You are here

Conditions of prescription of anti-TNF agents in newly treated patients with inflammatory bowel disease in France (2011–2013)

Digestive and Liver Disease, Volume 48, Issue 6, June 2016, Pages 620 - 625

Abstract

Background

Tumour necrosis factor inhibitors (anti-TNFs) are active but expensive drugs that induce and maintain remission in patients with Crohn's disease (CD) and ulcerative colitis (UC).

Aims

To assess the trends in anti-TNF prescription and the conditions of prescription of these drugs in patients with inflammatory bowel disease (IBD) in France.

Methods

Incidence study of anti-TNF use was performed based on French medico-administrative databases (SNIIRAM/PMSI). IBD patients who initiated adalimumab or infliximab between 2011 and 2013 were selected.

Results

The number of new anti-TNF users increased from 4571 to 5875 between 2011 and 2013 (+29%). More specifically, the number of patients not treated with immunosuppressants (IS) during the previous 12 months increased from 2100 to 3007 (+43%), among whom 379 patients in 2011 and 570 patients in 2013 started combination therapy (+50%). These trends were observed for both CD and UC. Patients who were naïve of IS were hospitalised more frequently than those treated with IS prior to anti-TNF therapy.

Conclusion

This study shows a rapid increase in new prescriptions of anti-TNF for both CD and UC in France between 2011 and 2013. These results suggest a change in medical practices, with anti-TNF agents prescribed more often as first-line maintenance treatment.

Keywords: Crohn's disease, Healthcare resource use, Trends, Ulcerative colitis.

1. Introduction

Inflammatory bowel diseases (IBD) include Crohn's disease (CD) and ulcerative colitis (UC). These diseases require lifelong care and often lead to hospitalisations and surgery. In France, health expenses for IBD are fully covered by French national health insurance. Randomized trials have shown that tumour necrosis factor alpha inhibitors (anti-TNF) are efficacious to induce and maintain remission in CD [1], [2], [3], [4], and [5] and UC [6] and [7]. These drugs have been approved in the European Union since 1999. Three anti-TNFs are currently indicated in IBD, infliximab (since 1999 for CD and 2006 for UC), adalimumab (since 2007 for CD and 2012 for UC), and golimumab (since the end of 2013 for UC). The European labelling states that these drugs are indicated in CD patients “who have not responded despite a full and adequate course of therapy with a corticosteroid and/or an immunosuppressant (IS); or who are intolerant to or have medical contraindications for such therapies”, and in UC patients “who have had an inadequate response to conventional therapy including corticosteroids and 6-mercaptopurine (6-MP) or azathioprine (AZA), or who are intolerant to or have medical contraindications for such therapies” [8] and [9]. In France, the cost of anti-TNF drugs is approximately €12,000 per year (excluding the additional cost of a nurse, and outpatient care in hospitals or clinics for infliximab). By comparison, azathioprine costs €360 per year. In the current era of escalating healthcare costs and growing constraints on healthcare budgets, it seems necessary to evaluate the evolution of prescription and to compare it to readily available evidence of the efficacy of these drugs.

The aim of the present study was to describe the trends in anti-TNF prescription and the conditions of prescription of these drugs in newly treated patients with IBD in France.

2. Methods

Incidence study of anti-TNF use was performed based on French medico-administrative databases.

2.1. Data source

Individual data were extracted from both the French national health insurance information system (i.e., the SNIIRAM (Système National d’Information InterRégimes de l’Assurance Maladie)) and the French hospital discharge database (i.e., the PMSI (Programme de Médicalisation des Systèmes d’Information)). These health databases can be linked by means of a unique anonymous number allocated to each individual, providing detailed information on real-life management of 99% of the population living in France (around 65,000,000 people) [10] and [11].

The SNIIRAM contains data concerning all outpatient services reimbursed by the national health insurance system, including drugs, physician visits, laboratory or imaging investigations. Patients with chronic diseases (Affections de longue durée, i.e., long-term diseases (LTD)) such as IBD are 100% reimbursed for their health expenditures, and the diagnosis is recorded in the SNIIRAM. The eligibility for an LTD (with its International Classification of Disease (ICD)-10 codes, and the date of start of the LTD) is established by a national health insurance expert physician, following a request from the patient's family practitioner.

