- Email: [email protected]
Economic impact of rheumatoid arthritis (RA) biotherapies in France
Joint Bone Spine 77 (2010) 319–324
Original article
Economic impact of rheumatoid arthritis (RA) biotherapies in France Milka Maravic ∗ Département d’information médicale, hôpital Léopold-Bellan, 19–21, rue Vercingétorix, 75674 Paris cedex 14, France
a r t i c l e
i n f o
Article history: Accepted 3 February 2010 Available online 20 May 2010 Keywords: Biotherapy Rheumatoid arthritis Cost
a b s t r a c t Objective: Determining the economic impact of rheumatoid arthritis (RA) biotherapies in France. Method: The number of patients on RA biotherapy in France was estimated from the French national medical information system program (PMSI) database using the 2007 hospital data. The cost of each biotherapy was calculated on a theoretical basis (French national health authority (HAS) recommendations) and on real-life setting, using ‘real-life’ setting data. In order to calculate the economic impact of the biotherapies, the cost of management with each biotherapy was applied to the RA patient population taking into account the market share of each biotherapy. Results: The number of patients with RA estimated was 15,873. Management costs ranged from 11,576 to 21,128 D for the theoretical management scenario and from 6,451 to 19,618 D for the real-life scenario. The overall cost was 222 million D (real-life setting). TNF antagonists (adalimumab, etanercept, infliximab) were prescribed for 82% of the patient population and accounted for 80% of the annual overall cost of theoretical management and 84% of the cost of the real-life setting, respectively. Other biotherapies (abatacept, rituximab) were prescribed for 18% of the patients and accounted for 20 and 16% of the annual overall cost of the theoretical setting and real-life setting. Outpatient biotherapy was prescribed for 61% of the patient population and generated 68 and 71% of the total costs. Conclusion: The data constitutes an initial inventory of the economic impact of RA biotherapies in France. © 2010 Société franc¸aise de rhumatologie. Published by Elsevier Masson SAS. All rights reserved.
1. Introduction Since 2004, major changes in the organization and financing of the French healthcare system have occurred: implementation of healthcare trajectories [1]; use of the medical procedure classification for coding and invoicing outpatient and hospital procedures [2]; and hospital funding based on the procedure tariff system (T2A) [3]. French healthcare establishments have undergone a veritable metamorphosis with the T2A system [3], the main principle of which is the introduction of procedure-based funding for outpatient and inpatient consultation. For hospitalization, the cost evaluation depends on numerous parameters including duration of hospitalization, and drug and/or medical device use. Certain expensive drugs and medical devices have been included in a costly drug and device list and are funded in addition to hospitalization funding. The list thus enables facilitated access to and funding of innovative and expensive therapies. Moreover, healthcare establishments have concluded a contract known as the ‘rational use contract’ with their regional hospital authority. The contracts have
∗ Tel.: +01 40 48 68 27; fax: +01 40 48 68 34. E-mail addresses: [email protected], [email protected].
a triple aim: use of treatments on the costly drug list in compliance with the reference systems; ensuring traceable and secure information from nominal prescription through administration and including dispensing; and expenditure monitoring [4]. Recently, the 2009 French national health insurance financing act and a ministerial circular set a maximum limit for the growth of drug and medical device expenditures for products on the costly drug and device list for 2009. Concomitantly, healthcare establishment monitoring is implemented and may give rise to sanctions in the event of an overshoot due to prescribing practices that are not medically justified [5–6]. All the biotherapies used to treat rheumatoid arthritis (RA) are included in the costly drug list. Some, however, due to their prescription status, are only administered in hospitals (hospital-use only) while others require a hospital prescription but are available in open-care pharmacies [7]. Since 2009, it has been theoretically possible to administer initial hospital prescription treatments in a hospital to outpatients. In this case, the hospital is reimbursed for the drug after reporting the administration [5]. In France, few data on the economic impact of RA biotherapies are available. It was thus considered important in the current context to evaluate the annual economic impact of RA biotherapies on the basis of the information available.
1297-319X/$ – see front matter © 2010 Société franc¸aise de rhumatologie. Published by Elsevier Masson SAS. All rights reserved. doi:10.1016/j.jbspin.2010.04.001
320
M. Maravic / Joint Bone Spine 77 (2010) 319–324
2. Method
2.3. Study point of view and costs considered
2.1. Definition of the rheumatoid arthritis patient population likely to receive biotherapy
The study point of view is that of the National Health Insurance Organization. Only the following direct medical costs were taken into account:
In order to estimate the target population, it was assumed that all hospitalized RA cases were severe RA or RA requiring disease-modifying antirheumatic drug (DMARD) treatment. Since only hospitalization data were available, those of the national public and private database of the medical information system program (PMSI) were used. The RA patient population was defined for 2007 from that database using the following criteria:
• entry of one of the diagnostic codes as the principal or related diagnosis (international classification of diseases (ICD) code M05, and M06.0, M06.8, M06.9) [8]; • patient age greater than or equal to 18 years; • treatment in a medical unit (hospitalization classification in a disease related-group, DRG); • hospitalization coverage by the national health insurance organization (enabling hospitalization linking and patient population determination).
The resulting RA patient population was considered to constitute the biotherapy target population since no outpatient data were available.
2.2. Biotherapies considered Several biotherapies are indicated for RA in France. They have all been taken into account except anakinra (Kineret® ), which is rarely prescribed France. The therapies differ in terms of administration conditions and frequency, dosage, which may or may not be based on patient body weight, indication for the type of RA treated and inclusion on the hospital-only treatment list (Table 1). First-line biotherapies are considered to be TNF antagonist biotherapies while second-line biotherapies consist in the other products. The National Health Authority (HAS) recommends use of abatacept or rituximab providing there are no contra-indications after failure of one or more antagonist biotherapies [9], in the event of active disease, and for patients having experiencing adverse events on other therapies.
• drug cost taking into account the dose and dosing frequency; • management cost: ◦ hospitalization (version 11 of the classification of homogeneous patient groups applicable since 1st March 2009, [12,13]): the tariff for day hospitalization for drugs on the hospital-only list was used. This is the national public tariff for DRG 28Z17Z (chemotherapy for non-neoplastic diseases, in sessions). This is homogeneous hospitalization group number 9616 (290.04 D ), ◦ outside of hospitalization: the open-care sector 1 specialized rheumatology consultation tariff was used for patients following that healthcare trajectory (28 D ) [14]; • for treatments administered subcutaneously, the cost of a subcutaneous injection administered by a state-registered nurse (SRN) was taken into account, or not taken into account, depending on the management contexts defined below (3 D ) [14]. The study was not able to take into account the cost of home-calls by SRN since no data are available. Costs considered the same for each of the biotherapies, such as concomitant treatment with methotrexate and other medications (corticosteroids, analgesics, anti-inflammatories) and laboratory monitoring were not taken into account. Given the study model, no direct non-medical costs such as transport expenses were taken into account. It is to be noted that the context considered is that of a patient with RA requiring biotherapy for a disease classified as a ‘long-term disease’ with 100% health insurance coverage. The costs have been calculated in 2009 Euros. 2.4. Definition of the biotherapy management trajectory and method of calculating the annual cost for a patient on treatment A ‘typical’ RA patient weighing 66 kg (mean weight of patients with RA) on biotherapy (excluding anakinra) was considered. The ‘typical’ patient has no contra-indication to any of the biotherapies. The patient receives treatment for 1 year, responds to treatment and tolerates treatment well. In that clinical context, two types of management are considered for each biotherapy:
Table 1 Biotherapies [9]. Biotherapy Abatacept (Orencia® ) Adalimumab (Humira® ) Etanercept (Enbrel® ) Infliximab (Remicade® ) Rituximab (Mabthera® )
Presentation–Route price (2008, incl. VAT)a 250-mg vial–IV 406.358 D 40-mg syringe–SC 1,137.06 D for 2 syringes 50-mg syringe–SCc 1,117.30 D for 4 syringes 100-mg vial–IV 569.138 D 500-mg vial–IV 1,439.61 D
Dosage as per SPC Dosing frequency 2 to 4 vials per injection W0, W2, W4, then monthly Every 2 weeks 50 mg/week 3 mg/kg per injection W0, W2, W6, then every 2 months 1st course: 1 g at W0 and 1 g at W2 2nd course: 1 g at an interval of 15 days ± after 6 to 12 months
Prescription type Administration Hospital only In hospital Initial hospital prescription Outside of hospital Initial hospital prescription Outside of hospitalb Hospital only In hospital Hospital prescription, sale by hospital pharmacy to patient In hospital*
IV: intravenous; SPC: summary of product characteristics; W: week; SC: subcutaneous; Route: administration route . a For treatments administered during hospitalization, the price is the current responsibility tariff published in the French Official Journal and includes VAT (2.1%). For treatments available from open-care pharmacies, the price is the retail price including VAT published in the Official Journal as at 30th April 2009 [10,11]. b These treatments may be administered during hospitalization but unlike the other treatments, they are available from open-care pharmacies. Day hospitalization of the patient simply for administration of these treatments is not justified in France in the regulations [12]. c This treatment is also available as a 25-mg strength administered subcutaneously twice weekly. A single weekly administration has been preferred. * This treatment is administered during hospitalization and is included in the costly drug list.
