUCTION severe hypoglycemia [6, 7]. The aim of this study was to estimate the cost of insulin-related The prevalence of diabetes is increasing world- hypoglycemia in Italy using the Local Impact of Hypoglycemia Tool (LIHT), a model used pre- wide. In Italy, 3.18 million people aged 15 years and over were estimated to be suffering from viously for analyses in the UK [5], Denmark [6] and Spain [7]. In the current analysis we also diabetes in 2016 (6.1% of the population for this age group) [1]. Type 1 diabetes (T1DM) and explore the potential variation in budget impact arising from different hypoglycemia rates. type 2 diabetes (T2DM) represent 5.4 and 91.9% of all diabetes cases in Italy, respectively, with the remaining 2.7% of cases made up of sec- METHODS ondary diabetes (2.1%) and ‘other’ types (0.5%) [2]. The LIHT was developed for the UK and the Hypoglycemia, a common adverse effect of methods have previously been described [5]. insulin therapy, is defined by the American Briefly, the cost and utilization of specific Diabetes Association as blood glucose levels healthcare resources are applied to a population below an alert value of 70 mg/dL (with clinically of adults with T1DM and T2DM and their cor- significant hypoglycemia defined as a blood responding severe and non-severe hypo- glucose level of \ 54 mg/dL) [3]. Hypoglycemia glycemia episode rates. In the current analysis, is defined as non-severe when the episode can the LIHT was used to estimate the cost of be self-treated with glucose or a sweet drink or hypoglycemia in Italy and the potential cost snack (15 g oral glucose according to Italian savings from a reduction in severe and non- treatment guidelines for diabetes) [4]. Severe severe hypoglycemia rates. hypoglycemia requires third-party assistance and treatment with oral or intravenous glucose Patients and Hypoglycemia Rates or intramuscular/subcutaneous injections of glucagon [4]. A severe episode may be treated by family or friends at home or work or may The population of Italy, number of adults, and require assistance from emergency healthcare number of adults with diabetes were obtained professionals (HCPs), in which case an ambu- from Italian National Institute of Statistics lance, accident and emergency (A&E) depart- (ISTAT) data for 2016 [1, 8, 9]. At that time the ment and hospital treatment may be required. total population of Italy was estimated to be The Associazione Medici Diabetologi and Soci- 60,579,000, with 86.3% aged C 15 years. Of eta` Italiana di Diabetologia recommend self- those aged C 15 years, 3,182,000 people suffered monitoring of blood glucose (SMBG) testing from diabetes, corresponding to a prevalence of every 15 min after the onset of hypoglycemia 6.1%. until at least two normal values are measured in All patients with T1DM (171,828) were the absence of further treatment between the assumed to be treated with insulin in this two measurements [4]. Patients may also con- analysis. For patients with T2DM, the HYPOS-1 sult their general practitioner (GP) or a diabetes study, an Italian, questionnaire-based, retro- specialist after a severe or non-severe hypo- spective study [10], was used to inform the glycemic episode. proportion of insulin-treated patients. In HYPOS-1, 30.1% of patients with T2DM were Diabetes Ther (2018) 9:1037–1047 1039 reported to be treated with insulin (15.1% with countries found that HCPs treated 33.3 and oral antidiabetic drugs [OADs] plus insulin and 43.2% of severe episodes experienced by people 15.0% with insulin alone); the remaining with T1DM and T2DM, respectively. For severe patients were treated with OADs (59.9%) and episodes treated by HCPs, it was assumed that lifestyle intervention (10.0%). Therefore, the cost of glucose and/or glucagon was inclu- 880,202 patients with T2DM were assumed to ded in the ambulance, A&E or hospitalization be treated with insulin in the current analysis tariff to avoid double counting. Due to a lack of (Table 1). available utilization data, the cost of glucose Hypoglycemia rates were also taken from and/or glucagon was assumed to be zero for HYPOS-1; patients with T1DM were