The PMSI database contains the procedures performed during all hospital stays, and the principal (DP), related (DR) or associated (DAS) diagnoses. The DP and DR correspond to the diseases, which have led to the hospitalisation (in particular, the DR is used when the patient is hospitalised for the administration of a medical treatment, such as intravenous anti-TNF). Procedures are coded according to the Classification Commune des Actes Médicaux (CCAM) (French medical classification for clinical procedures) and diagnoses are coded according to ICD-10. The PMSI database also contains information on the expensive drugs dispensed, such as anti-TNF agents, which have to be invoiced in addition to the stay when administered in hospital.

2.2. Study population

All IBD patients treated with adalimumab or infliximab between 2011 and 2013, and who had not received reimbursement for these drugs during the 12 months before the first prescription claim (index date) were included. At least one health service claim between 24 and 13 months before the index date was required to provide evidence of continuous enrolment for a one-year preindex period. Individuals were considered to have IBD when they had an IBD-related ICD-10 code (K50 (CD) or K51 (UC)) for an LTD starting before the index date, or when they were hospitalised for IBD during the 12 months prior to the index date (K50/K51 code for a DP/DR, or a DAS if the main diagnosis referred to an IBD complication). To avoid diagnostic ambiguity between CD and UC, patients with an LTD or hospital codes for both CD and UC were considered as a separate group (“IBD unclassified”). Patients who had an LTD starting before the index date or a hospital stay (DP, DR, DAS) during the 12 months prior to the index date for rheumatoid arthritis, juvenile arthritis, psoriatic arthritis, psoriasis or ankylosing spondylitis were excluded, as they may have received anti-TNF therapy for an indication other than IBD.

2.3. Variables of interest

Age, gender and commune of residence at the index date were extracted from the SNIIRAM databases. The Rey commune-level deprivation index was used [12]. IBD-related drugs dispensed during the 6 months (corticosteroids, salicylates, cholestyramine, antidiarrhoeal drugs, and analgesics) or 12 months (thiopurines, methotrexate, cyclosporin, tacrolimus) before the index date were identified from outpatient prescription claims. Bowel resections, surgical procedures for anorectal abscess or fistula and IBD-related hospitalisations for at least one night during the 12 months before the index date (and/or starting on the day of the initiation of anti-TNF therapy), were identified from the PMSI database. Hospitalisations for CD and UC were defined as hospital stays with either a DP/DR coded as K50 (CD) or K51 (UC), or a DAS coded as K50 (CD) or K51 (UC), when the DP/DR referred to an IBD complication (digestive obstruction/stenosis/perforation (K56, K631, K624), pain (R104), abscess (K61, K630), fistula (K60, K632), haemorrhage (K625, K922, D500) and peritonitis (K650)).

2.4. Statistical analysis

Statistical analyses were performed using SAS V9 software. For each year between 2011 and 2013, the numbers of new users of adalimumab and infliximab were calculated, for IBD, CD, UC and the IBD unclassified group. The numbers of new users were also calculated according to previous use of IS. The socio-demographic characteristics of the patients, the treatments they received and their hospitalisation rates during the year before the index date were described for 2013. Comparisons between IS-naïve new users and other patients were analysed with Student's t test for continuous variables and Chi-square test for dichotomous variables (Fisher's exact test for small numbers).

3. Results

3.1. Study population (Fig. 1)

A total of 183,000, 193,000 and 203,000 patients had a LTD or were hospitalised for IBD in France in 2011, 2012 and 2013, respectively. Among them, 16,500, 20,300 and 24,300 patients were treated with anti-TNF in 2011, 2012 and 2013, accounting for 9%, 11% and 12% of IBD patients, respectively. Finally, 4571, 5398 and 5875 IBD patients started a treatment with anti-TNF in 2011, 2012 and 2013, respectively.

gr1

Fig. 1 Inclusion flow chart (solid line) and number of inflammatory bowel disease (IBD) patients (dotted line) in 2011, 2012 and 2013.