M. Maravic / Joint Bone Spine 77 (2010) 319–324
321
Table 2 Management trajectories. Biotherapy Abatacept
Adalimumab
Management recommended in the SPC and reference systems (HAS/AFSSAPS) 3 vials per administration Hospital follow-up determined by injection frequency (n = 14 for the 1st year of treatment) Injection every 15 days (n = 26) Monthly follow-up outside of hospitalization (n = 12)
Etanercept
Weekly injection (n = 52) Monthly (n = 12)
Infliximab
2 vials per administration Hospital follow-up frequency determined by the injections (n = 8 for the 1st year of treatment) in alternation with follow-up outside of hospitalization (n = 5) 2 courses Hospital follow-up (n = 4) alternating with follow-up outside of hospitalization (n = 10)
Rituximab
‘Real-life’ setting management The same as for the ideal management but with 13 injections in the second year of treatment Injection of 40 mg every 15 days (n = 26) Monthly follow-up for the 1st quarter, then quarterly follow-up (n = 6) ± injection by SRN (17% of patientsa ) Weekly injection (n = 52) Monthly follow-up the 1st quarter, then quarterly follow-up outside of hospitalization (n = 6) ± injection by SRN (31% of patientsa ) Monthly dose of 150.2 mg (calculated as a function of dose and not as the number of vials)a 6.5 injections No follow-up outside of hospitalization 1st course (1 g + 1 g) for 100% of patients 2nd course (1 g + 1 g) for 31% of patients over a calendar yearb Quarterly follow-up outside of hospitalization (n = 4)
Sensitivity analysis on the basis of the real-life setting* Dose injected–13 injections lh: 2 vials per injection hh: 4 vials per injection SC injection by patient or SRN lh: SC injection by patient hh: SC injection by SRN SC injection by patient or SRN lh: SC injection by patient hh: SC injection by SRN
Treatment dosage lh: 3.9 mg/kg per injection (257.4 mg) hh: 7.5 mg/kg per injection (495 mg)
Number of courses lh: 1 course hh: 2 courses
SC: subcutaneous; SRN: state-registered nurse, injection tariff including transport: D 3. a Results of the PANEL A + A inquiry, March-April 2008. b Auto-immunity and rituximab RA (AIR-PR) registry [15]. * Low hypothesis (lh), high hypothesis (hh).
• theoretical RA management as defined by the HAS recommendations [9]; • ‘real-life’ setting management as per the dosage and dosing frequency used in everyday practice. With regard to theoretical management, monthly follow-up is recommended until the disease is brought under control by DMARD, when follow-up becomes quarterly. Since treatment monitoring is monthly, particularly with regard to the safety aspects, monthly follow-up by a rheumatologist is considered appropriate. The rheumatologist may be an open-care or hospital rheumatologist, or both, depending on the method of treatment administration. It is true that the recommendations introduce the concept of follow-up jointly by the reported attending physician, most frequently a general practitioner, and the other healthcare professionals involved, including the rheumatologist. This study is based on follow-up by the rheumatologist. The real-life setting data derives from two sources: • data from the A+A panel: the panel specializes in RA. The A+A Company panel consists of anonymous multi-subscribers (most of the people working in the sector are subscribers) and has operated for 6 years with two waves per year. The panel is cited by HAS in certain transparency opinions (transparency opinion on etanercept, 16th July 2008). In order to ensure that the panel is representative, it is constituted by the quota method using the database compiled by the Directorate of Research, Studies, Evaluation and Statistics and the National Council of the Order of Physicians (CNOM) (geographic distribution, age, gender distribution, type of establishment in which the physician practices, practice site: hospital, open-care or mixed). For each wave, 250 rheumatologists contribute, i.e. a 10% sampling rate (approximately 2500 rheumatologists in France) and include all patients with RA over a 2-week period, with a cut-off of 20 patients per physician. In all, 2000 patients, of whom 700 on biotherapy were analyzed; • the auto-immune and rituximab RA (AIR-PR) registry, which monitors a cohort of RA patients on rituximab treatment [16].
The primary objective of the registry is evaluation of the safety profile of rituximab. The secondary objective is evaluation of the efficacy of rituximab in the patients on treatment. The various modalities of retreatment in real-life settings are also studied. The cost of each of the treatments was calculated over 1 year and per patient as a function of the treatment modalities and hospital or open-care follow-up. The treatment share of each management method was determined. Table 2 shows the disease management trajectory over 1 year for each biotherapy. 2.5. Application of the management context to the rheumatoid arthritis patient population The following are available: the annual cost of each of the management methods (theoretical and real-life settings) by biotherapy and by patient; the distribution of biotherapy market share in the 3rd quarter 2008 (source: GFK Performance Tracker Q3 2008, a quarterly international study). Contributing rheumatologists were sampled using several criteria. The rheumatologists were to have 3 years of experience in the treatment of RA and there were to be at least 50 patients with RA among the rheumatologist’s patients. The rheumatologists were also to have experience with biotherapies and to have prescribed biotherapies to at least eight patients in the 3 months preceding the quarterly wave in order to participate. For each wave, 80 rheumatologists take part. Half of the participants are hospital rheumatologists and half mixed-practice rheumatologists. The participants include the ten patients who have most recently consulted, generating a total of 800 patients per wave. For each wave, new rheumatologists participate. The data used in this study were generated by the August 2008 wave. The patient distribution by biotherapy was as follows: 37% on etanercept, 30% on adalimumab, 15% on infliximab, 12% on rituximab and 6% on abatacept. To determine the overall cost of biotherapy in France, the cost of management by biotherapy and by patient was multiplied by the previously described distribution in the RA patient population defined above.
322
M. Maravic / Joint Bone Spine 77 (2010) 319–324
Table 3 Annual management cost by patient healthcare trajectory. Biotherapy
Abatacept
Management recommended in the SPC or reference systems (HAS/AFSSAPS) Annual cost (impact of acquisition treatment cost on overall cost) 21,128 D (81%)
Adalimumaba
15,118 D (98%)
Etanercepta
14,861 D (98%)
Infliximab
11,567 D (79%)
Rituximaba
12,957 D (89%)
‘Real-life’ setting management Annual cost (impact of acquisition treatment cost on overall cost) 19,618 D (81%) 14,950 D 15,464 D 14,693 D 14,849 D 12,144 D
if patient injection (99%) if SRN injection (98%) if patient injection (99%) if SRN injection (98%) (84%)
6,451 D if 1st course (89%) 12,789 D if 2nd course (90%)
Sensitivity analysis based on real-life setting* Annual cost (impact of acquisition treatment cost on overall cost) lh: 14,336 D (74%) hh: 24,901 D (85%) lh: 14,950 D (99 %) hh: 15,464 D (98 %) lh: 14,693 D (99%) hh: 14,849 D (98%) lh: 11,408 D (83%) hh: 20,198 D (91%) lh: 6,451 D (89%) hh: 12,789 D (90%)
SC: subcutaneous; SRN: state-registered nurse, injection tariff excluding travel: 3;◦ ◦ Results of the PANEL A+A inquiry, March-April 2008;◦ ◦ ◦ Auto-immunity and rituximab RA registry [16]. Refer to Table 2 for the percentage patients receiving injections by an SRN (adalimumab, etanercept) or receiving a second course of treatment (rituximab). a The annual cost calculation in the ‘real-life’ setting and in the sensitivity analysis are the same for the three biotherapies since the calculation was done for a given patient. * Low hypothesis (lh), high hypothesis (hh).