reported to severe episodes treated at work/home, and for experience an average of 0.49 severe and 53.3 non-severe episodes. non-severe episodes per year [11, 12], while Patients use extra SMBG tests following a patients with T2DM experienced an average of hypoglycemic episode to monitor their blood 0.09 severe and 9.3 non-severe episodes per year glucose more closely until it is stabilized. There [10, 12]. In line with Italian guidelines, severe was a lack of data for Italy on the number of hypoglycemia in the HYPOS-1 study was extra SMBG tests used, so this information was defined as an episode of hypoglycemia that led taken from a survey-based study of 1404 to unconsciousness or required intervention of patients with diabetes in the US, the UK, Ger- a third person. Non-severe episodes were many, and France [14]. It was found that, on defined as the onset of one or more of the fol- average, patients use 5.6 extra SMBG tests in the lowing symptoms, which resolved with the week following a non-severe episode. As no ingestion of food or a sugary drink: palpitations, equivalent data were identified for severe epi- tremors, sweating, difficulty concentrating, sodes, the value for non-severe episodes was dizziness, hunger, blurred vision, confusion, or assumed to also apply to severe episodes. This is difficulty moving [10, 11]. a conservative assumption, since patients are likely to monitor their blood glucose levels more closely following a severe episode than Resource Utilization following a non-severe episode. The treatment pathway and resource utilization values applied The HYPOS-1 study informed utilization values in the model are shown in Fig. 1. for ambulance, A&E, hospitalization, diabetes specialist visits, and GP visits [10–12]. As no Cost Inputs data were identified on the percentages of sev- ere episodes treated in different settings in Italy, these values were taken from a survey of people Costs for ambulance and hospitalization were with diabetes in Germany, Spain, and the UK taken from a study estimating the costs associ- [13]. A weighted average across the three ated with emergency care and hospitalization Table 1 Population in the model Population Number of people Total Italian population 60,579,000 Adult Italian population (aged C 15 years) 52,279,677 Adult Italian population with diabetes 3,182,000 T1DM (5.4% of diabetic population; all assumed to be insulin-treated) 171,828 T2DM (91.9% of diabetic population) 2,924,258 T2DM treated with insulin (30.1% of T2DM population) 880,202 T1DM type 1 diabetes mellitus, T2DM type 2 diabetes mellitus 1040 Diabetes Ther (2018) 9:1037–1047 Fig. 1 Treatment pathway and resource utilization. A&E diabetes mellitus. Dagger () Patients treated by HCPs accident and emergency, GP general practitioner, HCP received treatment in the community (e.g., from a healthcare professional, SMBG self-monitoring of blood paramedic or medical practitioner) or in hospital. Double glucose, T1DM type 1 diabetes mellitus, T2DM type 2 dagger () Assumed because no data were identified for severe hypoglycemia in Italy [15]; hospital- combination, to explore the potential cost sav- ization is a weighted average reimbursed cost ing/increase associated with a reduction/in- per hospitalized patient with diabetes that is crease in hypoglycemia rates. All model inputs based on Diagnosis Related Group reimburse- were also varied by ± 20% to assess the level of ment methodology. Costs for A&E and GP visits uncertainty. Finally, the diabetes prevalence were taken from a cost analysis of the HYPOS-1 was varied in order to further explore the study [12]. The cost of a diabetes specialist visit impact of hypoglycemia on the Italian health- was obtained from a study of the cost of man- care system. agement of severe hypoglycemia in Italy [15] This article does not contain any studies and is in line with the cost of visits to other with human participants or animals performed medical specialists published in Italian national by any of the authors. tariffs [16]. The cost of an extra SMBG test was taken from a study of the cost-effectiveness of RESULTS SMBG in patients