PharmacoEconomics Open (2018) 2:209–219 https://doi.org/10.1007/s41669-017-0050-3 OR IGINAL RESEARCH ARTIC L E Direct Medical Costs of Type 2 Diabetes in France: An Insurance Claims Database Analysis 1 2 3 4 • • • • Bernard Charbonnel Dominique Simon Jean Dallongeville Isabelle Bureau 5 5 6 4 • • • Sylvie Dejager Laurie Levy-Bachelot Julie Gourmelen Bruno Detournay Published online: 7 August 2017 The Author(s) 2017. This article is an open access publication Abstract 12.5; 53.9% male) matched with a control group of 76,406 Objectives Our objects was to estimate the direct health- individuals without diabetes. Overall per patient per year care costs of type 2 diabetes mellitus (T2DM) in France in medical expenditures were €6506 ± 10,106 in the T2DM 2013. group as compared with €3668 ± 6954 in the control group. Methods Data were drawn from a random sample of The cost difference between the two groups was €2838 per &600,000 patients registered in the French national health patient per year, mainly due to hospitalizations, medication and insurances database, which covers 90% of the French nursing care costs. Total per capita annual costs were lowest for population. An algorithm was used to select patients with patients receiving metformin monotherapy (€4153 ± 6170) T2DM. Direct healthcare costs from a collective perspec- and highest for those receiving insulin (€12,890). However, tive were derived from the database and compared with apart from patients receiving insulin, costs did not differ those from a control group to estimate the cost of diabetes markedly across the different oral treatment patterns. and related comorbidities. Overall direct costs were also Conclusion Extrapolating these results to the whole T2DM compared according to the diabetes therapies used population in France, total direct costs of diagnosed T2DM throughout the year 2013. in 2013 was estimated at over €8.5 billion. This estimate Results Cost analysis was available for a sample of 25,987 highlights the substantial economic burden of this condi- patients with T2DM (mean age 67.5 ± standard deviation tion on society. Electronic supplementary material The online version of this article (doi:10.1007/s41669-017-0050-3) contains supplementary material, which is available to authorized users. & Bruno Detournay Julie Gourmelen Bruno.email@example.com firstname.lastname@example.org Bernard Charbonnel Ho ˆ tel Dieu Hospital, Nantes, France Bernard.Charbonnel@univ-nantes.fr Diabetes Department and ICAN (Institute of Dominique Simon Cardiometabolism And Nutrition), Pitie ´ Hospital, Paris, email@example.com France Jean Dallongeville INSERM-U1167, Lille, France firstname.lastname@example.org Cemka-Eval, 43 Bd du Mare ´chal Joffre, 92 340 Isabelle Bureau Bourg-la-Reine, France email@example.com Merck Sharpe & Dohme, Courbevoie, France Sylvie Dejager firstname.lastname@example.org INSERM UMS 011, Villejuif, France Laurie Levy-Bachelot email@example.com 210 B. Charbonnel et al. Thus, we considered it timely to reassess the economic burden of T2DM in France. Recently, a number of public Key Points for Decision Makers health insurance databases have become available in France, which have made it possible to collect quasi-ex- In 2013, the average overall direct healthcare haustive information on healthcare resource utilisation in expenditure for a patient with type 2 diabetes representative samples of the French general population. mellitus (T2DM) in France was €6506. ´ ´ ´ ´ The EGB (Echantillon Generaliste des Beneﬁciaires) The direct cost of diabetes and related morbidities database is a representative sample of French National was estimated at €2838 per patient per year. insurance funds that covers around 95% of the French population . This database has been used in several Patients receiving insulin incurred the highest costs recent studies to document medication prescription, costs (€12,890). or outcomes in different disease groups in the French Costs did not differ markedly across the different general population [7–9], including in patients with dia- oral treatment patterns. betes . The objective of this study was to describe the characteristics of patients with T2DM in France and their treatments and to estimate their total direct healthcare expenditure as the cost directly related to diabetes care and 1 Introduction related comorbidities. Although some of the well-known diabetes drugs are As in most Western countries, the prevalence of diabetes in modestly priced generics, new brand-name drugs continue France has risen sharply over the last two decades, reaching to be introduced at higher prices. Their mechanisms of 4.7% of the adult general population in the French national action differ: some induce fewer side effects and others survey conducted in 2013  (updated to 5% in 2015 ). have greater efﬁcacy. It is therefore interesting to compare Over 2009–2013, the growth rate of the diabetic population total direct costs according to the antidiabetic agents pre- was estimated at 2.3% each year . On this basis, it can be scribed in a real-life setting. The cost of a drug is one estimated that around 3.3 million people had diabetes in criterion that may guide treatment choice among available France at the beginning of 2015. The majority of these glucose-lowering agents for a patient with T2DM [11, 12]. cases correspond to type 2 diabetes mellitus (T2DM), and However, the cost of treatment alone does not reﬂect the this population continues to grow because of the aging budget consequences of the drug choice. Therefore, we population and lifestyle factors. In the 2007 ENTRED were also interested in reporting hospital and community study , 92% of all cases of diabetes were T2DM, and costs for different pharmacological therapy options this proportion may have risen with the increases in obesity (monotherapy, dual, other) and