2.6. Sensitivity analysis The sensitivity analysis addressed the ‘real-life’ setting data using a low and a high hypothesis (Table 2). The aim was to evaluate the impact of patient body weight variation, which impacts the dosage prescribed for abatacept and infliximab, subcutaneous injection administration for adalimumab and etanercept, and the potential second course of rituximab treatment. 3. Results
the percentage for which a second course of treatment is administered (rituximab). Irrespective of the management trajectory considered, it will be observed that the biotherapy acquisition cost constitutes the greater part of the annual cost (79 to 98% of the annual cost, depending on the biotherapy, in the theoretical management setting). The variation in the treatment-related cost is greater if dosage varies as a function of patient body weight (abatacept, infliximab) or dosing frequency (rituximab). For biotherapies administered by the subcutaneous route (adalimumab, etanercept), injection or non-injection by a SRN has little impact on the overall cost.
3.1. Definition of the patient population In 2007, there were 23.4 million short-stay hospitalizations in the public and private sector in France. In all, 79,560 hospitalizations (0.34%) were hospitalizations in which RA was reported as the primary or linked diagnosis. Given the selection criteria defined above, the RA patient population was 15,873 (1 to 33 hospitalizations per patient; 39,549 hospitalizations in all). 3.2. Annual cost per patient by healthcare trajectory Table 3 shows the annual cost of each biotherapy by the defined healthcare trajectory. The data are presented as the cost of management by trajectory and indicate the percentage biotherapy acquisition cost. With regard to adalimumab, etanercept and rituximab, the calculations relate to the real-life setting management data and the sensitivity analyses are the same for a given patient. On the RA patient population scale, the cost of biotherapy management is modulated by application of the percentage of patients for which injection is administered by a nurse (adalimumab, etanercept) and
3.3. Biotherapy distribution in the patient population and calculation of the corresponding economic impact Table 4 shows the distribution of the various biotherapies in the RA patient population and the annual cost as a function of the type of management. As observed with the evaluation of the theoretical setting management trajectory annual cost, there is little variation between the theoretical management setting, real-life setting, and simulation scenarios for treatments independent of patient weight, dosage used and repeat-treatment, i.e. subcutaneous biotherapies. The impact of body weight varies as a function of the treatment considered. With abatecept, injection of three vials for a body weight fluctuating between 60 and 100 kg has less impact than modulation in terms of body weight and dose for infliximab (maximum dose: 7.5 mg/kg per administration). The results may also be ranked in terms of 1st- and 2nd-line treatment and as a function of the sector whose expenditures are impacted (open-care or hospital care). First-line biotherapies (adalimumab, etanercept and infliximab) account for 82% of the treated target population. In the context of theoretical or real-life
Table 4 Distribution of the patient population (number, percentage) and total annual cost of management (2009 Euros, percentage) by biotherapy for the entire patient population. Patient population as number of patients (%)a
Theoretical management cost, D (%)
Real-life setting cost, Euros (%)
Low hypothesis cost, D * (%)
High hypothesis cost, D * (%)
Abatacept Adalimumab Etanercept Infliximab Rituximab
952 (6) 4,762 (30) 5,873 (37) 2,381 (15) 1,905 (12)
20,113,856 (9) 71,991,916 (31) 87,278,653 (37) 27,541,027 (12) 24,683,085 (11)
18,676,336 (9) 71,252,500 (32) 86,575,567 (39) 28,914,864 (13) 16,029,434 (7)
13,647,872 (6) 71,191,900 (34) 86,291,549 (41) 27,162,448 (13) 12,289,155 (6)
23,705,752 (9) 73,639,568 (29) 87,208,177 (34) 48,091,438 (19) 24,363,045 (9)
Total
15,873 (100)
231,608,537 (100)
221,448,701 (100)
210,582,924 (100)
257,007,980 (100)
a *
The RA patient population was determined from the PMSI data. Based on the real-life setting data. The patient distribution by biotherapy is based on the market share data for the 3rd quarter 2008.
M. Maravic / Joint Bone Spine 77 (2010) 319–324
setting follow-up, 1st-line biotherapies generate 80 and 84% of management costs. Second-line biotherapies (abatacept and rituximab) are prescribed for the other 18% of the target population and generate 20 and 16% of the annual costs in the theoretical and reallife management settings, respectively. Open-care biotherapies are administered to 61% of the patient population and generate 68 and 71% of the total cost for the theoretical and real-life management settings, respectively. 4. Discussion and limitations of the study In the present context of reforms in France, it is important to determine the RA target population receiving biotherapy and the national cost generated, particularly if biotherapy use is to be controlled. Data interpretation is also to take into account the status of the drugs as defined by French regulations [5]. Both hospital-prescribed drugs (adalimumab, etanercept and rituximab) and hospital only drugs (abatacept and infliximab) were addressed. Biotherapies administered by the subcutaneous route impact opencare expenditures, while the other biotherapies impact hospital expenditures. It is also important to determine the biotherapy target population. Few epidemiological data on RA in France are available since all the data are not accessible, particularly those of the National Health Insurance Organization. While the National Health Insurance Organization publishes data, they are not sufficiently precise. The data does show that, at the end of 2007, there were 150,032 No. 22 long-term diseases (LTD) linked to RA [17]. Outside of the issues that may be raised by verification of diagnostic validity by comparing the LTD declaration with the patient’s file, the data are not informative with regard to the proportion of patients on biotherapy. A recent study of RA-related LTD in the Nord-Pas-deCalais region showed that 7,178 LTD were reported in 2005 [18] but 40% of the patients had no DMARD treatment. This point highlights to a probable overestimation of the number of cases of RA. Among the patients on DMARD treatment (the remaining 60%), biotherapies accounted for 45% of the DMARD treatments prescribed alone or in combination with more conventional DMARD treatments. Etanercept is the most widely prescribed biotherapy. Data on 2007 open-care drug consumption are available. The data shows that etanercept and adalimumab rank 7th (134.4 million D ) and 23rd (87.4 million D ), respectively, in terms of prescription and reimbursement, but the distribution by disease is not indicated [19]. The hospitalization data gives an idea of the number of hospitalizations for RA and theoretically enables targeting those associated with the costly drug list (which is the case for abatacept, infliximab and rituximab). While adalimumab and etanercept are also on the list, in practice they should only represent a minimal proportion of hospitalizations simply for drug administration. Information and particularly the links between hospitalization and drug consumption became mandatory in 2007 and are to be reported to a specific database [20]. With regard to the modalities for hospital selection, it is to be noted, as was the case for LTD, that it is not possible to compare the reported data with the patient files due to the anonymous nature of medical records. Moreover, considering that an RA case hospitalized at least once is a case liable to benefit from biotherapy is highly debatable. There is no evidence that all patients for whom biotherapy is discussed are hospitalized patients, particularly in the case of hospital prescribed biotherapies. A hospital prescription for an outpatient enables the biotherapy considered to be administered in an outpatient setting. Moreover, the diagnostic codes used for PMSI coding do not enable definition of the seriousness of RA. A patient with RA may also be hospitalized for another reason: confirmation of the diagnosis and initiation of conventional DMARD treatment, implementation of radioisotope synoviorthesis, etc. In addition, the present study addressed patients who respond to treatment and are
323