with T2DM in France, Ger- many, Italy, and Spain [17] and was inflated to The LIHT highlights the cost burden of hypo- 2017 Euros using health price indices for Italy glycemia in Italy for insulin-treated adults with [18]. Cost inputs are shown in Table 2. diabetes. The direct cost of a hypoglycemic episode ranges from €4.59 to €5,790.59, where Sensitivity Analyses the minimum cost corresponds to a non-severe episode for which the only cost incurred is Severe and non-severe hypoglycemia rates were additional SMBG testing and the maximum cost varied by ± 10–50%, both separately and in Diabetes Ther (2018) 9:1037–1047 1041 Table 2 Minimum and maximum cost of a hypoglycemic episode for type 1 diabetes mellitus and type 2 diabetes mellitus in Italy Resource Cost inputs (unit cost [reference]) Minimum cost of episode Maximum cost of episode Ambulance €205.00 [15]– €205.00 A&E €241.00 [12]– €241.00 Hospitalization €5317.00 [15]– €5317.00 GP visit €25.82 [12]– – Diabetes specialist visit €23.00 [15]– €23.00 Extra SMBG tests €0.82 [17, 18] €4.59 €4.59 Total n/a €4.59 €5790.59 A&E Accident and emergency, GP general practitioner, n/a not applicable, SMBG self-monitoring of blood glucose is for a severe episode that is treated in hospital episodes per year (79,218 vs. 84,196 for those and requires an ambulance, the A&E service, with T1DM). and a diabetes specialist (Table 2). The variation A reduction in hypoglycemia may be in cost per hypoglycemic episode arises due to achieved, for example, by switching a patient’s differences in healthcare resource use for indi- current insulin treatment to an insulin with a vidual patients. The estimated average cost of a better hypoglycemic profile. To explore the severe episode is €683.14 for T2DM and €129.77 potential variation in budget impact arising for T1DM, while the average cost of a non-sev- from differences in hypoglycemia event rate, ere episode is €4.59 for both T1DM and T2DM. the rates of severe and non-severe episodes were A breakdown of average costs by healthcare varied by ± 10–50%, both separately in a one- resource is shown in Fig. 2. way sensitivity analysis (Electronic Supplemen- The total direct cost of insulin-related tary Material [ESM] Table 1) and in combination hypoglycemia in Italy is estimated to be €144.7 in a two-way sensitivity analysis (ESM Table 2). million per year, of which €65 million is The results show that reductions in the rates of attributable to severe episodes and €79.6 mil- hypoglycemia events result in substantial cost lion is attributable to non-severe episodes. The savings. For example, in a sample general pop- total cost for a hypothetical general population ulation of 100,000, a 30% reduction leads to an of 100,000 people was also calculated to facili- annual cost saving of €32,211 for severe epi- tate the assessment of cost burden in a specific sodes, €39,442 for non-severe episodes and region or hospital; this is estimated to be €71,653 for both severe and non-severe episodes €238,843, with €107,370 and €131,473 due to combined. severe and non-severe episodes, respectively To assess the uncertainty surrounding the (Table 3). The total cost of hypoglycemia asso- model inputs, we performed one-way sensitivity ciated with T2DM is approximately €91.7 mil- analyses in which all unit costs and utilization lion, 1.7-fold higher than the cost associated values were varied by ± 20%. The most influ- with T1DM (€53 million; Table 3). This differ- ential variables were associated with treatment ence can be largely attributed to the approxi- setting for patients with T2DM (percentage of mately five-fold higher total cost of treating patients treated by HCPs and percentage of severe episodes in patients with T2DM than in patients hospitalized for a severe episode) and those with T1DM (approximately €54.1 million with additional SMBG testing after non-severe vs. €10.9 million), which is incurred despite episodes. The main drivers of the model are patients with T2DM experiencing fewer severe shown in Fig. 3. 