to use an example to to reach 95% or more. illustrate how a simple comparison of the direct costs of Given the large number of individuals affected and the diabetes treatment may result in misinterpretation. high cost of managing the complications of diabetes, the economic burden from diabetes is considerable. Data from the 2007 ENTRED study provided an estimate of €12.5 2 Methods billion for the total healthcare costs reimbursed by the National Sickness Fund for people with diabetes, whatever This was a retrospective study of healthcare resource the type. However, this ﬁgure is likely to have evolved consumption and associated costs generated by a repre- since then because of the increase in the treated population, sentative sample of patients with T2DM identiﬁed in the the introduction of new treatments since 2007—notably French general health insurance claims database in 2013. glucagon-like peptide-1 (GLP-1) analogues and dipep- Direct healthcare costs were estimated from a collective tidylpeptidase-4 (DPP-4) inhibitors (gliptins)—and the perspective regardless of the institution or individual. impact of treatment guidelines recommending more intensive treatment regimens earlier in the disease course 2.1 The EGB Database . More recent estimates were also published by the National Sickness Fund, but these only considered reim- The EGB (Echantillon Ge ´ne ´raliste des Be ´ne ´ﬁciaires) bursed expenditure. The total healthcare costs reimbursed database is a random sample of beneﬁciaries of the prin- for people with diabetes was estimated at €19 billion in cipal French public health insurance scheme, which covers 2012. Several methods were then used to estimate the approximately 95% of the total French population (66 burden of diabetes, resulting in estimates varying between million individuals) . The sample of 1/97 randomly €7.7 billion and €10 billion . selected individuals included in the EGB database Direct Medical Costs of Type 2 Diabetes in France 211 corresponds to around 600,000 individuals. All information in the database is anonymous. The EGB database contains limited sociodemographic and medical data on healthcare users but comprehensive reimbursement records on healthcare consumption in community and hospital care. Sociodemographic information is restricted to age, sex, and place of residence. All items of medical consumption in the public or private sector that are eligible for reimbursement are documented, notably consultations, paraclinical tests, medication, devices and, since 2005, hospitalisations . Items ineligible for reimbursement, such as over-the- counter drugs, are absent from the database and cannot be identiﬁed. In addition, information on inpatient rehabilita- tion is not available. For costing purposes, hospitalisations in acute care facilities (medicine, surgery or obstetrics) are coded in the EGB database through a speciﬁc diagnosis- related group (DRG). Medication is identiﬁed in the data- base through the relevant Anatomical Therapeutic Chem- ical (ATC) classiﬁcation code. Date of death is documented in the database but not the cause of death. The only types of data in the EGB database associated with an explicit diagnosis are hospitalisation and eligibility for full insurance coverage due to a severe chronic disease (Affection de Longue Dure ´e [ALD] status). In the case of hospitalisations, the diagnosis can be identiﬁed since each hospital stay is valued on the basis of a unique DRG, which is coded using the International Classiﬁcation of Disease, Fig. 1 Decision tree for identifying patients with T1DM and T2DM. tenth revision (ICD-10) codes . The reasons for hos- ALD affection de longue dure ´e (full insurance coverage due to a severe chronic disease), ICD-10 International Classiﬁcation of pitalisation are coded either as primary diagnoses (PD; the Diseases, tenth revision, N no, T1DM type 1 diabetes mellitus, condition for which the patient was hospitalised), related T2DM type 2 diabetes mellitus, Y yes diagnoses (RD; any underlying condition that may have history with diabetes as an identiﬁed diagnosis (PD, RD or been related to the PD) or as associated diagnoses (AD; comorbidities that may affect the course or cost of hospi- AD), ALD status for diabetes through the associated ICD- 10 code and insulin treatment history. The relevant ICD-10 talisation). In the case of ALD status, eligible diseases are identiﬁed on a restrictive list established by the French codes for diabetes applied in this decision tree were E10 for T1DM and E11 for T2DM. We assumed that all patients public health insurance schemes that speciﬁes the equiva- lent ICD-10 disease code. with T1DM had an ALD status or were hospitalized at least once over a 2-year period for diabetes. 2.2 Subjects 2.2.2 Identiﬁcation of Controls 2.2.1 Identiﬁcation of Patients with T2DM Beyond the cost of care for people with diabetes, the cost of diabetes care was estimated using a case–control Patients with diabetes were identiﬁed in the EGB database according to a criterion usually applied in France . This approach [16, 17]. A control sample was built, matched for age, sex, and criterion was EITHER reimbursement for three distinct prescriptions for antidiabetic medication (including insu- region of residence to the index diabetes case sample using the quota method. The overall demographic structure of the lin), or two prescriptions when large packs were delivered, on three different dates within 2 consecutive years or ALD diabetes sample was determined and quotas allocated for each age, sex, and geographic area class. Subjects without status for diabetes. All adult patients (aged C18 years) diabetes were then selected randomly from the EGB fulﬁlling this criterion during 2012 or 2013 were retained. database and assigned to each quota until three controls had Additionally, a decision tree was used to distinguish been identiﬁed for each case in each age, sex, and geo- patients with type 1 diabetes mellitus (T1DM) from those with T2DM (Fig. 1). This was based on hospitalisation graphic area class. 