free from adverse events. This may be remote from real-life settings given that patients may not respond and thus require a change in DMARD. The annual cost estimate only takes into account the direct medical costs and those liable to vary between biotherapies. Given that the context is a theoretical management context, it was important to develop that context using real-life setting data modulating the information, such as body weight and/or dose (abatacept and infliximab), SRN injection administration or not (adalimumab and etanercept) and repeat-treatment (rituximab). In the context of theoretical management, depending on the product, the treatment cost is equivalent to 79 to 98% of the annual cost. The annual cost varies, depending on the product considered, from 11,567 to 21,128 D . Those data differs little from the real-life setting data taken into account in the study. In the literature, the treatment cost is identified as a major component of direct medical costs [21–23]. Although the context is theoretical, the treatment cost does not differ from those reported by Jönsson et al. using 2006 sales data (irrespective of the indication for biotherapy): annual treatment cost of 15,000 to 19,000 D for abatacept; 10,000 to 15,000 D for adalimumab; 9,000 to 15,000 D for etanercept; 10,000 to 18,000 D for infliximab; and 7,000 to 10,000 D for rituximab [24]. Biotherapy cost distribution is to be interpreted as a function of the study context, year of the study and country in which it was conducted. US studies have evaluated biotherapy use on the basis of a preference questionnaire [25], database [26] and prospective study [27]. Irrespective of whether the evaluation is at a time t [25] or over several years [26,27], the results vary depending on the methodology. Depending on the study, the results are in favour of etanercept [26–27] or infliximab [27]. However, the studies are of limited pertinence in that they only consider TNF antagonist biotherapies. In conclusion, it would seem important for every country to have pertinent information on the epidemiology of patients with diseases requiring the use of expensive treatments and, more specifically, patients liable to have need of them since they meet the criteria defined by the recommendations. This study attempted an optimal estimation of the impact of RA biotherapies on the basis of the information available in France. However, the findings are only an initial approach. In a context in which healthcare expenditure regulation is evolving, in particular with the set-up of expenditure ceilings for costly drugs and medical devices, the information available on patient trajectories and the impact of the necessary drugs or devices by disease will enable decision-makers to adjust access to certain therapies as a function of the population’ real needs. Conflict of interest None of the authors has any conflict of interest to declare. References [1] Arrêté 3 février 2005 portant approbation de la convention nationale des médecins généralistes et des médecins spécialistes. Journal Officiel 2005. [2] La classification Commune des actes médicaux. Site Ameli : http://www.ameli. fr/professionnels-de-sante/medecins/exercer-au-quotidien/nomenclatureset-codage/codage-des-actes-medicaux-ccam.php. [3] Maravic M, Le Bihan C, Landais P. La tarification à l’activité : définition, modalités et tarifs en rhumatologie. Rev Rhum 2004;71:642–4. [4] Décret n◦ 2008-1121 du 31 octobre 2008 relatif au contrat de bon usage des médicaments et des produits et prestations mentionné à l’article L. 162-22-7 du code de la sécurité sociale. Journal Officiel 2008. [5] Décret no 2004-546 du 15 juin 2004 relatif aux catégories de médicaments à prescription restreinte et à la vente de médicaments au public par certains établissements de santé et modifiant le code de la santé publique et le code de la sécurité sociale (deuxième partie : décrets en conseil d’État). Journal Officiel 2004. [6] Loi n◦ 2008-1330 du 17 décembre 2008 de financement de la sécurité sociale pour 2009 (1). Journal Officiel 2008 http://www.legifrance.gouv.fr/affichTexte.
324
[7]
[8] [9]
[10]
[11]
[12]
[13]
[14]
[15]
M. Maravic / Joint Bone Spine 77 (2010) 319–324 do;jsessionid=E4F710107B3AF52A2CB42D2373DDA587.tpdjo10v 1?cidTexte= LEGITEXT000019945312&dateTexte=20090329. Circulaire N◦ DHOS/F2/F3/F1/DSS/1A/2009/78 du 17 mars 2009 relative à la campagne tarifaire 2009 des établissements de santé et son annexe IV : évolution du modèle de financement des activités de court séjour. Ministère de la Santé et des Sports. Classification statistique internationale des maladies et des problèmes de santé connexes. 10e révision. Organisation Mondiale de la santé 1993. Site Haute Autorité de la santé : polyarthrite rhumatoïde : diagnostic et prise en charge initiale, 2007 : http://www.has-sante.fr/portail/jcms/c 606479/ polyarthrite-rhumatoide-diagnostic-et-prise-en-charge-initiale. Polyarthrite Rhumatoïde : prise en charge en phase d’état, 2007 : http://www.has-sante.fr/ portail/jcms/c 606483/polyarthrite-rhumatoide-prise-en-charge-en-phasedetat. Site AFSSAPS : tarification à l’activité (T2A) et médicaments hors groupe homogène de séjours : http://www.afssaps.fr/Dossiers-thematiques/ Tarification-a-l-activite-T2A-medicaments/Accueil-T2A/%28offset%29/0. Site Assurance Maladie : base des médicaments et informations tarifaires : http://www.codage.ext.cnamts.fr/codif/bdm it/index tele ucd.php?p site= AMELI. Circulaire n◦ DHOS/F1/MTAA/2006/376 du 31 août 2006 relative aux conditions de facturation d’un GHS pour les prises en charge hospitalières en zone de surveillance de très courte durée ainsi que pour les prises en charge de moins d’une journée. Ministère de la Santé et des Solidarités. Arrêté du 19 février 2009 relatif à la classification et à la prise en charge des prestations d’hospitalisation pour les activités de médecine, chirurgie, obstétrique et odontologie et pris en application de l’article L. 162-22-6 du code de la sécurité sociale. Journal Officiel 2009. Arrêté du 27 février 2009 fixant pour l’année 2009 les ressources d’assurance maladie des établissements de santé exerc¸ant une activité de médecine, chirurgie, obstétrique et odontologie. Journal Officiel 2009. Site Assurance Maladie : Nomenclature générale des actes professionnels http://www.ameli.fr/professionnels-de-sante/infirmiers/exercer-auquotidien/n.g.a.p.-et-l.p.p./la-n.g.a.p.php.
[16] Registre AIR-PR http://www.rhumatologie.asso.fr/12-Recherche/registres. asp?intSM=SM3&strLien=Lien30, http://www.abstracts2view.com/eular/view. php?nu=EULAR08L THU018. [17] Assurance Maladie. Les personnes en affection de longue durée au 31 décembre 2007. Points de repère n◦ 20, Novembre 2008. [18] Biver E, Beague V, Verloop D, et al. Low and stable prevalence of rheumatoid arthritis in northern France. Joint Bone Spine 2009;76(5):497–500. [19] Assurance Maladie.Médicaments remboursés : analyse de la consommation en 2007. 1er août 2008. [20] DHOS: précisions complémentaires concernant les fichiers FichComp et FichSup que doivent transmettre les établissements antérieurement sous dotation globale en 2007 : http://www.atih.sante.fr/?id=0002200074FF. [21] Maravic M. Economic burden of rheumatoid arthritis in France. Exp Rev Pharmacoeconomics Outcome Res 2006;6:25–32. [22] Fautrel B, Woronoff-Lemsi MC, Etghen M, et al. Influence des pratiques médicales sur les coûts de traitement de la polyarthrite rhumatoïde par biothérapie anti-TNF␣ en France. Rev Rhum 2005;72:1306–13. [23] Kobelt G, Woronoff AS, Richard B, et al. Disease status, costs and quality of life of patients with rheumatoid arthritis in France: the ECO-PR Study. Joint Bone Spine 2008;75(4):408–15. [24] Jönsson B, Kobelt G, Smolen J. The burden of rheumatoid arthritis and access to treatment: uptake of new therapies. Eur J Health Econ 2008;8(Suppl2): S61–86. [25] Kamal KM, Madhavan SS, Allen Hornsby JAA, et al. Use of tumor necrosis factor inhibitors in rheumatoid arthritis: a national survey of practicing United States rheumatologists. Joint Bone Spine 2006;73:718–24. [26] Yazici Y, Shi N, John A. Utilization of biological agents in rheumatoid arthritis in the United States. Analysis of prescribing patterns in 16,752 newly diagnosed patients and patients new to biologic therapy. NYU Hosp Jt Dis 2008;66: 77–85. [27] Gibofsky A, Palmer WR, Goldman JA, et al. Real-world utilization of DMARDs and biologics in rheumatoid arthritis: the RADIUS (rheumatoid arthritis disease-modifying anti-rheumatic drug intervention and utilization study) study. Cur Med Res Opin 2006;22:169–83.
Original article
Economic impact of rheumatoid arthritis (RA) biotherapies in France Milka Maravic ∗ Département d’information médicale, hôpital Léopold-Bellan, 19–21, rue Vercingétorix, 75674 Paris cedex 14, France
a r t i c l e
i n f o
Article history: Accepted 3 February 2010 Available online 20 May 2010 Keywords: Biotherapy Rheumatoid arthritis Cost
a b s t r a c t Objective: Determining the economic impact of rheumatoid arthritis (RA) biotherapies in France. Method: The number of patients on RA biotherapy in France was estimated from the French national medical information system program (PMSI) database using the 2007 hospital data. The cost of each biotherapy was calculated on a theoretical basis (French national health authority (HAS) recommendations) and on real-life setting, using ‘real-life’ setting data. In order to calculate the economic impact of the biotherapies, the cost of management with each biotherapy was applied to the RA patient population taking into account the market share of each biotherapy. Results: The number of patients with RA estimated was 15,873. Management costs ranged from 11,576 to 21,128 D for the theoretical management scenario and from 6,451 to 19,618 D for the real-life scenario. The overall cost was 222 million D (real-life setting). TNF antagonists (adalimumab, etanercept, infliximab) were prescribed for 82% of the patient population and accounted for 80% of the annual overall cost of theoretical management and 84% of the cost of the real-life setting, respectively. Other biotherapies (abatacept, rituximab) were prescribed for 18% of the patients and accounted for 20 and 16% of the annual overall cost of the theoretical setting and real-life setting. Outpatient biotherapy was prescribed for 61% of the patient population and generated 68 and 71% of the total costs. Conclusion: The data constitutes an initial inventory of the economic impact of RA biotherapies in France. © 2010 Société franc¸aise de rhumatologie. Published by Elsevier Masson SAS. All rights reserved.
1. Introduction Since 2004, major changes in the organization and financing of the French healthcare system have occurred: implementation of healthcare trajectories [1]; use of the medical procedure classification for coding and invoicing outpatient and hospital procedures [2]; and hospital funding based on the procedure tariff system (T2A) [3]. French healthcare establishments have undergone a veritable metamorphosis with the T2A system [3], the main principle of which is the introduction of procedure-based funding for outpatient and inpatient consultation. For hospitalization, the cost evaluation depends on numerous parameters including duration of hospitalization, and drug and/or medical device use. Certain expensive drugs and medical devices have been included in a costly drug and device list and are funded in addition to hospitalization funding. The list thus enables facilitated access to and funding of innovative and expensive therapies. Moreover, healthcare establishments have concluded a contract known as the ‘rational use contract’ with their regional hospital authority. The contracts have
∗ Tel.: +01 40 48 68 27; fax: +01 40 48 68 34. E-mail addresses: [email protected], [email protected].
a triple aim: use of treatments on the costly drug list in compliance with the reference systems; ensuring traceable and secure information from nominal prescription through administration and including dispensing; and expenditure monitoring [4]. Recently, the 2009 French national health insurance financing act and a ministerial circular set a maximum limit for the growth of drug and medical device expenditures for products on the costly drug and device list for 2009. Concomitantly, healthcare establishment monitoring is implemented and may give rise to sanctions in the event of an overshoot due to prescribing practices that are not medically justified [5–6]. All the biotherapies used to treat rheumatoid arthritis (RA) are included in the costly drug list. Some, however, due to their prescription status, are only administered in hospitals (hospital-use only) while others require a hospital prescription but are available in open-care pharmacies [7]. Since 2009, it has been theoretically possible to administer initial hospital prescription treatments in a hospital to outpatients. In this case, the hospital is reimbursed for the drug after reporting the administration [5]. In France, few data on the economic impact of RA biotherapies are available. It was thus considered important in the current context to evaluate the annual economic impact of RA biotherapies on the basis of the information available.
1297-319X/$ – see front matter © 2010 Société franc¸aise de rhumatologie. Published by Elsevier Masson SAS. All rights reserved. doi:10.1016/j.jbspin.2010.04.001
320
M. Maravic / Joint Bone Spine 77 (2010) 319–324
2. Method
2.3. Study point of view and costs considered
2.1. Definition of the rheumatoid arthritis patient population likely to receive biotherapy
The study point of view is that of the National Health Insurance Organization. Only the following direct medical costs were taken into account:
In order to estimate the target population, it was assumed that all hospitalized RA cases were severe RA or RA requiring disease-modifying antirheumatic drug (DMARD) treatment. Since only hospitalization data were available, those of the national public and private database of the medical information system program (PMSI) were used. The RA patient population was defined for 2007 from that database using the following criteria:
• entry of one of the diagnostic codes as the principal or related diagnosis (international classification of diseases (ICD) code M05, and M06.0, M06.8, M06.9) [8]; • patient age greater than or equal to 18 years; • treatment in a medical unit (hospitalization classification in a disease related-group, DRG); • hospitalization coverage by the national health insurance organization (enabling hospitalization linking and patient population determination).
The resulting RA patient population was considered to constitute the biotherapy target population since no outpatient data were available.
2.2. Biotherapies considered Several biotherapies are indicated for RA in France. They have all been taken into account except anakinra (Kineret® ), which is rarely prescribed France. The therapies differ in terms of administration conditions and frequency, dosage, which may or may not be based on patient body weight, indication for the type of RA treated and inclusion on the hospital-only treatment list (Table 1). First-line biotherapies are considered to be TNF antagonist biotherapies while second-line biotherapies consist in the other products. The National Health Authority (HAS) recommends use of abatacept or rituximab providing there are no contra-indications after failure of one or more antagonist biotherapies [9], in the event of active disease, and for patients having experiencing adverse events on other therapies.
• drug cost taking into account the dose and dosing frequency; • management cost: ◦ hospitalization (version 11 of the classification of homogeneous patient groups applicable since 1st March 2009, [12,13]): the tariff for day hospitalization for drugs on the hospital-only list was used. This is the national public tariff for DRG 28Z17Z (chemotherapy for non-neoplastic diseases, in sessions). This is homogeneous hospitalization group number 9616 (290.04 D ), ◦ outside of hospitalization: the open-care sector 1 specialized rheumatology consultation tariff was used for patients following that healthcare trajectory (28 D ) [14]; • for treatments administered subcutaneously, the cost of a subcutaneous injection administered by a state-registered nurse (SRN) was taken into account, or not taken into account, depending on the management contexts defined below (3 D ) [14]. The study was not able to take into account the cost of home-calls by SRN since no data are available. Costs considered the same for each of the biotherapies, such as concomitant treatment with methotrexate and other medications (corticosteroids, analgesics, anti-inflammatories) and laboratory monitoring were not taken into account. Given the study model, no direct non-medical costs such as transport expenses were taken into account. It is to be noted that the context considered is that of a patient with RA requiring biotherapy for a disease classified as a ‘long-term disease’ with 100% health insurance coverage. The costs have been calculated in 2009 Euros. 2.4. Definition of the biotherapy management trajectory and method of calculating the annual cost for a patient on treatment A ‘typical’ RA patient weighing 66 kg (mean weight of patients with RA) on biotherapy (excluding anakinra) was considered. The ‘typical’ patient has no contra-indication to any of the biotherapies. The patient receives treatment for 1 year, responds to treatment and tolerates treatment well. In that clinical context, two types of management are considered for each biotherapy:
Table 1 Biotherapies [9]. Biotherapy Abatacept (Orencia® ) Adalimumab (Humira® ) Etanercept (Enbrel® ) Infliximab (Remicade® ) Rituximab (Mabthera® )
Presentation–Route price (2008, incl. VAT)a 250-mg vial–IV 406.358 D 40-mg syringe–SC 1,137.06 D for 2 syringes 50-mg syringe–SCc 1,117.30 D for 4 syringes 100-mg vial–IV 569.138 D 500-mg vial–IV 1,439.61 D
Dosage as per SPC Dosing frequency 2 to 4 vials per injection W0, W2, W4, then monthly Every 2 weeks 50 mg/week 3 mg/kg per injection W0, W2, W6, then every 2 months 1st course: 1 g at W0 and 1 g at W2 2nd course: 1 g at an interval of 15 days ± after 6 to 12 months
Prescription type Administration Hospital only In hospital Initial hospital prescription Outside of hospital Initial hospital prescription Outside of hospitalb Hospital only In hospital Hospital prescription, sale by hospital pharmacy to patient In hospital*
IV: intravenous; SPC: summary of product characteristics; W: week; SC: subcutaneous; Route: administration route . a For treatments administered during hospitalization, the price is the current responsibility tariff published in the French Official Journal and includes VAT (2.1%). For treatments available from open-care pharmacies, the price is the retail price including VAT published in the Official Journal as at 30th April 2009 [10,11]. b These treatments may be administered during hospitalization but unlike the other treatments, they are available from open-care pharmacies. Day hospitalization of the patient simply for administration of these treatments is not justified in France in the regulations [12]. c This treatment is also available as a 25-mg strength administered subcutaneously twice weekly. A single weekly administration has been preferred. * This treatment is administered during hospitalization and is included in the costly drug list.