1042 Diabetes Ther (2018) 9:1037–1047 Fig. 2 Cost breakdown by resource for an average severe and non-severe hypoglycemic episode. Hospital costs is the cost of A&E and hospitalization; HCP consultations is the cost of GP and diabetes specialist visits. HCP healthcare professional Table 3 Cost of insulin-related hypoglycemia in Italy Population (n) Severity of episode Total Severe Non-severe Italian general population (60,579,000) T1DM (171,828) €10,926,180 €42,055,522 €52,981,702 T2DM (2,924,258, of whom 880,202 are receiving insulin) €54,117,474 €37,589,540 €91,707,014 Total €65,043,654 €79,645,062 €144,688,715 Sample general population (100,000) T1DM (284) €18,036 €69,423 €87,459 T2DM (4827, of whom 1453 are receiving insulin) €89,334 €62,050 €151,384 Total €107,370 €131,473 €238,843 n number of patients, T1DM type 1 diabetes mellitus, T2DM type 2 diabetes mellitus ISTAT data for 2016 show that the overall associated with a total cost of €313,932, which prevalence of diabetes in Italy varies between is an increase of €75,089 compared with the regions, ranging from 3.2% in the Autonomous default model prevalence of 6.1%. A prevalence Province of Trento in the north, to 7.9% in the of 3.0% is associated with a cost of €117,724 southern region of Calabria [19]. A sensitivity (reduction of €121,119 versus the default analysis exploring the impact of diabetes model). prevalence on the cost of insulin-related hypo- Finally, alternative model inputs for glucose glycemia was therefore also performed (ESM and/or glucagon utilization, and SMBG unit Table 3). A disease prevalence of 8.0% in a cost, were tested in scenario analyses (ESM sample general population of 100,000 is Table 4). In the base case analysis the cost of Diabetes Ther (2018) 9:1037–1047 1043 Fig. 3 Tornado diagram showing the main drivers of the SMBG self-monitoring of blood glucose, T1DM type 1 model for a sample general population of 100,000. A&E diabetes mellitus, T2DM type 2 diabetes mellitus accident and emergency, HCP healthcare professional, glucose and/or glucagon was conservatively The higher average cost of a severe episode assumed to be zero for severe hypoglycemic for patients with T2DM compared with that for episodes treated at home or work as no data on those with T1DM likely reflects the fact that, on the utilization of treatment were identified. The average, people with T2DM have a shorter potential impact of this assumption was inves- duration of diabetes and are therefore less tigated by applying the average utilization and familiar with hypoglycemia and the treatment costs of the UK and Danish LIHT analyses [5, 6]; required. These individuals may be more likely this had a small effect on the overall results, than those with T1DM to require medical increasing the total cost of T1DM and T2DM by assistance for a severe episode instead of €22,724 and €18,207, respectively, from the receiving treatment from family or friends. This, default values. Applying alternative SMBG unit along with the association of T2DM with older costs had a larger effect on the results, with the age (and therefore increased frailty) and the greatest impact observed when this cost was presence of comorbidities, is likely to increase taken from a study in patients with T2DM that the need for hospital admission. Indeed, for used data on SMBG unit cost from two regions patients with diabetes experiencing severe in northern Italy [20]. hypoglycemia, T2DM was recently found to be associated with more frequent hospitalizations and longer length of hospital stay compared DISCUSSION with T1DM [12]. In the sensitivity analyses, the proportion of In this study we highlight the cost to the Italian patients with T2DM who received treatment healthcare system of insulin-related hypo- from a HCP for a hypoglycemic episode, the glycemia in adults with diabetes. The results proportion of patients with T2DM who were show that, although frequently overlooked, treated in hospital for a severe episode, and the non-severe hypoglycemia incurs the greatest number of extra SMBG tests used following cost, with an estimated total of €79.6 million non-severe episodes, had the largest effect on per year (55% of the total annual cost). The total the overall