212 B. Charbonnel et al. 2.3 Data Collection categorical data are presented as frequency counts and percentages. The occurrence of comorbidities and com- For each eligible patient, information was extracted from plications was compared between the diabetes cases and the database on demographics (sex and age in 2013), ALD the matched controls using the Chi-squared (v ) test or status for diabetes or other chronic diseases, and comor- Fisher’s exact test as appropriate. bidities or complications. Costs were compared between cases and controls for Comorbidities and complications of interest were each individual cost component and for total costs using ischaemic heart disease, stroke, cardiac failure, treated the Mann–Whitney U test. The differential cost between hypertension, treated dyslipidemia, kidney transplantation, the cases and controls was calculated as a measure of the chronic kidney failure, haemodialysis, retinopathy, hyper- health economic burden of T2DM, including complications glycaemia, sleep apnoea and cancer. These were identiﬁed and related comorbidities. This burden was extrapolated to from three sources: hospitalisations in which these the whole French population using national diabetes comorbidities were identiﬁed as a diagnosis through the prevalence data . relevant ICD-10 disease classiﬁcation code, the presence of The cost of diabetes was reported for the different serial reimbursements for prescription of relevant speciﬁc pharmacological therapy options (monotherapy, dual, medications, and ALD status identiﬁed through the rele- other) according to hospital and community costs. Fur- vant ICD-10 disease classiﬁcation code. thermore, an exploratory analysis of average annual con- All healthcare resource consumption documented in the sumption of care in patients with T2DM was conducted in EGB database between 1 January and 31 December 2013 patients receiving dual therapy throughout the year 2013 were identiﬁed. Healthcare resource variables of interest (quarters 1–4) depending on the type of dual therapy were medication (for diabetes and other conditions) and (metformin ? DPP-4 inhibitor and metformin ? sulfony- other reimbursed pharmacy products, hospitalisations, lurea). For this analysis, patients treated with the same dual consultations (specialists, general practitioners and den- therapy throughout the year 2013 (same regimen during the tists), paramedical care (nurses and physiotherapists), lab- quarters 1–4) were selected. Overall direct costs of oratory tests, medical devices, medical transport and other healthcare were compared between the two groups. community care costs. As patients may have different characteristics according Costs presented for reimbursement were identiﬁed for to their treatment group, we conducted an adjusted analysis each item in the EGB database. Ambulatory costs were (see Appendix I in the Electronic Supplementary Material directly estimated using reimbursement data from the EGB [ESM]). A regression analysis (generalized linear model database. For hospitalisations, cost per DRG was estimated ﬁtted with a Gamma distribution, after logarithmic trans- using the National Cost database per DRG . All costs formation) was set up to explain the total healthcare con- were estimated using a collective perspective (collective sumption. Only patients who were treated throughout 2013 perspective limited to direct healthcare costs) according to were analysed. The model took into account the following the current guidelines for health economic evaluation in variables: France . • Dual therapy in the last quarter of 2013: metformin ? DPP-4 inhibitor or metformin ? sulfonylurea 2.4 Statistical Analysis • Age in four groups (\55, 55–64, 65–75, C75 years) • Sex Two main analysis populations were considered in this • Presence of full coverage for a longstanding illness study. For the description of the study population, all (ALD status) patients with T2DM identiﬁed in the EGB database were • Simpliﬁed Charlson Comorbidity Index score based on considered. For the cost analysis, patients with T2DM and diseases requiring hospitalizations in 2012–2013 (with- controls who had died before the end of the cost assessment out taking into account patients age, which was already period (31 December 2013) were excluded to avoid bias considered in the model) due to differences in follow-up duration. A third population • Patient area of residence used to identify antidiabetic medication delivered during • Dual therapy duration (from the initial prescription of the last quarter of 2013 (the last period assessable in the the dual therapy to the end of 2013). study) was evaluated considering patients diagnosed with All statistical analyses were performed using SAS T2DM by the third quarter of 2013 at the latest. Our presentation of the study population is principally software, version 9.2 (Cary, NC, USA). A bilateral prob- ability threshold of 0.05 was used to determine statistical descriptive. Continuous data are presented as mean val- ues ± standard deviation (SD) or as median values, and signiﬁcance. Direct Medical Costs of Type 2 Diabetes in France 213 2.5 Ethical Considerations Table 1 Study population demographics Characteristic Main study Cost analysis Since this was a retrospective study of an anonymised population population database and had no inﬂuence on patient care, ethics (N = 28,708) (N = 25,987) committee approval was not required. Age (years) 67.2 ± 12.9 67.3 ± 12.5 Median (range) 67 (18–112) 67 (18–104) \45 years 1246 (4.4) 987 (3.8) 3 Results Sex Men (%) 15,538 (54.1) 14,001 (53.9) 3.1 Study Population Data are presented