Therefore, between 2011 and 2013, the number of IBD patients newly treated with adalimumab or infliximab increased by 29% (Fig. 2). This increase concerned both UC (+39%) and CD (+29%), and all age groups (Supplemental Table 1). The number of new prescriptions for adalimumab increased, with nearly twice as many patients initiating adalimumab in 2013 than in 2011, while the number of infliximab prescriptions remained stable (Supplemental Table 1).

gr2

Fig. 2 Number of inflammatory bowel disease (IBD) patients initiating anti-TNF therapy with adalimumab or infliximab in 2011, 2012 and 2013 in France.

Table 1 shows the numbers of new users according to previous use of IS. Overall, the number of anti-TNF users who had received IS during the 12 months before initiation of anti-TNF therapy increased from 2471 to 2868 (+16%) between 2011 and 2013. The number of patients who did not receive IS prior to anti-TNF increased more markedly, from 2100 to 3007 patients (+43%). The number of patients who initiated a treatment with both anti-TNF and IS (combination therapy) increased by 50%. This was observed for both adalimumab and infliximab (Supplemental Table 2), in patients with CD, UC and unclassified IBD.

Table 1 Number of inflammatory bowel disease (IBD) patients initiating anti-TNF therapy according to immunosuppressant (IS) use during the year before initiation of anti-TNF, in 2011, 2012 and 2013.

Patients initiating anti-TNF therapy with adalimumab or infliximaba % increase over the study period
2011
N (%)
2012
N (%)
2013
N (%)
All IBD 4571 (100) 5398 (100) 5875 (100)
 Use of IS before initiation of anti-TNF therapyb 2471 (54) 2769 (51) 2868 (49) +16%
  Azathioprine, mercaptopurine 2227 (9) 2540 (47) 2648 (45)
  Ciclosporin, tacrolimus, methotrexate 357 (8) 321 (6) 312 (5)
 No use of IS before initiation of anti-TNF therapyb 2100 (46) 2629 (49) 3007 (51) +43%
  Combination therapy c 379 (8) 471 (9) 570 (10) +50%
  Anti-TNF alone 1721 (38) 2158 (40) 2437 (41) +41%
Crohn's disease 3003 (100) 3591 (100) 3878 (100)
 Use of IS before initiation of anti-TNF therapyb 1574 (52) 1781 (50) 1838 (47) +17%
  Azathioprine, mercaptopurine 1402 (47) 1624 (45) 1698 (44)
  Ciclosporin, tacrolimus, methotrexate 228 (8) 203 (6) 180 (5)
 No use of IS before initiation of anti-TNF therapyb 1429 (48) 1810 (50) 2040 (53) +43%
  Combination therapy c 255 (9) 316 (8) 391 (10) +53%
  Anti-TNF alone 1174 (39) 1494 (42) 1649 (43) +41%
Ulcerative colitis 959 (100) 1172 (100) 1335 (100)
 Use of IS before initiation of anti-TNF therapyb 536 (56) 647 (55) 701 (53) +31%
  Azathioprine, mercaptopurine 496 (52) 602 (51) 651 (49)
  Ciclosporin, tacrolimus, methotrexate 80 (8) 78 (7) 86 (6)
 No use of IS before initiation of anti-TNF therapyb 423 (44) 525 (45) 634 (48) +50%
  Combination therapyc 79 (8) 99 (8) 116 (9) +47%
  Anti-TNF alone 344 (36) 426 (36) 518 (39) +51%
IBD unclassifiedd 609 (100) 635 (100) 662 (100)
 Use of IS before initiation of anti-TNF therapyb 361 (59) 341 (54) 329 (50) −9%
  Azathioprine, mercaptopurine 329 (54) 314 (49) 299 (45)
  Ciclosporin, tacrolimus, methotrexate 49 (8) 40 (6) 46 (7)
 No use of IS before initiation of anti-TNF therapyb 248 (41) 294 (46) 333 (50) +34%
  Combination therapyc 45 (7) 56 (9) 63 (10) +40%
  Anti-TNF alone 203 (33) 238 (37) 270 (41) +33%

a No use of adalimumab or infliximab during the 12 months before initiation of anti-TNF therapy.

b Between 12 months and 1 month before initiation of anti-TNF therapy. Some patients may have been prescribed both thiopurines and other IS in the 12 months before anti-TNF therapy.

c Dispensing of IS during the month before or the month after initiation of anti-TNF therapy.

d Patients with long-term disease status or hospital ICD-10 codes for both UC and CD.