results; these are therefore the inputs cost of hypoglycemia associated with T2DM is associated with the greatest uncertainty. approximately 1.7-fold higher than that associ- Utilization of HCPs, A&E, and hospital ated with T1DM. 1044 Diabetes Ther (2018) 9:1037–1047 treatment were also identified as key drivers of insulin-related hypoglycemia; however, while the model. SMBG testing and the utilization of hypoglycemia is most commonly associated services such as HCPs and hospitals were also with insulin treatment, it is also a complication key model drivers in analyses performed using of other glucose-lowering medications, such as the LIHT in the UK [5], Denmark, [6] and Spain sulfonylureas and glinides [28]. [7]. Secondly, the hypoglycemia rates used in the Hypoglycemia is a common adverse effect of analysis are likely to be conservative. Rates of insulin therapy and despite pharmacological severe and non-severe hypoglycemia were taken and technological advances, event rates in from HYPOS-1 [10–12], a retrospective, ques- clinical practice remain high [21]. Hypo- tionnaire-based study that relied on patient glycemia is a key barrier to the intensification of recall. Hypoglycemic episodes are frequently insulin therapy; physicians acknowledge that under-reported by patients in real-world prac- many insulin-treated patients do not have ade- tice; indeed, a study of insulin-treated diabetes quate glycemic control, and 70% would treat across seven European countries found that more aggressively were hypoglycemia not a 65% of patients with T1DM and 50–59% of concern [22]. Hypoglycemia risk can be reduced patients with T2DM rarely or never inform their by switching a patient’s current insulin treat- GP or diabetes specialist about non-severe ment to insulin therapy with a better hypo- hypoglycemic events [29]. Patients may be glycemic profile; indeed, reductions of up to reluctant to report hypoglycemia for several 50% have been observed upon switching from reasons, including implications for employ- neutral protamine Hagedorn (NPH) insulin to ment, the risk of losing their driving license, insulin analogues [23, 24], or from older to and the fear that they will be perceived to have newer insulin analogues [25, 26]. However, poor disease control. The model also does not other factors also contribute to hypoglycemia account for asymptomatic hypoglycemia, an rates. In particular, structured education and aspect of diabetes associated with impaired training programs in self-management of dia- awareness of hypoglycemia. Approximately betes are now recognized as an important 50% of patients with T1DM [29, 30] and 45% strategy for minimizing hypoglycemia risk [27]. with T2DM [29, 31] were reported to experience The LIHT allows healthcare decision-makers to impaired awareness of hypoglycemia in ques- explore how introducing new insulin therapies tionnaire-based studies. or self-management programs could impact Thirdly, indirect costs incurred when healthcare budgets by enabling them to balance absence from work due to hypoglycemia results these costs with the savings that can be realized in lost productivity are not considered in the when hypoglycemia rates are reduced. In the model, as they do not directly affect the clini- current study, sensitivity analyses exploring the cian or budget holder. The financial impact on impact of varying hypoglycemia rates show that the patient and their family and friends is reductions in severe and non-severe events therefore an additional cost to society that is result in substantial cost savings. not accounted for in the current study. The costs of hypoglycemia in Italy are unli- Finally, while hypoglycemia is an acute kely to be overestimated in this study because of complication of diabetes, it may also impose a several conservative assumptions. Firstly, only long-term cost burden, for example when fear an adult, insulin-treated population is consid- of hypoglycemia leads patients to reduce or ered in the model. There are far fewer children omit an insulin dose, leading to sub-optimal than adults with diabetes in Italy, and it is dif- glucose control