as mean ± standard deviation or n (%) unless otherwise indicated Overall, 30,155 patients with diabetes were identiﬁed in the EGB database. After application of the decision tree 3.2 Comorbidities and Complications rules, 28,708 (95.2%) were considered to have T2DM. The majority of these patients (N = 23,182; 80.8%) were The 2013 prevalence of complications and comorbidities classiﬁed on the basis of an explicit ICD-10 code for was analysed using the cost analysis population in both T2DM associated with a hospitalisation record or ALD diabetes cases and controls (Table 2). The proportion of status; 5244 patients (18.3%) were identiﬁed on the basis subjects with an ischaemic complication, heart failure, of a prescription for antidiabetic medication alone. kidney disease or retinopathy or who were receiving Among this T2DM population, 19.4% were receiving medication for the treatment of hypertension or dyslipi- insulin. After 2721 patients who had died by 31 demia was markedly higher for cases than for controls December 2013 were excluded from the study popula- (Table 2). No difference was observed for all cancers tion, 25,987 (90.5%) remained for the cost analysis. The combined, but the frequency of hepatobiliary cancers and distribution of subjects across the analysis populations is pancreatic cancers was higher in cases, whereas the fre- presentedinFig. 2. quency of prostate cancers was higher in controls. Overall, 54.1% of patients with T2DM were men, and the mean age of the sample was 67 years (Table 1). The 3.3 Antidiabetic Medication majority (74.4%) beneﬁted from ALD status (full reim- bursement of all related care) for their diabetes (64% for Antidiabetic medication delivered during the last quarter of \6 years), and 42.1% were classed as such for another 2013 (the last period assessable in the study) was evaluated pathology, most frequently hypertension (N = 3723; by class (Table 3) among patients diagnosed with T2DM 13.0%) and ischaemic heart disease (N = 1847; 6.4%). by the third quarter of 2013 at the latest. Among the 27,829 patients with T2DM considered in this cross-sectional analysis, 3777 patients (13.6%) had no medication deliv- ered. Around half of the patients receiving treatment 30,155 paents idenﬁed with diabetes in the EGB database in (48.9%) were prescribed a monotherapy (including insulin 2012-2013 (100%) monotherapy), principally metformin (27.3%), and 25.6% were prescribed dual therapy. Insulin (alone or in combi- nation) was prescribed for 19.4% of patients, and a GLP-1 MAIN STUDY POPULATION analogue (alone or in combination) was prescribed for 28,708 paents fulﬁlling criteria for 3.5% of patients . type 2 diabetes (95.2%) 3.4 Costs 2,721 paents died before 31/12/2013 The total annual per capita cost incurred by patients with (9.5%) T2DM was €6506. These costs were 1.77 times higher than those incurred by the matched control group (€3668). The COST ANALYSIS POPULATION CONTROLS (without diabetes) 25,987 paents surviving at 76,406 paents surviving at speciﬁc cost associated with T2DM and its related 31/12/2013 (90.5%) 31/12/2013 comorbidities or complications (difference between the two groups) was €2838 per patient per year (pppy). Fig. 2 Patient distribution and analysis populations. Percentages Ambulatory costs accounted for around two-thirds of costs, were calculated with respect to the previous line. EGB Echantillon and hospital costs accounted for the remaining third ´ ´ ´ ´ Generaliste des Beneﬁciaires database 214 B. Charbonnel et al. Table 2 Comorbidities and complications in diabetes cases and controls Comorbidities and complications Cases (N = 25,987) Controls (N = 76,406) p value Ischaemic heart disease 3160 (12.2) 4977 (6.5) \0.0001 Incident stroke in 2013 141 (0.5) 298 (0.4) 0.0011 Heart failure ALD 481 (1.9) 953 (1.2) \0.0001 Hospitalisation 2009–2013 main diagnosis 851 (3.3) 983 (1.3) \0.0001 Hospitalisation 2009–2013, secondary diagnosis 332 (1.3) 449 (0.6) \0.0001 Either ALD or hospitalisation 1471 (5.7) 2068 (2.7) \0.0001 Treated hypertension in 2013 20,192 (77.7) 36,773 (48.1) \0.0001 Treated dyslipidemia in 2013 15,441 (59.4) 21,853 (28.6) \0.0001 Kidney transplantation in 2013 2 (\0.1) 2 (\0.1) 0.2681 Chronic kidney disease 249 (1.0) 415 (0.5) \0.0001 Haemodialysis (C45 sessions per year) 118 (0.5) 96 (0.1) \0.0001 Terminal kidney disease 120 (0.5) 98 (0.1) \0.0001 Retinal laser treatment in 2013 87 (0.3) 28 (\0.1) \0.0001 Retinopathy in 2013 609 (2.3) 734 (1.0) \0.0001 Hypoglycaemia In 2013 162 (0.6) – \0.0001 Between 2009 and 2013 477 (1.8) 3 (\0.1) \0.0001 Sleep apnoea Hospitalisation in 2013 581 (2.2) 485 (0.6) \0.0001 Reimbursement for CPAP in 2013 1470 (5.7) 1587 (2.1) \0.0001 Either hospitalisation or CPAP in 2013 1732 (6.7) 1835 (2.4) \0.0001 Cancer 3174 (12.2) 9000 (11.8) 0.0615 Prostate 593 (18.7) 1964 (21.8) 0.0002 Breast 523 (16.5) 1595 (17.7) 0.1118 Colon 237 (7.5) 622 (6.9) 0.2931 Bladder 206 (6.5) 521 (5.8) 0.1517 Lung 131 (4.1) 353 (3.9) 0.6112 Other skin cancers 126 (4.0) 359 (4.0) 0.9622 Rectum 99 (3.1) 231 (2.6) 0.0994 Liver or biliary cancer 68 (2.1) 71 (0.8) \0.0001 Kidney 87 (2.7) 232 (2.6) 0.6206 Mouth 71 (2.2) 200 (2.2) 0.9615 Pancreas 54 (1.7) 54 (0.6) \0.0001 Thyroid 68 (2.1) 160 (1.8) 0.1926 Melanoma 70 (2.2) 180 (2.0) 0.4829 Data are presented as n (%) unless otherwise indicated ALD Affection de Longue Dure ´e, CPAP continuous positive airway pressure (Table 4). The highest individual costs incurred related to Costs were reported between the principal therapeutic hospitalisations (33.2% of total cost), medications (23.7%) patterns delivered in the last quarter of 2013 (Fig. 3). Total and nursing care (10.9%). For each individual cost com- per capita annual costs were lowest in patients receiving ponent, expenditure for diabetes cases was signiﬁcantly metformin monotherapy (€4153 ± 6170) and highest