M. Maravic / Joint Bone Spine 77 (2010) 319–324
321
Table 2 Management trajectories. Biotherapy Abatacept
Adalimumab
Management recommended in the SPC and reference systems (HAS/AFSSAPS) 3 vials per administration Hospital follow-up determined by injection frequency (n = 14 for the 1st year of treatment) Injection every 15 days (n = 26) Monthly follow-up outside of hospitalization (n = 12)
Etanercept
Weekly injection (n = 52) Monthly (n = 12)
Infliximab
2 vials per administration Hospital follow-up frequency determined by the injections (n = 8 for the 1st year of treatment) in alternation with follow-up outside of hospitalization (n = 5) 2 courses Hospital follow-up (n = 4) alternating with follow-up outside of hospitalization (n = 10)
Rituximab
‘Real-life’ setting management The same as for the ideal management but with 13 injections in the second year of treatment Injection of 40 mg every 15 days (n = 26) Monthly follow-up for the 1st quarter, then quarterly follow-up (n = 6) ± injection by SRN (17% of patientsa ) Weekly injection (n = 52) Monthly follow-up the 1st quarter, then quarterly follow-up outside of hospitalization (n = 6) ± injection by SRN (31% of patientsa ) Monthly dose of 150.2 mg (calculated as a function of dose and not as the number of vials)a 6.5 injections No follow-up outside of hospitalization 1st course (1 g + 1 g) for 100% of patients 2nd course (1 g + 1 g) for 31% of patients over a calendar yearb Quarterly follow-up outside of hospitalization (n = 4)
Sensitivity analysis on the basis of the real-life setting* Dose injected–13 injections lh: 2 vials per injection hh: 4 vials per injection SC injection by patient or SRN lh: SC injection by patient hh: SC injection by SRN SC injection by patient or SRN lh: SC injection by patient hh: SC injection by SRN
Treatment dosage lh: 3.9 mg/kg per injection (257.4 mg) hh: 7.5 mg/kg per injection (495 mg)
Number of courses lh: 1 course hh: 2 courses
SC: subcutaneous; SRN: state-registered nurse, injection tariff including transport: D 3. a Results of the PANEL A + A inquiry, March-April 2008. b Auto-immunity and rituximab RA (AIR-PR) registry [15]. * Low hypothesis (lh), high hypothesis (hh).
• theoretical RA management as defined by the HAS recommendations [9]; • ‘real-life’ setting management as per the dosage and dosing frequency used in everyday practice. With regard to theoretical management, monthly follow-up is recommended until the disease is brought under control by DMARD, when follow-up becomes quarterly. Since treatment monitoring is monthly, particularly with regard to the safety aspects, monthly follow-up by a rheumatologist is considered appropriate. The rheumatologist may be an open-care or hospital rheumatologist, or both, depending on the method of treatment administration. It is true that the recommendations introduce the concept of follow-up jointly by the reported attending physician, most frequently a general practitioner, and the other healthcare professionals involved, including the rheumatologist. This study is based on follow-up by the rheumatologist. The real-life setting data derives from two sources: • data from the A+A panel: the panel specializes in RA. The A+A Company panel consists of anonymous multi-subscribers (most of the people working in the sector are subscribers) and has operated for 6 years with two waves per year. The panel is cited by HAS in certain transparency opinions (transparency opinion on etanercept, 16th July 2008). In order to ensure that the panel is representative, it is constituted by the quota method using the database compiled by the Directorate of Research, Studies, Evaluation and Statistics and the National Council of the Order of Physicians (CNOM) (geographic distribution, age, gender distribution, type of establishment in which the physician practices, practice site: hospital, open-care or mixed). For each wave, 250 rheumatologists contribute, i.e. a 10% sampling rate (approximately 2500 rheumatologists in France) and include all patients with RA over a 2-week period, with a cut-off of 20 patients per physician. In all, 2000 patients, of whom 700 on biotherapy were analyzed; • the auto-immune and rituximab RA (AIR-PR) registry, which monitors a cohort of RA patients on rituximab treatment [16].
The primary objective of the registry is evaluation of the safety profile of rituximab. The secondary objective is evaluation of the efficacy of rituximab in the patients on treatment. The various modalities of retreatment in real-life settings are also studied. The cost of each of the treatments was calculated over 1 year and per patient as a function of the treatment modalities and hospital or open-care follow-up. The treatment share of each management method was determined. Table 2 shows the disease management trajectory over 1 year for each biotherapy. 2.5. Application of the management context to the rheumatoid arthritis patient population The following are available: the annual cost of each of the management methods (theoretical and real-life settings) by biotherapy and by patient; the distribution of biotherapy market share in the 3rd quarter 2008 (source: GFK Performance Tracker Q3 2008, a quarterly international study). Contributing rheumatologists were sampled using several criteria. The rheumatologists were to have 3 years of experience in the treatment of RA and there were to be at least 50 patients with RA among the rheumatologist’s patients. The rheumatologists were also to have experience with biotherapies and to have prescribed biotherapies to at least eight patients in the 3 months preceding the quarterly wave in order to participate. For each wave, 80 rheumatologists take part. Half of the participants are hospital rheumatologists and half mixed-practice rheumatologists. The participants include the ten patients who have most recently consulted, generating a total of 800 patients per wave. For each wave, new rheumatologists participate. The data used in this study were generated by the August 2008 wave. The patient distribution by biotherapy was as follows: 37% on etanercept, 30% on adalimumab, 15% on infliximab, 12% on rituximab and 6% on abatacept. To determine the overall cost of biotherapy in France, the cost of management by biotherapy and by patient was multiplied by the previously described distribution in the RA patient population defined above.
322
M. Maravic / Joint Bone Spine 77 (2010) 319–324
Table 3 Annual management cost by patient healthcare trajectory. Biotherapy
Abatacept
Management recommended in the SPC or reference systems (HAS/AFSSAPS) Annual cost (impact of acquisition treatment cost on overall cost) 21,128 D (81%)
Adalimumaba
15,118 D (98%)
Etanercepta
14,861 D (98%)
Infliximab
11,567 D (79%)
Rituximaba
12,957 D (89%)
‘Real-life’ setting management Annual cost (impact of acquisition treatment cost on overall cost) 19,618 D (81%) 14,950 D 15,464 D 14,693 D 14,849 D 12,144 D
if patient injection (99%) if SRN injection (98%) if patient injection (99%) if SRN injection (98%) (84%)
6,451 D if 1st course (89%) 12,789 D if 2nd course (90%)
Sensitivity analysis based on real-life setting* Annual cost (impact of acquisition treatment cost on overall cost) lh: 14,336 D (74%) hh: 24,901 D (85%) lh: 14,950 D (99 %) hh: 15,464 D (98 %) lh: 14,693 D (99%) hh: 14,849 D (98%) lh: 11,408 D (83%) hh: 20,198 D (91%) lh: 6,451 D (89%) hh: 12,789 D (90%)
SC: subcutaneous; SRN: state-registered nurse, injection tariff excluding travel: 3;◦ ◦ Results of the PANEL A+A inquiry, March-April 2008;◦ ◦ ◦ Auto-immunity and rituximab RA registry [16]. Refer to Table 2 for the percentage patients receiving injections by an SRN (adalimumab, etanercept) or receiving a second course of treatment (rituximab). a The annual cost calculation in the ‘real-life’ setting and in the sensitivity analysis are the same for the three biotherapies since the calculation was done for a given patient. * Low hypothesis (lh), high hypothesis (hh).