and increased risk of long-term ficult to speculate about the cost of hypo- complications [32, 33]. Such long-term cost glycemia in these patients compared with implications of hypoglycemia are not consid- adults. However, there will inevitably be addi- ered in the current analysis. tional costs of treating hypoglycemia for the There are some limitations associated with 21,000 children aged 0–14 years with diabetes the LIHT analysis due to a lack of resource uti- in Italy [1]. The model also only considers lization data for Italy. Firstly, data on the Diabetes Ther (2018) 9:1037–1047 1045 percentages of severe episodes treated in differ- decision-makers in their treatment choices by ent settings (by family and friends at home/- allowing the costs of insulin therapies or dia- work or by HCPs in the community/hospital) betes self-management programs to be balanced were taken from a survey of people with severe with the savings provided by reductions in hypoglycemia in Germany, Spain, and the UK hypoglycemia rates. [13]. Secondly, due to a lack of utilization data, glucose and/or glucagon costs are conserva- tively assumed to be zero for non-severe epi- ACKNOWLEDGEMENTS sodes and for severe episodes treated at home/work. When tested in a scenario analysis by applying the average utilization and costs Funding. This study, article processing from the UK and Danish LIHT analyses [5, 6], charges, and editorial assistance were funded by this assumption was found to have only a small Novo Nordisk, Crawley, UK. impact on the overall results. For severe epi- sodes treated in the community/hospital, glu- Editorial Assistance. Editorial support was cose and/or glucagon costs are assumed to be provided by DRG Abacus (funded by Novo included in the cost of ambulance, A&E, or Nordisk). hospitalization tariff to avoid double counting. Finally, as Italian data on the number of extra Authorship. All authors had full access to all SMBG tests used following hypoglycemic epi- of the data in this study and take complete sodes were not available, a survey of patients responsibility for the integrity of the data and with diabetes experiencing non-severe episodes accuracy of the data analysis. All named authors in the US, UK, Germany, and France [14] was meet the International Committee of Medical used. The values for non-severe episodes are Journal Editors (ICMJE) criteria for authorship assumed to also apply to severe episodes, a for this manuscript, take responsibility for the integrity of the work as a whole, and have given conservative assumption given that patients are likely to monitor their blood glucose levels their approval for this version to be published. more frequently following a severe episode than Disclosures. Witesh Parekh is an employee a non-severe episode. Given the potential for of Novo Nordisk, UK. Roberta Montagnoli is an regional variation in SMBG unit costs in Italy employee of Novo Nordisk, Italy. Paolo Nico- [34], and the observed impact of extra SMBG ziani is an employee of Novo Nordisk, Italy. testing in sensitivity analyses, alternative costs Giulio Marchesini participates in advisory of SMBG testing were also explored in scenario boards for Eli Lilly and Gilead, has received analyses. The greatest impact on the results was honoraria from Sanofi-Aventis, Merck Sharp & seen when the SMBG unit cost reported by Dohme and Novartis, and has been involved in Afonso et al. [20] was applied; this study was not clinical studies for Novo Nordisk, Boehringer selected for the base case analysis as the SMBG Ingelheim, Sanofi-Aventis, Eli Lilly, Gilead, unit cost reflects two northern Italian regions Genfit, Janssen, and AstraZeneca. Sophie Stree- only. ton is an employee of DRG Abacus. James Baker- Knight is an employee of DRG Abacus. CONCLUSIONS Compliance with Ethics Guidelines. This This study highlights the substantial economic article does not contain any studies with burden of insulin-related hypoglycemia for human participants or animals performed by adults with diabetes in Italy. The LIHT is a any of the authors. useful model for estimating the costs of severe Data Availability. The datasets generated and non-severe hypoglycemic episodes for Italy during and/or analyzed during the current as a whole, as well as for individual regions or hospitals. The tool may also aid healthcare 1046 Diabetes Ther (2018) 9:1037–1047 8. Italian National Institute of Statistics (ISTAT). study are available from the corresponding Demographic indicators 2016. http://www.istat.it/ author on reasonable request. en/archive/197555. Accessed November 2017. Open Access. This article is distributed 9. Italian National Institute of Statistics (ISTAT). Data under the terms of the Creative Commons table 2016. Population and households: popula- tion: demographic indicators. Percentage of popu- Attribution-NonCommercial 4.0 International lation aged 15 and over. http://dati.istat.it/?lang= License (http://creativecommons.org/licenses/ en. Accessed November 2017. by-nc/4.0/), which permits any non- commercial use, distribution, and reproduction 10. Giorda CB, Ozzello A, Gentile S, et al. Incidence and correlates of hypoglycemia in type 2 diabetes. The in any medium, provided you give appropriate Hypos-1 study. J Diabetes Metab. 2014;5:344. credit to the original author(s) and the source, provide a link to the Creative Commons license, 11. Giorda CB, Ozzello A, Gentile S, et al. Incidence and and indicate if changes were made. risk factors for severe and symptomatic hypo- glycemia in type 1 diabetes. Results of the HYPOS-1 study. Acta Diabetol. 2015;52:845–53. 12. Giorda CB, Rossi MC, Ozzello O, et al. Healthcare resource use, direct and indirect costs of hypo- REFERENCES glycemia in type 1 and type 2 diabetes, and nationwide projections. Results of the HYPOS-1 study. Nutr Metab Cardiovasc Dis. 2017;27:209–16. 1. Italian National Institute of Statistics (ISTAT). Data table 2016. Health statistics: health conditions: 13. Hammer M, Lammert M, Mejı´as SM, Kern W, Frier health status: age class. Number of people aged 15 BM. Costs of managing severe hypoglycaemia in and over suffering from diabetes. http://dati.istat.it/ three European countries. J Med Econ. ?lang=en. Accessed November 2017. 2009;12:281–90. 2. Associazione Medici Diabetologi (AMD). AMD 14. Brod M, Christensen T, Thomsen TL, Bushnell DM. annals 2010: quality indicators in diabetes care in The impact of non-severe hypoglycemic events on Italy. http://aemmedi.it/files/ANNALI-AMD/2010/ work productivity and diabetes management. Value Annali%202010%20inglese.pdf. Accessed Novem- Health. 2011;14:665–71. ber 2017. 15. Veronese G, Marchesini G, Forlani G, et al. Costs 3. American Diabetes Association. Standards of medi- associated with emergency care and hospitalization cal care in diabetes—2017. Diabetes Care. 40[Suppl for severe hypoglycemia. Nutr Metab Cardiovasc 1];S1–S135. Dis. 2016;26:345–51. 4. Associazione Medici Diabetologi (AMD) and Societa 16. Italian Ministry of Health. Tariff for specialist out- Italiana di Diabetologia (SID). Standard italiani per la patient care (Allegato 3). http://www.trovanorme. cura del diabete mellito 2016. http://www. salute.gov.it/norme/renderPdf.spring?seriegu=SG& standarditaliani.it/skin/www.standarditaliani.it/pdf/ datagu=28/01/2013&redaz=13A00528&artp=3&art= STANDARD_2016_June20.pdf 1&subart=1&subart1=10&vers=1&prog=001. Acces- sed November 2017. 5. Parekh WA, Ashley D, Chubb B, Gillies H, Evans M. Approach to assessing the economic impact of 17. Tunis SL, Willis WD, Foos V. Self-monitoring of insulin-related hypoglycaemia using the novel local blood glucose (SMBG) in patients with type 2 dia- impact of hypoglycaemia tool. Diabetic Med. betes on oral anti-diabetes drugs: cost-effectiveness 2015;32:1156–66. in France, Germany, Italy, and Spain. Curr Med Res Opin. 2010;26:163–75. 6. Hoskins N, Tikkanen CK, Pedersen-Bjergaard U. The economic impact of insulin-related hypoglycemia 18. European Central Bank. Health price indices for in Denmark: an analysis using the Local Impact of Italy, October 2017. https://www.ecb.europa.eu/ Hypoglycemia Tool. J Med Econ. 2017:20(4): stats/ecb_statistics/escb/html/table.en.html?id=JDF_ 363–370. ICP_COICOP_INX. Accessed November 2017. 7. Parekh W, Hoskins N, Baker-Knight J, de Arellano 19. Italian National Institute of Statistics (ISTAT). Data AR, Raya PM. The economic burden of insulin-re- table 2016. Health statistics: health conditions: lated hypoglycemia in Spain. Diabetes Ther. health status: regions and type of municipality. 2017;8(4):899–913. Diabetes Ther (2018) 9:1037–1047 1047 http://dati.istat.it/?lang=en. Accessed November presented at the American Diabetes Association, 2017. 76th annual scientific sessions, 10–14 June 2016. New Orleans, LA, USA. 20. Afonso M, Ryan F, Pitcher A, Lew E. Evaluating drug cost per responder and number needed to treat 27. Iqbal A, Heller SR. The role of structured education associated with lixisenatide on top of glargine when in the management of hypoglycaemia. Diabetolo- compared to rapid-acting insulin intensification gia. 2017:61(4):751–760. regimens on top of glargine, in patients with type 2 diabetes in the UK, Italy, and Spain. J Med Econ. 28. Inzucchi SE, Bergenstal RM, Buse JB, et al. Man- 2017;20:633–9. agement of hyperglycemia in type 2 diabetes, 2015: a patient-centered approach: update to a position 21. Khunti K, Alsifri S, Aronson R, et al. Rates and statement of the American Diabetes Association predictors of hypoglycaemia in 27 585 people from and the European Association for the Study of 24 countries with insulin-treated type 1 and type 2 Diabetes. Diabetes Care. 2015;38:140–9. diabetes: the global HAT study. Diabetes Obes Metab. 2016;18:907–15. 29. Ostenson CG, Geelhoed-Duijvestijn P, Lahtela J, Weitgasser R, Markert Jensen M, Pedersen-Bjergaard 22. Peyrot M, Barnett AH, Meneghini LF, Schumm- U. Self-reported non-severe hypoglycaemic events Draeger PM. Insulin adherence behaviours and in Europe. Diabetic Med. 2014;31:92–101. barriers in the multinational global attitudes of patients and physicians in insulin therapy study. 30. Pedersen-Bjergaard U, Pramming S, Thorsteinsson Diabetic Med. 2012;29:682–9. B. Recall of severe hypoglycaemia and self-esti- mated state of awareness in type 1 diabetes. Dia- 23. Horvath K, Jeitler K, Berghold A, et al. Long-acting betes Metab Res Rev. 2003;19:232–40. insulin analogues versus NPH insulin (human iso- phane insulin) for type 2 diabetes mellitus. 31. Akram K, Pedersen-Bjergaard U, Carstensen B, Cochrane Database Syst Rev. 2007:CD005613. Borch-Johnsen K, Thorsteinsson B. Frequency and risk factors of severe hypoglycaemia in insulin- 24. Monami M, Marchionni N, Mannucci E. Long-act- treated Type 2 diabetes: a cross-sectional survey. ing insulin analogues vs. NPH human insulin in Diabetic Med. 2006;23:750–6. type 1 diabetes. A meta-analysis. Diabetes Obes Metab. 2009;11:372–8. 32. Leiter LA, Yale JF, Chiasson JL, Harris S, Kleinstiver P, Sauriol L. Assessment of the impact of fear of 25. Lane W, Gerety G; Gumprecht J, Philis-Tsimikas A, hypoglycemic episodes on glycemic and hyp- Hansen CT, Nielsen TSS, Warren M. SWITCH 1: ogycemia management. Can J Diabetes. reduced hypoglycemia with insulin degludec (IDeg) 2005;29:186–92. versus insulin glargine (IGlar), both U100, in patients with T1D at high risk of hypoglycemia: a 33. Wild D, von Maltzahn R, Brohan E, Christensen T, randomized, double-blind, crossover trial. Poster Clauson P, Gonder-Frederick L. A critical review of presented at the American Diabetes Association, the literature on fear of hypoglycemia in diabetes: 76th annual scientific sessions, 10–14 June 2016. implications for diabetes management and patient New Orleans, LA, USA. education. Patient Educ Couns. 2007;68:10–5. 26. Wysham CB, Chaykin L, de la Rosa R, Handelsman 34. Italian National Agency for Regional Healthcare. Y, Troelsen LN, Kvist K, Norwood P. SWITCH 2: Outpatient services: comparison of rates. http://www. Reduced hypoglycemia with insulin degludec agenas.it/prestazioni-specialistiche-ambulatoriali- (IDeg) versus insulin glargine (IGlar), both U100, in confronto-tra-tariffe. Accessed November 2017. patients with T2D at high risk of hypoglycemia: a randomized, double-blind, crossover trial. Poster

Journal

Diabetes TherapySpringer Journals

Published: Mar 29, 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