in higher (p \ 0.0001) than for controls by a factor ranging those receiving insulin (€12,890 ± 14,735). However, from 1.2-fold (for physician consultations, interventions apart from patients receiving insulin, costs did not differ and physiotherapy) to fourfold (for nursing costs). markedly across the different treatment patterns. Direct Medical Costs of Type 2 Diabetes in France 215 Table 3 Medications delivered during last quarter of 2013 3.5 Costs According to Treatment Pattern: Patients Treated with Dual Therapy Treatment N = 27,829 No documented treatment 3777 A speciﬁc analysis of average annual consumption of care Monotherapy in patients with T2DM was conducted in patients receiving Metformin 6568 (27.3) dual therapy throughout the year 2013 (quarters 1–4), Sulphonylurea 2120 (8.8) comparing patients treated with metformin ? DPP-4 inhi- Other 1228 (5.1) bitor (N = 1846) and those receiving metformin ? sul- Dual therapy fonylurea (N = 1811). Metformin ? sulphonylurea 2381 (9.9) In the real-life setting, the average cost of dual therapy with metformin ? a DPP-4 inhibitor was estimated at €605 Metformin ? DPP-4 inhibitor 2399 (10.0) pppy (all taxes included), and the average cost of dual Sulphonylurea ? DPP-4 inhibitor 431 (1.8) therapy with metformin ? a sulfonylurea was estimated at Other 943 (3.9) €270 pppy, a signiﬁcant difference of €335 pppy Triple therapy (p\ 0.0001, ?124%). Metformin ? sulphonylurea ? DPP-4 inhibitor 2019 (8.4) When considering average overall direct healthcare Other 1031 (4.3) costs, the gap between patients treated with metformin ? a Other multi-therapies, excluding insulin 243 (1.0) DDP-4 inhibitor and those treated with metformin ? a Insulin regimens sulfonylurea reduced to €167 per year (p \ 0.0001; ?4%). Insulin alone 1864 (7.7) The difference in costs for hypoglycaemic agents was Insulin ? metformin 741 (3.1) partially offset by the reduced need for paramedics Insulin ? DPP-4 inhibitor 91 (0.4) (p = 0.0131), including nursing (p = 0.0004), and a non- Insulin ? metformin ? DPP-4 inhibitor 247 (1.0) signiﬁcant reduction in inpatient costs (p = 0.1436) Insulin ? sulphonylurea 131 (0.5) (Table 5). Other 1615 (6.7) Finally, patients treated throughout 2013 with met- Data are presented as n (%) unless otherwise indicated. Percentages formin ? a DPP-4 inhibitor were younger than those are calculated with respect to the 24,052 patients with a documented treated with metformin ? a sulfonylurea (65.0 ± 10.6 vs. treatment 67.7 ± 11.0 years; p\ 0.0001); their geographical distri- DPP-4 dipeptidylpeptidase-4 bution (p \ 0.0001) and ALD status coverage (83 vs. 86%, Table 4 Per capita costs Costs Cases (N = 25,987) Controls (N = 76,406) p value presented for reimbursement by diabetes cases and controls Hospital costs 2159 (33.2) ± 6502 1304 (35.5) ± 4632 \0.0001 Ambulatory costs Medication 1541 (23.7) ± 2057 731 (19.9) ± 1693 \0.0001 Physician consultations 233 (3.6) ± 213 191 (5.2) ± 198 \0.0001 Home visits 58 (0.9) ± 159 33 (0.9) ± 110 \0.0001 Interventions 319 (4.9) ± 735 275 (7.5) ± 634 \0.0001 Nursing care 712 (10.9) ± 2468 182 (4.9) ± 1120 \0.0001 Physiotherapy 150 (2.3) ± 506 122 (3.3) ± 420 \0.0001 Medical devices 583 (8.9) ± 1146 309 (8.4) ± 744 \0.0001 Dental care 145 (2.2) ± 489 179 (4.9) ± 556 \0.0001 Laboratory tests 201 (3.1) ± 247 119 (3.2) ± 191 \0.0001 Transportation 236 (3.6) ± 1325 107 (2.9) ± 612 \0.0001 Total ambulatory costs Total community costs 4347 (66.8) ± 5230 2364 (64.5) ± 3421 \0.0001 Total costs 6506 (100) ± 9955 3668 (100) ± 6854 \0.0001 Median (IQR) 3093 (1627–7069) 1530 (665–558) Costs are presented in €, year 2013 values, as mean (%) ± standard deviation unless otherwise indicated IQR interquartile range, SD standard deviation 216 B. Charbonnel et al. Fig. 3 Per capita annual costs by patients with diabetes according to the principal therapeutic patterns (last 2013 quarter only). The numbers at the end of the horizontal columns represent total costs, the ﬁlled bars hospital costs and the open bars community costs. DPP-4-I inhibitor of dipeptidylpeptidase-4, SU sulphonylurea Table 5 Per capita costs presented for reimbursement per dual therapy (entire year) Costs per item per year Patients treated with Patients treated with metformin p value metformin ? a sulfonylurea ? a DPP-4 inhibitor N 1811 (100.0) 1846 (100.0) Total amount presented for reimbursement 3969 ± 5337 4136 ± 5693 \0.0001 Hospitalisations 1002 ± 3228 917 ± 3270 0.1436 Ambulatory care Medications 1181 ± 1697 1542 ± 1998 \0.0001 Hypoglycaemic agents 605 ± 231 270 ± 204 \0.0001 Medical fees 660 ± 847 633 ± 777 0.1561 Paramedics 373 ± 1476 257 ± 1024 0.0131 Nursing care 262 ± 1344 143 ± 815 0.0004 Medical devices 353 ± 693 352 ± 724 0.2466 Transportation 91 ± 494 99 ± 764 0.0003 Laboratory tests 164 ± 160 165 ± 166 0.3762 Dental care 138 ± 479 162 ± 499 0.0013 Costs are presented in €, year 2013 values as N (%) or mean ± standard deviation DPP-4 dipeptidylpeptidase-4 p = 0.0052) also differed, and these results suggest the two No signiﬁcant difference in hospitalisations was populations are not strictly comparable in a real-life set- observed between the groups of patients treated with a dual ting. Therefore, we used a multivariate regression model to therapy (p = 0.98). Although using a DPP-4 inhibitor was compare costs according to the dual therapy prescribed more expensive, this was partly offset by reduced medical during 2013 as described in the methodology: analyses fees (honoraria) (-5.8%; p = 0.04) and need for para- were adjusted on age, sex, a simpliﬁed Charlson Comor- medics (-25.5%; p \ 0.0001) in patients treated with bidity Index score, ALD status and area of residence metformin ? a DPP-4 inhibitor. (Appendix 1 in the ESM). The adjusted average overall healthcare costs of patients treated with metformin ? a DPP-4 inhibitor were 6.3% 4 Discussion (95% conﬁdence interval [CI] 0.4–12.6%) higher than the average overall healthcare costs of patients treated with In this study, we analysed direct healthcare costs accrued metformin ? a sulfonylurea in 2013 (p = 0.0348). Good- by patients with T2DM in 2013 from a collective per- ness of ﬁt of the adjusted model was estimated to be spective using a bottom-up approach in a representative acceptable (v statistic [or the deviance] divided by the sample of the national health insurance claims database degrees of freedom was 0.82). (EGB). A sample of 28,708 patients fulﬁlling criteria for Direct Medical Costs of Type 2 Diabetes in France 217 T2DM using a decision tree was identiﬁed. Both selection healthcare expenditure for people with diabetes reaches and classiﬁcation algorithms used in this study have their €19.5 billion. Such results were remarkably close to those own limitations. The selection process excluded the least of de Lagasnerie et al.  despite methodological differ- severe cases (without both antidiabetic medication and ences (e.g. this study considered both T1DM and T2DM, ALD status). The classiﬁcation algorithm is likely to be and estimates are based on 2012 reimbursed expenditures imperfect. However, this will have only a marginal impact only). on the quality of the results. The diabetes burden is likely to have increased since Direct costs identiﬁed were those ﬁguring in the national 2013, mainly due to the rising prevalence (in France, the health insurance database. Some costs supported by average annual increase in diabetes prevalence was ?2.3% patients and their families were not considered, such as between 2009 and 2013 ). A series of pricing measures some over-the-counter drugs. However, such costs are has limited the pace of price increase. usually considered to be very low , especially for In general, patients with diabetes tended to be more ill patients with diabetes who have 100% reimbursement than controls, with signiﬁcantly higher rates of a broad status or are covered by a supplementary mutual health range of comorbidities. Furthermore, as the analysis con- insurance. sidered only patients who did not die during the year, The cost analysis was performed for all patients alive at estimates provided in Table 2 may underestimate the full 31 December 2013 and thus took into account expenditure burden of these comorbidities. As such, it is interesting that for a full 12-month period. The mean annual per capita cost the two most frequent reasons for hospitalisation for accrued by patients with T2DM was €6506, which was patients with diabetes were haemodialysis and cancer nearly twice as high as that of a matched sample of control chemotherapy. Both kidney disease in general and regular subjects without diabetes. These data can be compared with haemodialysis (C45 sessions/year) in particular were more a previous estimate of the cost of T2DM in France deter- frequent in diabetes cases than in controls, consistent with mined in the 2007 ENTRED survey , which was €4890 the well-characterised renal complications of diabetes . (€10,413 in patients treated with insulin and €3625 in On the other hand, although certain cancers were more patients not using insulin). The costs of diabetes would thus frequent in diabetes cases (hepatobiliary and pancreatic have increased by 30% between 2007 and 2013. This cancers), the most frequent cancer type, prostate cancer, increase concerned all individual cost components. How- was in fact more frequent in controls. Again, this is con- ever, differences in methodologies between the two studies sistent with the known association of these cancers with may introduce bias in the comparison (i.e. costing in the diabetes . ENTRED survey was not conducted using the method- Excluding patients who died during the study period may ological guidelines for economic evaluation published in have introduced a bias. Some evidence suggests that patients 2012 ; as an example, hospital stays were valuated at the end of life drive healthcare spending, but this remains using tariffs and not costs, and the cost perspective was that controversial . Conversely, regardless of the period of the National Sickness Fund). considered (year, month, etc.), including the costs of people We observed little change in the pattern of antidiabetic who died early in the period would underestimate the costs of medication use since the 2007 ENTRED study , with illness. Finally, we consider that end-of-life care is an the exception of the appearance of a signiﬁcant proportion important and challenging issue that needs to be addressed of patients receiving a DPP-4 inhibitor, which reached independently. A speciﬁc study on the cost of care at the end 25.3% in 2013. This was accompanied by a fall in pre- of life among people with diabetes is necessary. scriptions for sulphonylureas, from 49 to 30%. The pro- The largest individual cost components were related to portion of patients receiving insulin remained stable at hospitalisations, medications and nursing care. This raises 19%, and the proportion of patients receiving monotherapy many questions, both about the organisation of care and with an oral antidiabetic drug was also stable, at 42%. The about medication prescribing practices in France. use of metformin increased. With regards to the healthcare system, it is well known Patients with diabetes generated an additional per capita that patients with diabetes in France are often referred to cost of €2838 compared with the matched controls in 2013. hospital even for issues that could be managed in an This cost represents both the cost of T2DM care and the ambulatory setting. In the French Healthcare system, pri- costs of diabetes-related comorbidities and complications. vate nursing care is often used to provide an answer to Extrapolated to the estimated total French