2.6. Sensitivity analysis The sensitivity analysis addressed the ‘real-life’ setting data using a low and a high hypothesis (Table 2). The aim was to evaluate the impact of patient body weight variation, which impacts the dosage prescribed for abatacept and infliximab, subcutaneous injection administration for adalimumab and etanercept, and the potential second course of rituximab treatment. 3. Results
the percentage for which a second course of treatment is administered (rituximab). Irrespective of the management trajectory considered, it will be observed that the biotherapy acquisition cost constitutes the greater part of the annual cost (79 to 98% of the annual cost, depending on the biotherapy, in the theoretical management setting). The variation in the treatment-related cost is greater if dosage varies as a function of patient body weight (abatacept, infliximab) or dosing frequency (rituximab). For biotherapies administered by the subcutaneous route (adalimumab, etanercept), injection or non-injection by a SRN has little impact on the overall cost.
3.1. Definition of the patient population In 2007, there were 23.4 million short-stay hospitalizations in the public and private sector in France. In all, 79,560 hospitalizations (0.34%) were hospitalizations in which RA was reported as the primary or linked diagnosis. Given the selection criteria defined above, the RA patient population was 15,873 (1 to 33 hospitalizations per patient; 39,549 hospitalizations in all). 3.2. Annual cost per patient by healthcare trajectory Table 3 shows the annual cost of each biotherapy by the defined healthcare trajectory. The data are presented as the cost of management by trajectory and indicate the percentage biotherapy acquisition cost. With regard to adalimumab, etanercept and rituximab, the calculations relate to the real-life setting management data and the sensitivity analyses are the same for a given patient. On the RA patient population scale, the cost of biotherapy management is modulated by application of the percentage of patients for which injection is administered by a nurse (adalimumab, etanercept) and
3.3. Biotherapy distribution in the patient population and calculation of the corresponding economic impact Table 4 shows the distribution of the various biotherapies in the RA patient population and the annual cost as a function of the type of management. As observed with the evaluation of the theoretical setting management trajectory annual cost, there is little variation between the theoretical management setting, real-life setting, and simulation scenarios for treatments independent of patient weight, dosage used and repeat-treatment, i.e. subcutaneous biotherapies. The impact of body weight varies as a function of the treatment considered. With abatecept, injection of three vials for a body weight fluctuating between 60 and 100 kg has less impact than modulation in terms of body weight and dose for infliximab (maximum dose: 7.5 mg/kg per administration). The results may also be ranked in terms of 1st- and 2nd-line treatment and as a function of the sector whose expenditures are impacted (open-care or hospital care). First-line biotherapies (adalimumab, etanercept and infliximab) account for 82% of the treated target population. In the context of theoretical or real-life
Table 4 Distribution of the patient population (number, percentage) and total annual cost of management (2009 Euros, percentage) by biotherapy for the entire patient population. Patient population as number of patients (%)a
Theoretical management cost, D (%)
Real-life setting cost, Euros (%)
Low hypothesis cost, D * (%)
High hypothesis cost, D * (%)
Abatacept Adalimumab Etanercept Infliximab Rituximab
952 (6) 4,762 (30) 5,873 (37) 2,381 (15) 1,905 (12)
20,113,856 (9) 71,991,916 (31) 87,278,653 (37) 27,541,027 (12) 24,683,085 (11)
18,676,336 (9) 71,252,500 (32) 86,575,567 (39) 28,914,864 (13) 16,029,434 (7)
13,647,872 (6) 71,191,900 (34) 86,291,549 (41) 27,162,448 (13) 12,289,155 (6)
23,705,752 (9) 73,639,568 (29) 87,208,177 (34) 48,091,438 (19) 24,363,045 (9)
Total
15,873 (100)
231,608,537 (100)
221,448,701 (100)
210,582,924 (100)
257,007,980 (100)
a *
The RA patient population was determined from the PMSI data. Based on the real-life setting data. The patient distribution by biotherapy is based on the market share data for the 3rd quarter 2008.
M. Maravic / Joint Bone Spine 77 (2010) 319–324
setting follow-up, 1st-line biotherapies generate 80 and 84% of management costs. Second-line biotherapies (abatacept and rituximab) are prescribed for the other 18% of the target population and generate 20 and 16% of the annual costs in the theoretical and reallife management settings, respectively. Open-care biotherapies are administered to 61% of the patient population and generate 68 and 71% of the total cost for the theoretical and real-life management settings, respectively. 4. Discussion and limitations of the study In the present context of reforms in France, it is important to determine the RA target population receiving biotherapy and the national cost generated, particularly if biotherapy use is to be controlled. Data interpretation is also to take into account the status of the drugs as defined by French regulations [5]. Both hospital-prescribed drugs (adalimumab, etanercept and rituximab) and hospital only drugs (abatacept and infliximab) were addressed. Biotherapies administered by the subcutaneous route impact opencare expenditures, while the other biotherapies impact hospital expenditures. It is also important to determine the biotherapy target population. Few epidemiological data on RA in France are available since all the data are not accessible, particularly those of the National Health Insurance Organization. While the National Health Insurance Organization publishes data, they are not sufficiently precise. The data does show that, at the end of 2007, there were 150,032 No. 22 long-term diseases (LTD) linked to RA [17]. Outside of the issues that may be raised by verification of diagnostic validity by comparing the LTD declaration with the patient’s file, the data are not informative with regard to the proportion of patients on biotherapy. A recent study of RA-related LTD in the Nord-Pas-deCalais region showed that 7,178 LTD were reported in 2005 [18] but 40% of the patients had no DMARD treatment. This point highlights to a probable overestimation of the number of cases of RA. Among the patients on DMARD treatment (the remaining 60%), biotherapies accounted for 45% of the DMARD treatments prescribed alone or in combination with more conventional DMARD treatments. Etanercept is the most widely prescribed biotherapy. Data on 2007 open-care drug consumption are available. The data shows that etanercept and adalimumab rank 7th (134.4 million D ) and 23rd (87.4 million D ), respectively, in terms of prescription and reimbursement, but the distribution by disease is not indicated [19]. The hospitalization data gives an idea of the number of hospitalizations for RA and theoretically enables targeting those associated with the costly drug list (which is the case for abatacept, infliximab and rituximab). While adalimumab and etanercept are also on the list, in practice they should only represent a minimal proportion of hospitalizations simply for drug administration. Information and particularly the links between hospitalization and drug consumption became mandatory in 2007 and are to be reported to a specific database [20]. With regard to the modalities for hospital selection, it is to be noted, as was the case for LTD, that it is not possible to compare the reported data with the patient files due to the anonymous nature of medical records. Moreover, considering that an RA case hospitalized at least once is a case liable to benefit from biotherapy is highly debatable. There is no evidence that all patients for whom biotherapy is discussed are hospitalized patients, particularly in the case of hospital prescribed biotherapies. A hospital prescription for an outpatient enables the biotherapy considered to be administered in an outpatient setting. Moreover, the diagnostic codes used for PMSI coding do not enable definition of the seriousness of RA. A patient with RA may also be hospitalized for another reason: confirmation of the diagnosis and initiation of conventional DMARD treatment, implementation of radioisotope synoviorthesis, etc. In addition, the present study addressed patients who respond to treatment and are
323
free from adverse events. This may be remote from real-life settings given that patients may not respond and thus require a change in DMARD. The annual cost estimate only takes into account the direct medical costs and those liable to vary between biotherapies. Given that the context is a theoretical management context, it was important to develop that context using real-life setting data modulating the information, such as body weight and/or dose (abatacept and infliximab), SRN injection administration or not (adalimumab and etanercept) and repeat-treatment (rituximab). In the context of theoretical management, depending on the product, the treatment cost is equivalent to 79 to 98% of the annual cost. The annual cost varies, depending on the product considered, from 11,567 to 21,128 D . Those data differs little from the real-life setting data taken into account in the study. In the literature, the treatment cost is identified as a major component of direct medical costs [21–23]. Although the context is theoretical, the treatment cost does not differ from those reported by Jönsson et al. using 2006 sales data (irrespective of the indication for biotherapy): annual treatment cost of 15,000 to 19,000 D for abatacept; 10,000 to 15,000 D for adalimumab; 9,000 to 15,000 D for etanercept; 10,000 to 18,000 D for infliximab; and 7,000 to 10,000 D for rituximab [24]. Biotherapy cost distribution is to be interpreted as a function of the study context, year of the study and country in which it was conducted. US studies have evaluated biotherapy use on the basis of a preference questionnaire [25], database [26] and prospective study [27]. Irrespective of whether the evaluation is at a time t [25] or over several years [26,27], the results vary depending on the methodology. Depending on the study, the results are in favour of etanercept [26–27] or infliximab [27]. However, the studies are of limited pertinence in that they only consider TNF antagonist biotherapies. In conclusion, it would seem important for every country to have pertinent information on the epidemiology of patients with diseases requiring the use of expensive treatments and, more specifically, patients liable to have need of them since they meet the criteria defined by the recommendations. This study attempted an optimal estimation of the impact of RA biotherapies on the basis of the information available in France. However, the findings are only an initial approach. In a context in which healthcare expenditure regulation is evolving, in particular with the set-up of expenditure ceilings for costly drugs and medical devices, the information available on patient trajectories and the impact of the necessary drugs or devices by disease will enable decision-makers to adjust access to certain therapies as a function of the population’ real needs. Conflict of interest None of the authors has any conflict of interest to declare. References [1] Arrêté 3 février 2005 portant approbation de la convention nationale des médecins généralistes et des médecins spécialistes. Journal Officiel 2005. [2] La classification Commune des actes médicaux. Site Ameli : http://www.ameli. fr/professionnels-de-sante/medecins/exercer-au-quotidien/nomenclatureset-codage/codage-des-actes-medicaux-ccam.php. [3] Maravic M, Le Bihan C, Landais P. La tarification à l’activité : définition, modalités et tarifs en rhumatologie. Rev Rhum 2004;71:642–4. [4] Décret n◦ 2008-1121 du 31 octobre 2008 relatif au contrat de bon usage des médicaments et des produits et prestations mentionné à l’article L. 162-22-7 du code de la sécurité sociale. Journal Officiel 2008. [5] Décret no 2004-546 du 15 juin 2004 relatif aux catégories de médicaments à prescription restreinte et à la vente de médicaments au public par certains établissements de santé et modifiant le code de la santé publique et le code de la sécurité sociale (deuxième partie : décrets en conseil d’État). Journal Officiel 2004. [6] Loi n◦ 2008-1330 du 17 décembre 2008 de financement de la sécurité sociale pour 2009 (1). Journal Officiel 2008 http://www.legifrance.gouv.fr/affichTexte.
324
[7]
[8] [9]
[10]
[11]
[12]
[13]
[14]
[15]
M. Maravic / Joint Bone Spine 77 (2010) 319–324 do;jsessionid=E4F710107B3AF52A2CB42D2373DDA587.tpdjo10v 1?cidTexte= LEGITEXT000019945312&dateTexte=20090329. Circulaire N◦ DHOS/F2/F3/F1/DSS/1A/2009/78 du 17 mars 2009 relative à la campagne tarifaire 2009 des établissements de santé et son annexe IV : évolution du modèle de financement des activités de court séjour. Ministère de la Santé et des Sports. Classification statistique internationale des maladies et des problèmes de santé connexes. 10e révision. Organisation Mondiale de la santé 1993. Site Haute Autorité de la santé : polyarthrite rhumatoïde : diagnostic et prise en charge initiale, 2007 : http://www.has-sante.fr/portail/jcms/c 606479/ polyarthrite-rhumatoide-diagnostic-et-prise-en-charge-initiale. Polyarthrite Rhumatoïde : prise en charge en phase d’état, 2007 : http://www.has-sante.fr/ portail/jcms/c 606483/polyarthrite-rhumatoide-prise-en-charge-en-phasedetat. Site AFSSAPS : tarification à l’activité (T2A) et médicaments hors groupe homogène de séjours : http://www.afssaps.fr/Dossiers-thematiques/ Tarification-a-l-activite-T2A-medicaments/Accueil-T2A/%28offset%29/0. Site Assurance Maladie : base des médicaments et informations tarifaires : http://www.codage.ext.cnamts.fr/codif/bdm it/index tele ucd.php?p site= AMELI. Circulaire n◦ DHOS/F1/MTAA/2006/376 du 31 août 2006 relative aux conditions de facturation d’un GHS pour les prises en charge hospitalières en zone de surveillance de très courte durée ainsi que pour les prises en charge de moins d’une journée. Ministère de la Santé et des Solidarités. Arrêté du 19 février 2009 relatif à la classification et à la prise en charge des prestations d’hospitalisation pour les activités de médecine, chirurgie, obstétrique et odontologie et pris en application de l’article L. 162-22-6 du code de la sécurité sociale. Journal Officiel 2009. Arrêté du 27 février 2009 fixant pour l’année 2009 les ressources d’assurance maladie des établissements de santé exerc¸ant une activité de médecine, chirurgie, obstétrique et odontologie. Journal Officiel 2009. Site Assurance Maladie : Nomenclature générale des actes professionnels http://www.ameli.fr/professionnels-de-sante/infirmiers/exercer-auquotidien/n.g.a.p.-et-l.p.p./la-n.g.a.p.php.
[16] Registre AIR-PR http://www.rhumatologie.asso.fr/12-Recherche/registres. asp?intSM=SM3&strLien=Lien30, http://www.abstracts2view.com/eular/view. php?nu=EULAR08L THU018. [17] Assurance Maladie. Les personnes en affection de longue durée au 31 décembre 2007. Points de repère n◦ 20, Novembre 2008. [18] Biver E, Beague V, Verloop D, et al. Low and stable prevalence of rheumatoid arthritis in northern France. Joint Bone Spine 2009;76(5):497–500. [19] Assurance Maladie.Médicaments remboursés : analyse de la consommation en 2007. 1er août 2008. [20] DHOS: précisions complémentaires concernant les fichiers FichComp et FichSup que doivent transmettre les établissements antérieurement sous dotation globale en 2007 : http://www.atih.sante.fr/?id=0002200074FF. [21] Maravic M. Economic burden of rheumatoid arthritis in France. Exp Rev Pharmacoeconomics Outcome Res 2006;6:25–32. [22] Fautrel B, Woronoff-Lemsi MC, Etghen M, et al. Influence des pratiques médicales sur les coûts de traitement de la polyarthrite rhumatoïde par biothérapie anti-TNF␣ en France. Rev Rhum 2005;72:1306–13. [23] Kobelt G, Woronoff AS, Richard B, et al. Disease status, costs and quality of life of patients with rheumatoid arthritis in France: the ECO-PR Study. Joint Bone Spine 2008;75(4):408–15. [24] Jönsson B, Kobelt G, Smolen J. The burden of rheumatoid arthritis and access to treatment: uptake of new therapies. Eur J Health Econ 2008;8(Suppl2): S61–86. [25] Kamal KM, Madhavan SS, Allen Hornsby JAA, et al. Use of tumor necrosis factor inhibitors in rheumatoid arthritis: a national survey of practicing United States rheumatologists. Joint Bone Spine 2006;73:718–24. [26] Yazici Y, Shi N, John A. Utilization of biological agents in rheumatoid arthritis in the United States. Analysis of prescribing patterns in 16,752 newly diagnosed patients and patients new to biologic therapy. NYU Hosp Jt Dis 2008;66: 77–85. [27] Gibofsky A, Palmer WR, Goldman JA, et al. Real-world utilization of DMARDs and biologics in rheumatoid arthritis: the RADIUS (rheumatoid arthritis disease-modifying anti-rheumatic drug intervention and utilization study) study. Cur Med Res Opin 2006;22:169–83.