IBD, inflammatory bowel disease; IS: immunosuppressant.

3.3. Characteristics of CD and UC new users in 2013 (Table 2)

Overall, patients who were prescribed anti-TNF without having received IS during the 12 months before initiation of anti-TNF therapy were very similar to the other patients in terms of age, gender and deprivation index of the area of residence. As expected, the disease was registered as an LTD for majority of the patients. Characteristics of infliximab and adalimumab new users are shown in Supplemental Tables 3 and 4.

Table 2 Socio-demographic characteristics and long-term disease status of patients with Crohn's disease or ulcerative colitis initiating anti-TNF therapy in 2013, according to immunosuppressant use during the year before initiation.

Crohn's disease Ulcerative colitis
Previous use of ISa No previous use of ISa All patients Previous use of ISa No previous use of ISa All patients
N = 1838 (%) N = 2040 (%) N = 3878 (%) N = 701 (%) N = 634 (%) N = 1335 (%)
Age at anti-TNF initiation (mean) 34.4 36.4 35.5 41.7 41.7 41.7
Male gender 832 (45) 888 (43) 1720 (44) 382 (55) 333 (53) 715 (54)
Deprivation index of area of residence (town)
 Quintile 1 (least deprived) 281 (17) 319 (17) 600 (17) 90 (15) 97 (18) 187 (16)
 Quintile 2 289 (18) 366 (20) 655 (19) 124 (21) 118 (22) 242 (21)
 Quintile 3 302 (18) 339 (18) 641 (18) 143 (24) 106 (20) 249 (22)
 Quintile 4 364 (22) 343 (19) 707 (20) 131 (22) 109 (20) 240 (21)
 Quintile 5 (most deprived) 409 (25) 474 (26) 883 (25) 115 (19) 105 (20) 220 (19)
Missing 193 199 392 98 99 197
IBD registered as a long term disease 1715 (93) 1858 (91) 3573 (92) 646 (92) 548 (86) 1194 (89)
Time since the start of long term disease
 ≤4 years 1008 (59) 1098 (59) 2106 (59) 397 (62) 341 (62) 738 (62)
  <1 year 393 (23) 695 (37) 1088 (31) 136 (21) 203 (37) 339 (28)
  1–4 years 615 (36) 403 (22) 1018 (29) 261 (40) 138 (25) 399 (33)
 5–9 years 311 (18) 303 (16) 614 (17) 120 (19) 109 (20) 229 (19)
 ≥10 years 396 (23) 457 (25) 853 (24) 129 (20.0) 98 (18) 227 (19)

a No use of immunosuppressant between 12 months and 1 month before initiation of anti-TNF therapy.

IBD, inflammatory bowel disease; IS: immunosuppressant.

3.4. Resource use before the initiation of anti-TNF therapy according to the previous dispensation of IS (supplemental Table 5 and Table 3)

New anti-TNF users naive of IS had received corticosteroids less often than patients treated with IS during the 12 months before the index date (Supplemental Table 5). This was observed for both CD (60% vs. 66%, p < 0.001) and UC patients (80% vs. 84%, p = 0.12). On the contrary, salicylates, antidiarrhoeal drugs and analgesics were prescribed more often to IS-naive anti-TNF users, in particularly for UC patients.

Table 3 Resource use among immunosuppressant-naive anti-TNF new users, according to whether or not they received combination therapy.

Crohn's disease
N = 2040
Ulcerative colitis
N = 634
Anti-TNF alonea Combination therapyb p-value Anti-TNF alonea Combination therapyb p-value
N = 1649 N = 391 N = 518 N = 116
Corticosteroids during the previous six months 996 (60) 224 (57) ns 404 (78) 106 (91) 0.001
Overnight hospitalization for CD (CD patients) or UC (UC patients) during the previous 12 months 696 (42) 214 (55) <0.001 210 (41) 70 (60) <0.001
Surgery for anorectal abscess or fistula during the previous 12 months 82 (5) 43 (11) <0.001 6 (1.2) 3 (3) ns
Bowel resection during the previous 12 months 95 (6) 8 (2) 0.002 4 (0.8) 0 (0.0) ns

a No use of immunosuppressant between 12 months and 1 month before initiation of anti-TNF therapy.

b Dispensing of immunosuppressant during the month before or the month after initiation of anti-TNF therapy.

ns, non significant; CD: Crohn's disease; UC: ulcerative colitis

In addition, IS-naïve anti-TNF users with CD had been more frequently hospitalised for CD (45% vs. 39%, p < 0.001) and operated for anorectal abscess or fistula (6% vs. 4%, p < 0.001) during the previous 12 months than those who had received IS prior to starting anti-TNF (Supplemental Table 5). Similarly, IS-naïve anti-TNF users with UC had been more frequently hospitalised for UC during the previous 12 months than the other patients (44% vs. 31%, p < 0.001).

A recent history of surgery and hospitalisation was more frequent among patients initiating combination therapy (Table 3). In particular, 11% of CD patients initiating combination therapy had undergone surgery for anorectal abscess or fistula during the previous 12 months, and 60% of UC patients had been hospitalised during the previous year.

Overall, 366 CD patients (9%) and 51 UC patients (4%) had neither corticosteroids, nor IS, nor surgery nor hospitalisation during the 12 months before initiation of anti-TNF therapy. These numbers were 65 (6%) and 7 (2%) among the 1088 CD patients and 339 UC patients who had an LTD for less than one year, respectively.

Patients newly treated with infliximab had been more frequently hospitalised than those newly treated with adalimumab, for both CD and UC (Supplemental Tables 6 and 7). The difference was particularly marked for IS-naive UC patients: 22% were hospitalised during the 12 months before initiation of adalimumab vs. 64% for infliximab.

4. Discussion

This study shows a marked increase in new prescriptions of anti-TNF in IBD patients, between 2011 and 2013, in France. This trend was particularly pronounced among patients not treated with IS during the previous 12 months, for both CD and UC. These patients were more frequently hospitalised or operated for IBD during the 12 months before initiation of anti-TNF therapy than those previously treated with IS.

Such a trend is unlikely to be due to an increase in IBD incidence [13] and [14]. In our source population, the number of IBD patients increased by 10% between 2011 and 2013, lower than the trend observed for anti-TNF use in this study. It is also unlikely that the severity of IBD increased over the study period. Interestingly, in a United-States study, also based on insurance claims data, the proportion of anti-TNF users among patients receiving health care for IBD between 2010 and 2012 (11%) [15] was similar to that reported in the present study.

The increase in new prescriptions of anti-TNF therapy was particularly marked among patients not treated with IS during the previous 12 months. Anti-TNFs can be used to induce clinical remission in steroid-refractory patients, but a marked increase in this category of patients seems unlikely. Our results rather suggest a change in medical practices, with anti-TNF agents prescribed more often as first-line maintenance treatment, instead of thiopurines. IS-naive patients at the index date (particularly those starting a combination therapy) had a history of surgery and hospitalisations more often than the patients previously treated with IS, suggesting that IS-naïve patients had more severe IBD.

Few controlled trials have compared anti-TNF to IS. A single randomized trial, performed in IS-naive CD patients, has shown that endoscopic healing is more often obtained with azathioprine and infliximab than with infliximab alone or azathioprine alone [16]. In the REACT open-label cluster randomized controlled trial performed in community gastroenterology practices, corticosteroid-refractory CD patients (either corticosteroid-dependent or resistant) randomised to adalimumab combined with IS had a similar rate of steroid-free remission at 12 months (primary endpoint) than those randomised to conventional management, but a lower risk of major adverse outcomes at 24 months (surgery and hospitalisation; both secondary outcomes) [17]. A single trial has compared azathioprine, infliximab, and a combination of both in IS-naive UC patients. It showed that patients treated with infliximab combined with azathioprine were more likely to achieve corticosteroid-free remission at 16 weeks than those receiving a monotherapy [18]. Combination therapy also led to significantly better mucosal healing than azathioprine monotherapy. No such trial has been conducted with adalimumab.

Overall, data supporting first-line maintenance treatment with anti-TNF, rather than after failure of azathioprine, are limited. The REACT trial suggests that the early use of anti-TNF is associated with a reduced risk of surgery and hospitalisation [17]. Yet, its conclusion is limited to CD patients who do not respond to corticosteroids or relapse when the dose is tapered. In addition, it was an open trial and it was not possible to blind the study [17]. Therefore, further trials are needed to assess the benefits of early vs. delayed introduction of anti-TNF, in both CD and UC. Beyond clinical issues, the steady increase in new prescriptions of anti-TNF, as documented in the present study, calls for cost-effectiveness analysis of the various treatment options in a long time horizon [19] and [20].

Regarding the use of other IBD-related medications, our analysis showed a high rate of salicylate use among CD patients, which is no longer recommended by ECCO [21] consensus guidelines. Similar results have also been reported in the US population [15]. In contrast, budesonide was prescribed much more frequently than rectal steroids in CD, and the reverse was true in UC, in line with ECCO guidelines [21] and [22].

The main strength of this study is that it was conducted on two nationwide, comprehensive databases providing complementary data, prospectively and independently collected. We also used identical algorithms to describe anti-TNF use in 2011, 2012, and 2013, allowing evaluation of the changing trends in the conditions of prescriptions of anti-TNF over this period.

This study has a number of limitations inherent to the use of health claims databases. Diagnoses cannot be formally confirmed and are based on information encoded for reimbursement purposes. Some misclassifications between CD and UC diagnoses may have occurred, and a substantial number of patients could not be classified. We also had to develop an algorithm to identify CD- or UC-related hospitalisations, but which cannot be formally validated. The lack of perspective in the database also prevented us from identifying patients who had never been treated with IS. The proportion of patients not treated with IS during the previous 12 months was considered to be a proxy to investigate the trend in first-line use of anti-TNF over the study period.

In conclusion, this study shows a rapid increase in new prescriptions of anti-TNF, for both CD and UC, in France between 2011 and 2013. Our results suggest a change in medical practices, with anti-TNF agents prescribed more often as first-line maintenance treatment, instead of thiopurines.

Conflict of interest

Franck Carbonnel has served on Advisory Board for Genentech, Otsuka, Vifor, enterome, MSD (oncology), Ferring (2015), has received lecture fees from Abbvie, Mayoly Spindler, Takeda, Hospira, Pfizer. Laurent Peyrin-Biroulet has received consulting fees from Merck, Abbvie, Janssen, Genentech, Ferring, Norgine, Tillots, Vifor, Shire, Therakos, Pharmacosmos, Pilège, BMS, UCB-Pharma, Hospira, Celltrion, Takeda, Biogaran, Boerhinger-Ingelheim, Lilly, Pfizer, and HAC-Pharma; and lecture fees from Merck, Abbvie, Takeda, Janssen Cilag, Ferring, Norgine, Tillots, Vifor, Therakos, HAC-Pharma, and Mitsubishi. The other authors declare no competing interest.

Appendix A. Supplementary data

The following are the supplementary data to this article:

References

  • [1] S.B. Hanauer, B.G. Feagan, G.R. Lichtenstein, et al. Maintenance infliximab for Crohn's disease: the ACCENT I randomised trial. Lancet. 2002;359:1541-1549
  • [2] B.E. Sands, F.H. Anderson, C.N. Bernstein, et al. Infliximab maintenance therapy for fistulizing Crohn's disease. New England Journal of Medicine. 2004;350:876-885
  • [3] S.B. Hanauer, W.J. Sandborn, P. Rutgeerts, et al. Human anti-tumor necrosis factor monoclonal antibody (adalimumab) in Crohn's disease: the CLASSIC-I trial. Gastroenterology. 2006;130:323-333
  • [4] W.J. Sandborn, S.B. Hanauer, P. Rutgeerts, et al. Adalimumab for maintenance treatment of Crohn's disease: results of the CLASSIC II trial. Gut. 2007;56:1232-1239
  • [5] J.F. Colombel, W.J. Sandborn, P. Rutgeerts, et al. Adalimumab for maintenance of clinical response and remission in patients with Crohn's disease: the CHARM trial. Gastroenterology. 2007;132:52-65
  • [6] P. Rutgeerts, W.J. Sandborn, B.G. Feagan, et al. Infliximab for induction and maintenance therapy for ulcerative colitis. New England Journal of Medicine. 2005;353:2462-2476
  • [7] W. Reinisch, W.J. Sandborn, D.W. Hommes, et al. Adalimumab for induction of clinical remission in moderately to severely active ulcerative colitis: results of a randomised controlled trial. Gut. 2011;60:780-787
  • [8] European Medicine Agency, Summary of product characteristics, Remicade. http://www.ema.europa.eu/ema/index.jsp?curl=pages/medicines/human/medicines/000240/human_med_001023.jsp&mid=WC0b01ac058001d124.
  • [9] European Medicine Agency, Summary of product characteristics, Humira. http://www.ema.europa.eu/docs/en_GB/document_library/EPAR_-_Product_Information/human/000481/WC500050870.pdf.
  • [10] P. Tuppin, L. de Roquefeuil, A. Weill, et al. French national health insurance information system and the permanent beneficiaries sample. Revue d Epidemiologie et de Sante Publique. 2010;58:286-290
  • [11] G. Moulis, M. Lapeyre-Mestre, A. Palmaro, et al. French health insurance databases: what interest for medical research?. Revue de Medecine Interne. 2015;36:411-417
  • [12] G. Rey, E. Jougla, A. Fouillet, et al. Deprivation index, France ecological association between a deprivation index and mortality in France over the period 1997–2001: variations with spatial scale, degree of urbanicity, age, gender and cause of death. BMC Public Health. 2009;22:33
  • [13] J. Cosnes, C. Gower-Rousseau, P. Seksik, et al. Epidemiology and natural history of inflammatory bowel diseases. Gastroenterology. 2011;140:1785-1794
  • [14] C. Gower-Rousseau, F. Vasseur, M. Fumery, et al. Epidemiology of inflammatory bowel diseases: new insights from a French population-based registry (EPIMAD). Digestive and Liver Disease. 2013;45:89-94
  • [15] W.K. van Deen, M.G. van Oijen, K.D. Myers, et al. A nationwide 2010–2012 analysis of U.S. health care utilization in inflammatory bowel diseases. Inflammatory Bowel Diseases. 2014;20:1747-1753
  • [16] J.F. Colombel, W.J. Sandborn, W. Reinisch, et al. Infliximab, azathioprine, or combination therapy for Crohn's disease. New England Journal of Medicine. 2010;362:1383-1395
  • [17] R. Khanna, B. Bressler, B.G. Levesque, et al. Early combined immunosuppression for the management of Crohn's disease (REACT): a cluster randomised controlled trial. Lancet. 2015; 10.1016/S0140-6736(15)00068-9
  • [18] R. Panaccione, S. Ghosh, S. Middleton, et al. Combination therapy with infliximab and azathioprine is superior to monotherapy with either agent in ulcerative colitis. Gastroenterology. 2014;146:392-400
  • [19] J. Dretzke, R. Edlin, J. Round, et al. A systematic review and economic evaluation of the use of tumour necrosis factor-alpha (TNF-α) inhibitors, adalimumab and infliximab, for Crohn's disease. Health Technology Assessment. 2011;15:1-244
  • [20] M. Marchetti, N.L. Liberato. Biological therapies in Crohn's disease: are they cost-effective? A critical appraisal of model-based analyses. Expert Review of Pharmacoeconomics & Outcomes Research. 2014;14:815-824
  • [21] A. Dignass, G. Van Assche, J.O. Lindsay, et al. The second European evidence-based consensus on the diagnosis and management of Crohn's disease: current management. Journal of Crohn's and Colitis. 2010;4:28-62
  • [22] A. Dignass, J.O. Lindsay, A. Sturm, et al. Second European evidence-based consensus on the diagnosis and management of ulcerative colitis. Part 2: Current management. Journal of Crohn's and Colitis. 2012;6:991-1030

Footnotes

a French National Health Insurance (CNAMTS), Paris, France

b Université Paris Sud and Gastroenterology Unit, Hôpitaux Universitaires Paris Sud, CHU de Bicêtre, AP-HP, Le Kremlin Bicêtre, France

c Inserm U954, Department of Hepato-Gastroenterology, University Hospital of Nancy, Lorraine University, Vandoeuvre-Les-Nancy, France

Corresponding author at: Caisse Nationale de l’Assurance Maladie (CNAMTS), 50, avenue du Professeur Andre Lemierre, 75986 Paris Cedex 20, France. Tel.: +33 1 72601721; fax: +33 1 72601724.

1 These authors contributed equally to this work.