population with social and medical issues encountered by patients with treated T2DM of nearly 3 million individuals (4.7%), this diabetes. represents a total annual cost of €8.5 billion, corresponding In our analysis, medication costs included those related to around 5% of all healthcare expenditure in France (es- both to antidiabetic therapy and to other drugs that may be timated at €186.7 billion in 2013 ). The overall prescribed to treat comorbidities or related cardiovascular 218 B. Charbonnel et al. risk factors. In the USA , the cost of antidiabetic agents importance of public health programmes aimed at reducing and diabetes supplies was estimated to represent 12% of the incidence of T2DM through the promotion of healthy the total direct medical costs, and prescription medications lifestyles and at the prevention of diabetic complications for the complications of diabetes or comorbidities was (better glycaemic control, less therapeutic inertia and better estimated at 18%. In France, 23.7% of diabetes costs are compliance with lifestyle measures and drugs) and of related to medications. However, when comparing the developing integrated care programmes for patients with costs of antidiabetic agents, it is interesting to consider not diabetes that may be less costly for society. only the price of drugs but also the budget impact of total care associated with the use of the drug. Our analysis Author Contributions BC, SD and LLB conceptualized the study provides an exploratory illustration of this based on a cost idea. BD designed the study, developed the data analysis plan and analysis of two dual therapies. The yearly cost of dual provided critical feedback at the design stage, wrote the ﬁrst version therapy with metformin ? a DPP-4 inhibitor appears to be of the manuscript, and is the guarantor of the study. IB performed all more than double that of metformin ? a sulfonylurea when the analyses. All authors critically reviewed the manuscript and approved the ﬁnal version. considering only the costs of the antidiabetic therapy. This estimate may be debatable because it was based on the Compliance with Ethical Standards public price of drugs, all taxes included, in France in 2013. Unknown rebates and paybacks are negotiated between Data Availability Statement Data that supported these analyses are available from Dr. Bruno Detournay. payers and pharmaceutical companies for licensed drugs, resulting in lower prices in practice. However, considering Funding MSD France provided funding for this project. the overall annual healthcare expenditure, the adjusted difference among the two populations was only 6% in Conﬂict of interest IB and BD are employed by Cemka-Eval, a contract research organisation that was contracted by MSD France for 2013. Higher costs of treatment are partially offset by this study. BD has also received honoraria for consultancy from MSD, savings on other cost items. Novo-Nordisk, Sanoﬁ. LLB and SD are employed by MSD France. Health-related retrospective databases, particularly BC, DS and JD have received honoraria from MSD France for par- claims databases, continue to be an important data source ticipating on the scientiﬁc board for this study. BC has also received honoraria for consultancy from AstraZeneca, Boehringer-Ingelheim, for outcomes research. A search of PubMed ((claims Janssen Pharmaceuticals, Eli Lilly, MSD, Novartis, Novo-Nordisk analysis[MeSH Terms]) OR (Claims[Title/Abstract]) AND and Sanoﬁ. DS has served as an expert for Sanoﬁ Aventis and Takeda (‘‘2015/01/01’’[Date-MeSH] : ‘‘2015/12/31’’[Date- and has been a member of a board for Astellas, MSD and Novartis, MeSH])) indicated that analyses of insurance claims data and received fees for all these activities. JG has no conﬂicts of interest. were published in at least 500 articles in 2015. In France, EGB contains exhaustive information on reimbursement Open Access This article is distributed under the terms of the claims for a representative sample of the national health Creative Commons Attribution-NonCommercial 4.0 International insurance database covering 95% of the French population. License (http://creativecommons.org/licenses/by-nc/4.0/), which per- Nevertheless, like other databases, EGB has some limita- mits any noncommercial use, distribution, and reproduction in any medium, provided you give appropriate credit to the original tions: diagnoses rely on algorithms instead of adjudicated author(s) and the source, provide a link to the Creative Commons events, some patient subgroups, such as students and civil license, and indicate if changes were made. servants, are not well represented in the database; diag- noses are only documented if the patient was hospitalised or eligible for full reimbursement for an ALD. Other lim- itations include the absence of information on medications, References tests or interventions that were prescribed but never delivered; limited documentation of sociodemographic and 1. Mandereau-Bruno L, Denis P, Fagot-Campagna AF, Fosse-Edorh clinical characteristics of the insuree; and the risk of S. Pre ´valence du diabe `te traite ´ pharmacologiquement et disparite ´s incomplete or inaccurate coding of medical events. territoriales en France en 2012. Bull Epide ´miol Hebd. 2014;30–31:493–9. 2. Sante ´ Publique France. Pre ´valence et incidence du diabe `te. http:// invs.santepubliquefrance.fr/Dossiers-thematiques/Maladies- 5 Conclusion chroniques-et-traumatismes/Diabete/Donnees-epidemiologiques/ Prevalence-et-incidence-du-diabete. Accessed 7 Feb 2017. 3. Druet C, Roudier C, Romon I, Assogba F, Bourdel-Marchasson I, This insurance claims database study estimated the total Eschwege ER-F, M., et al. Echantillon national temoin cost of T2DM (including related comorbidities and com- ´ ´ representatif des personnes diabetiques, Entred 2007–2010. Car- plications) in France in 2013 to be €8.5 billion, corre- ´ ´ ´ ´ acteristiques, etat de sante, prise en charge et poids economique sponding to 5% of total public expenditure on healthcare. ´ des personnes diabetiques. Saint-Maurice: Institut de veille san- itaire, 2012. This large economic burden from diabetes highlights the Direct Medical Costs of Type 2 Diabetes in France 219 4. Nathan DM, Buse JB, Davidson MB, Ferrannini E, Holman RR, 14. Caisse Nationale de l’Assurance Maladie des Travailleurs Sal- ´ ´ ´ ´ Sherwin R, et al. Medical management of hyperglycemia in type aries. Rapport au ministre charge de la Securite sociale et au 2 diabetes: a consensus algorithm for the initiation and adjust- Parlement sur l’evolution des charges et des produits de l’As- ment of therapy: a consensus statement of the American Diabetes surance Maladie. 2014. Association and the European Association for the Study of Dia- 15. Quantin C, CNAMTS. Etude des algorithmes de de ´ﬁnition de betes. Diabetes Care. 2009;32(1):193–203. pathologies dans le syste `me national d’information inter-re ´gimes 5. de Lagasnerie G, Aguade ´ AS, Denis P, Fagot-Campagna A, de l’assurance maladie (SNIIRAM). 2014. Gastaldi-Menager C. The economic burden of diabetes to French 16. Jo ¨ nsson B. The economic impact of diabetes. Diabetes Care. national health insurance: a new cost-of-illness method based on 1998;21(Suppl 3):C7–10. a combined medicalized and incremental approach. Eur J Health 17. Williams R. Estimating the national and global costs of diabetes. Econ. doi:10.1007/s10198-017-0873-y (Epub: 11 Feb 2017). Diabetes Res Clin Pract. 2016;119:118–20. 6. Tuppin P, de Roquefeuil L, Weill A, Ricordeau P, Merlie `re Y. 18. Echelle Nationale des cou ˆ ts. ENCC. http://www.atih.fr. French national health insurance information system and the 19. HAS. Guide me ´thodologique. Choix me ´thodologiques pour permanent beneﬁciaries sample. Rev Epidemiol Sante ´ Publique. l’e ´valuation e ´conomique a ` la HAS. Octobre 2011. http://www. 2010;54(4):286–90. has-sante.fr/. 7. Dupouy J, Fournier JP, Jouanjus E, Palmaro A, Poutrain JC, 20. Simon D, Dallongeville J, Charbonnel B, Bureau I, Detournay B, Oustric S, et al. Baclofen for alcohol dependence in France: Leproust S, Levy-Bachelot L, Gourmelen J. Patterns of phar- incidence of treated patients and prescription patterns—a cohort macological treatment in type 2 diabetes in france in 2013 Poster study. Eur Neuropsychopharmacol. 2014;24(2):192–9. PDB14. ISPOR 18th Annual European Congress, 7–11 Novem- 8. Fautrel B, Cukierman G, Joubert JM, Laurendeau C, Gourmelen ber 2015 MiCo—Milano Congressi, Milan, Italy. J, Fagnani F. Healthcare service utilisation costs attributable to 21. French Health Ministry. Direction de la recherche, des e ´tudes, de rheumatoid arthritis in France: analysis of a representative l’e ´valuation et des statistiques. Les de ´penses de sante ´ en 2015. ´ ´ ´ national claims database. Joint Bone Spine. 2016;83(4):461–2. Resultats des comptes de la sante. Edition 2016, Republique 9. Vorilhon C, Chenaf C, Mulliez A, Pereira B, Clerfond G, Authier Franc ¸aise. N, et al. Heart failure prognosis and management in over-80- 22. Ricci P, Chantry M, Detournay B, Poutignat N, Kusnik-Joinville ˆ ´ year-old patients: data from a French national observational ret- O, Raimond V, et al. Cout des soins rembourses par l’Assurance ´ ` rospective cohort. Eur J Clin Pharmacol. 2015;71(2):251–60. maladie aux personnes traitees pour diabete : Etudes Entred 2001 10. Detournay B, Halimi S, Robert J, Deschaseaux C, Dejager S. et 2007. Bull Epidemiol Hebd. 2009;42–43:464–9. Hypoglycemia hospitalization frequency in patients with type 2 23. Robert J, Roudier C, Poutignat N, Fagot-Campagna A, Weill A, diabetes mellitus: a comparison of dipeptidyl peptidase 4 inhi- Rudnichi A, et al. Prise en charge des personnes diabetiques de bitors and insulin secretagogues using the French health insur- type 2 en France en 2007 et tendances par rapport a ` 2001. Bull ance database. Vasc Health Risk Manag. 2015;11:417–25. Epide ´miol Hebd. 2009;42–43:455–60. 11. Inzucchi SE, Bergenstal RM, Buse JB, Diamant M, Ferrannini E, 24. Comptes nationaux de la sante ´. Consommation de soins et biens Nauck M, Peters AL, Tsapas A, Wender R, Matthews DR. me ´dicaux. DREES. 2013. Management of hyperglycaemia in type 2 diabetes: a patient- 25. Detournay B, Simon D, Guillausseau PJ, Joly D, Verges B, Attali centered approach. Position statement of the American Diabetes C, et al. Chronic kidney disease in type 2 diabetes patients in Association (ADA) and the European Association for the Study France: prevalence, inﬂuence of glycaemic control and implica- of Diabetes (EASD). Diabetologia. 2012;55(6):1577–96. tions for the pharmacological management of diabetes. Diabetes 12. Inzucchi SE, Bergenstal RM, Buse JB, Diamant M, Ferrannini E, Metab. 2012;38(2):102–12. Nauck M, Peters AL, Tsapas A, Wender R, Matthews DR. 26. Tsilidis KK, Kasimis JC, Lopez DS, Ntzani EE, Ioannidis JP. Management of hyperglycaemia in type 2 diabetes: a patient- Type 2 diabetes and cancer: umbrella review of meta-analyses of centered approach. Update to a position statement of the Amer- observational studies. BMJ. 2015;350:g7607. ican Diabetes Association and the European Association for the 27. Aldridge M, Kelley AS. The myth regarding the high cost of end- Study of Diabetes. Diabetes Care. 2015;38(1):140–9. of-life care. Am J Public Health. 2015;105(12):2411–5. 13. World Health Organization. International classiﬁcation of dis- 28. American Diabetes Association. Economic costs of diabetes in eases 10th revision. 4th ed. Geneva: WHO; 2010. USA in 2012. Diabetes Care. 2013;36(4):1033–46.
PharmacoEconomics - Open – Springer Journals
Published: Aug 7, 2017
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
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera