Objective Evaluate the cost-effectiveness of laparoscopic ileocaecal resection compared with infliximab in patients with ileocaecal Crohn’s disease failing conventional therapy.
Design A multicentre randomised controlled trial was performed in 29 centres in The Netherlands and the UK. Adult patients with Crohn’s disease of the terminal ileum who failed >3 months of conventional immunomodulators or steroids without signs of critical strictures were randomised to laparoscopic ileocaecal resection or infliximab. Outcome measures included quality-adjusted life-years (QALYs) based on the EuroQol (EQ) 5D-3L Questionnaire and the Inflammatory Bowel Disease Questionnaire (IBDQ). Costs were measured from a societal perspective. Analyses were performed according to the intention-to-treat principle. Missing cost and effect data were imputed using multiple imputation. Cost-effectiveness planes and cost-effectiveness acceptability curves were estimated to show uncertainty.
Results In total, 143 patients were randomised. Mean Crohn’s disease total direct healthcare costs per patient at 1 year were lower in the resection group compared with the infliximab group (mean difference €−8931; 95% CI €−12 087 to €−5097). Total societal costs in the resection group were lower than in the infliximab group, however not statistically significant (mean difference €−5729, 95% CI €−10 606 to €172). The probability of resection being cost-effective compared with infliximab was 0.96 at a willingness to pay (WTP) of €0 per QALY gained and per point improvement in IBDQ Score. This probability increased to 0.98 at a WTP of €20 000/QALY gained and 0.99 at a WTP of €500/point of improvement in IBDQ Score.
Conclusion Laparoscopic ileocaecal resection is a cost-effective treatment option compared with infliximab.
Clinical trial registration number Dutch Trial Registry NTR1150; EudraCT number 2007-005042-20 (closed on 14 October 2015).
- Crohn’s disease
- laparoscopic ileocecal resection
- economic evaluation
Statistics from Altmetric.com
Significance of this study
What is already known on this subject?
In the majority of patients with ileocaecal Crohn’s disease who fail conventional therapy, treatment is intensified with biologics following a step-up approach.
The LIR!C Trial demonstrated that laparoscopic ileocaecal resection is a valid alternative to infliximab induction and maintenance therapy in terms of quality of life measured 1 year after start of treatment.
Infliximab treatment is associated with a considerable economic burden on society.
What are the new findings?
Laparoscopic ileocaecal resection is associated with significantly lower total Crohn’s disease related direct healthcare costs (€−8,931, 95% CI €−12 087 to €5097) compared with infliximab treatment.
Laparoscopic ileocaecal resection is cost-effective compared with infliximab.
As compared with biosimilars, societal costs of resection were non-significantly higher after 1 year. The break-even point after which biosimilars would become rapidly more expensive compared with laparoscopic ileocaecal resection was estimated at 14 months.
How might it impact on clinical practice in the foreseeable future?
Given the improved quality of life and reduced total societal costs after 1 year, laparoscopic ileocaecal resection should be considered a valid alternative to infliximab therapy in eligible patients with Crohn’s disease failing conventional therapy.
Crohn’s disease is a chronic IBD that can occur at any site in the GI tract and leads to lifelong morbidity.1 2 The incidence and prevalence of Crohn’s disease in Europe, 12.7/100 000/year and 322/100 000, respectively, have increased over the last decade.3 Treatment is focused on symptom control and intestinal mucosal healing in order to optimise quality of life, reduce disability, facilitate resumption of daily activities and prevent complications. Treatment consists primarily of medical therapy following a step-up approach ranging from budesonide and immunomodulators to biologic therapy.4 5 Despite a rapid expansion in the therapeutic armamentarium, to date a curative treatment is still lacking. Moreover, discontinuation of treatment is associated with high relapse rates.6–8 Hence, most patients with Crohn’s disease require lifelong therapy. The total annual economic burden of Crohn’s disease was estimated to be €16.7 billion in Europe.9 Since the introduction of biologics in the treatment of IBD, healthcare expenditures have shifted from hospitalisation and surgery to biologics as the dominant contributor to Crohn’s disease related healthcare costs.3 10 Surgical rates have decreased over the last decades.11 12 Nine years after diagnosis, up to 25% of patients with Crohn’s disease have undergone surgery.11 In a third of patients with Crohn’s disease, only the terminal ileum is affected.13 14 Patients failing conventional therapy are traditionally upscaled to antitumour necrosis factor therapy, such as infliximab. However, the LIR!C Trial showed that laparoscopic ileocaecal resection proved to be a valid alternative to infliximab with respect to improving quality of life.15 Since the use of infliximab is associated with a considerable economic burden on society, it was hypothesised that laparoscopic ileocaecal resection would result in considerably lower healthcare resource utilisation and costs, and thus societal costs, while achieving a higher quality of life.
Therefore, the aim of the present study was to evaluate the cost-effectiveness and cost utility of laparoscopic ileocaecal resection compared with 1 year induction and maintenance treatment with infliximab from a societal perspective.
This economic evaluation was conducted as part of the LIR!C Trial, a multicentre, randomised controlled 1:1 parallel group trial comparing laparoscopic ileocaecal resection with infliximab therapy for patients with Crohn’s terminal ileitis failing conventional therapy. This trial was conducted between 2 May 2008 and 14 October 2015 in 28 centres in the Netherlands and one in the UK. A protocol of this study was published and a detailed description of the trial design and clinical outcomes was published recently.15 16 In brief, adult patients with Crohn’s disease of the terminal ileum who failed >3 months of immunomodulator or steroid treatment without signs of a critical stricture were eligible for inclusion. Patients with a prior ileocaecal resection, an affected segment >40 cm, abdominal abscesses or fluid collections, or an American Society of Anesthesiologists Score of III/IV were excluded. The LIR!C Trial was powered on its primary clinical outcome measure, the Inflammatory Bowel Disease Questionnaire (IBDQ), assuming a clinically relevant between-group effect size of 0.5, which required a sample size of 143. All participants provided written informed consent. Monitoring of the trial was performed by an independent monitor from the Clinical Research Unit of the academic medical centre.
Economic evaluation consisted of cost-effectiveness analysis (CEA) and cost-utility analysis (CUA) from a societal perspective. In CEA, health effects are expressed as disease-specific outcomes, such as the IBDQ. CUA is a special type of CEA in which health effects of the medical intervention are expressed in quality-adjusted life-years (QALYs). Costs considered were Crohn’s disease related direct healthcare, other healthcare costs, travel costs and lost productivity costs.
Source of cost data
All costs were adjusted to €2014 using price indices. Crohn’s disease related direct healthcare costs included costs of the index treatment (resection or infliximab) as well as the costs of readmissions and re-interventions related to the index treatment. This information was obtained using patient diaries, supplemented by the patient’s medical charts. Other healthcare costs included costs related to other primary and secondary healthcare services and medication. Other healthcare costs as well as travel expenses were assessed using questionnaires. Dutch standard costs were used to value healthcare utilisation and travel costs.17 If unavailable, prices according to professional organisations were used. Costs per unit of resource use are presented in online supplementary table 1. Lost productivity costs (days of sick leave) were assessed by monthly telephone follow-up. The cumulative total number of sick leave days was calculated for the 12-month study period. Sick leave was valued according to the human capital approach, using sex- specific and age-specific wage rates for the Dutch population.17
Health-related quality of life was assessed using EQ-5D-3L at baseline, weeks 2 and 6 and months 3, 6, 9 and 12.18 19 EQ-5D-3L health states were converted to utility scores using the Dutch EQ-5D-3L tariff. Utility scores represent the relative desirability of a health state and are anchored at 0 (death) and 1 (full health). QALYs are calculated as the product-sum of health utilities and the lengths of the period between measurements during follow-up. Changes between follow-up measurements are considered to be linear.
In addition to the general quality of life questionnaires, we used the IBD-specific questionnaire for measuring quality of life: the IBDQ.20 21 It comprises 32 questions, each with a 1–7 scale, generating a total score of 32–224 points with higher scores indicating better quality of life.
Data are presented as mean and SD in case of continuous data and number and percentage in case of categorical and dichotomous data. All analyses were performed according to the intention-to-treat principle. Missing cost and effect data were imputed using multiple imputation by chained equations. A fully conditional imputation model was specified including variables that differed between groups at baseline, between patients with and without complete follow-up and that were associated with predefined outcomes (QALYs, IBDQ Score and societal costs). Predictive mean matching was used to account for the skewed distribution of the cost data, and 10 complete data sets were generated. Each data set was analysed separately as described below, after which results were pooled using Rubin’s rules.
Differences in costs and effects between laparoscopic ileocaecal resection and infliximab were estimated using bivariate regression analysis, which enabled correction for a possible correlation between costs and effects. Effect differences were corrected for baseline differences. Incremental cost-effectiveness ratios (ICERs) were calculated as the difference in costs divided by the difference in effects and indicate the incremental cost per additional unit of effect (QALY gained or point of improvement in IBDQ Score) of an intervention relative to the comparator. Statistical uncertainty around the cost and effect differences and the ICERs was estimated using bias-corrected accelerated bootstrapping with 5000 replications. Uncertainty around the ICERs was presented using cost-effectiveness planes (CE-planes) and cost-effectiveness acceptability curves (CEACs). In a CE-plane, the difference in costs between the intervention groups is plotted on the y axis and the difference in effects on the x axis, resulting in four quadrants: north-east (resection is more expensive and more effective than infliximab), south-east (resection is less expensive and more effective than infliximab, that is, resection is dominant over infliximab), south-west (resection is less expensive and less effective than infliximab) and north-west (resection is more expensive and less effective than infliximab, ie, resection is dominated by infliximab). The CEA curve shows the probability that laparoscopic ileocaecal resection is cost-effective in comparison with infliximab for willingness-to-pay (WTP) values between 0 and infinity. WTP is defined as the amount of money society is willing to pay to gain one unit of effect extra with the intervention relative to the comparator. In the UK, the National Institute for Health and Care Excellence (NICE) recommends a cost-effectiveness threshold of £20 000–30 000 per QALY gained (€22 756–38 028).22
In a first sensitivity analysis, the costs of infliximab (Remicade) were replaced by the much lower costs of a biosimilar (€17 543 for an average single infusion). The break-even point at which the costs of laparoscopic ileocaecal resection were equal to the biosimilar was calculated assuming that healthcare and lost productivity costs, and effects after the trial were similar to those during the trial. Moreover, no Crohn’s disease related direct healthcare costs were included after 1 year except for the costs of the biosimilar. In a second sensitivity analysis, analyses were stratified according to age of onset of disease. The cut-off was set at 40 years of age (A3 versus A2 or A1, conform to the Montreal classification). This was done because patients with a younger age of onset tend to have a more aggressive phenotype compared with an older age of onset. This might impact cost differences between both groups.
Demographics and baseline characteristics were described in detail previously.15 In brief, between 2 May 2008 and 14 October 2015, 143 patients (32.9% male) with a median age of 27 years (IQR 22–40) and median disease duration of 14 months were included (online supplementary figure 1). Seventy-three patients were randomly allocated to laparoscopic ileocaecal resection and 70 to infliximab. Baseline characteristics are presented in table 1. QALY and IBDQ data were complete for 130, 96, 96, 97, 81, 93 and 84 patients at baseline, weeks 2 and 6 and months 3, 6, 9 and 12, respectively. Information on Crohn’s disease related direct healthcare costs was complete for all patients. All other cost data were complete for 80, 62, 76, 60, 57 and 59 patients for months 1 and 2, months 3 and 4, months 5 and 6, months 7 and 8, months 9 and 10, and months 11 and 12, respectively.
Crohn’s disease related direct healthcare costs were statistically significantly lower in patients who underwent a laparoscopic ileocaecal resection in comparison to those who received infliximab (mean difference (MD) €−8931, 95% CI €−12 087 to €−5097), see table 2. Medication costs (MD €54 95% CI €36 to €77) and absenteeism costs (MD €2596 95% CI €579 to €4719) were statistically significantly higher in the resection group compared with the infliximab group (see table 3). Primary care costs and total societal costs were lower in the resection group compared with the infliximab group, although these differences did not reach statistical significance.
The course of the utility scores over 12 months is shown in figure 1. The average utility score in the laparoscopic ileocaecal resection group was lower compared with the infliximab group during the first weeks of the trial. However, after 13 weeks the utility scores in the resection group were higher than in the infliximab group. The difference in mean utility scores between the two groups reached a maximum of 0.10 after 26 weeks and remained relatively stable after that. The mean number of QALYs after 12 months in the laparoscopic ileocaecal resection group was statistically significantly larger than in the infliximab group (0.81 vs 0.74, MD 0.07, 95% CI (0.01 to 0.12)).
After 12 months, the mean IBDQ Score was 180.8 in the resection group and 173.5 in the infliximab group, which resulted in a statistically non-significant difference between the groups (adjusted mean difference 6.3 (95% CI −1 to 13.4)).
CEA and CUA
Table 4 shows the results of the CEAs. The ICER for QALYs was −85 802, indicating that laparoscopic ileocaecal resection was dominant over infliximab. In the CE-plane, the majority of the cost-effect pairs (95.1%) was located in the south-east quadrant confirming that laparoscopic ileocaecal resection was on average less costly and more effective than infliximab (figure 2A). The CEACs showed that the probability of laparoscopic ileocaecal resection being cost-effective in comparison with infliximab was 0.96 at a WTP of €0/QALY gained. This probability increased to 0.98 at a WTP of €20 000/QALY gained (figure 3A).
The ICER for the IBDQ was −5729, indicating that laparoscopic ileocaecal resection was dominant over infliximab. Again, the majority of cost-effect pairs (91.5%) was located in the south-east quadrant (figure 2B). The probability of a laparoscopic ileocaecal resection being cost-effective in comparison with infliximab was 0.96 at a WTP of €0/point of improvement and increased to 0.99 at a WTP of €500/point of improvement (figure 3B).
Sensitivity analyses with biosimilars and stratified according to disease age of onset
Using the price of a biosimilar changed the results of the CEA considerably. Although Crohn’s disease related direct healthcare costs were still lower in the laparoscopic ileocaecal resection group as compared with the infliximab group, this difference was not statistically significant anymore (MD €−1938, 95% CI €−5036 to €1160). Moreover, the difference in total societal costs turned around—with higher costs in the resection group compared with the biosimilar group—although this did not reach statistical significance (table 4). The majority (63.8%) of cost-effect pairs was located in the north-east quadrant (ie, resection is more costly and more effective than the biosimilar) for both QALYs and IBDQ (figure 2C,D). For both outcomes, the probability of cost-effectiveness was 0.35 at a WTP of €0/additional unit of effect, after which it gradually increased with increasing values of WTP (eg, to 0.82 and 0.84 at a WTP of €80000/QALY gained and €1000/point of improvement on the IBDQ, respectively) (figure 3C,D). The break-even point in societal costs at which biosimilars would become rapidly more expensive than resection was reached at approximately 14 months after the start of treatment with the biosimilar.
In a secondary sensitivity analysis we examined the effect of age of disease onset on differences in cost-effectiveness between both treatment strategies. Even though the point estimates of the cost and effect differences, as well as that of the ICERs, slightly differed across the stratified analyses, the overall conclusion of the current study would not change when solely looking at patients with a disease onset <40 years or ≥40 years of age (online supplementary table 2).
This economic evaluation showed that laparoscopic ileocaecal resection is cost-effective compared with infliximab therapy for patients with Crohn’s disease of the terminal ileum failing conventional therapy. This is an important finding for clinical practice, because with the introduction of biologics, there has been a major shift in IBD-related direct healthcare costs from hospitalisation and surgery towards medical treatment with biologics.10 23 Surgery and the biologic infliximab were directly compared in a prospective fashion including both direct (healthcare-related) and indirect (lost productivity) costs. Surgery may be expected to lead to increased direct costs due to postoperative complications, which prolong hospitalisation and potentially lead to repeated surgery. Nevertheless, Crohn’s disease related direct healthcare costs were significantly lower in the resection group compared with the infliximab group. This was despite the occurrence of a serious adverse event (Dindo-Clavien classification ≥IIIa) in four patients (5.7%) in the resection group.24 Two were transferred to the intensive care unit and three received a temporary ileostomy. This was in accordance with the number recently reported by the REMIND Study group.25 Absenteeism costs were significantly higher in the resection group as compared with the infliximab group. After 2 months however, absenteeism was similar in both groups.
Patients in the resection group experienced significantly more QALYs based on EQ-5D-3L over 12 months compared with patients in the infliximab group. At 1 year, quality of life, expressed in terms of utilities, was significantly better in the resection group compared with the infliximab group (0.88 vs 0.73). This is in line with the previously reported effect analyses that showed that patients had significantly better general quality of life (physical Short Form 36 (SF-36) Scale).15 No statistically significant difference was observed in disease-specific quality of life as measured by IBDQ. Of note, in the current trial-based economic evaluation, due to the differences in applied analytical methods, the adjusted mean difference of IBDQ at 12 months between both groups differed slightly from that reported in the primary effect analyses (MD 6.3 95% CI −1 to 13.4 vs MD 6.1 95% CI −4.2 to 16.4, respectively). Differences include multiple imputation for dealing with missing cost and effect data, instead of maximum likelihood estimation, and a correction for the possible correlation between costs and effects by bivariate regression analyses. Both methods are recommended in economic evaluations, even though they may not necessarily be recommended for efficacy studies, due to the distinct nature of cost and effect data.26–28 Although this trial evaluated cost-effectiveness at 1 year, data regarding the evolution of costs beyond this time point would be valuable. In the UK, NICE guidelines advise discontinuation of infliximab treatment after 1 year unless there is clear evidence of ongoing active Crohn’s disease.29 Conversely, the European Crohn’s and Colitis Organisation does not provide specific guidance on duration of infliximab treatment, mainly due to high relapse rates after discontinuation.5 6 8 Therefore, in most European countries, patients with Crohn’s disease who are started on infliximab require long-term treatment. In this line of thought, a further continuous increase in total costs may be expected for the infliximab-treated patients. Conversely, cost evolution in the resection group is difficult to predict. Surgery might also be associated with increasing costs due to recurrence rates following the first year after surgery. The collection and analysis of long-term follow-up data with regard to time to disease recurrence requiring a therapeutic intervention is currently underway. Nevertheless, the observed difference in healthcare expenditure between both groups will most likely further increase over a long-term horizon.
Previous economic evaluations of biologics in IBD—which mainly compared biologics to conventional treatment—have debated the cost-effectiveness of long-term infliximab maintenance therapy.30 Even though a high WTP of $100 000/QALY (±€74 112/QALY) was used, only two out of six included studies concluded that after 1 year infliximab maintenance therapy was cost-effective as compared with conventional treatment. In another four studies assessing budget impact, infliximab was cost saving in only one study.30 For this study, we focused on the frequent situation in which there is failure on conventional therapy and treatment needs to be upscaled.
A few limitations of this trial should be acknowledged. First, the inclusion duration of 7 years might decrease the external validity of our results due to the putative risk of selection bias. However, the demographic characteristics of the study subjects was quite similar to those of a recent Dutch population-based cohort.14 Second, this trial was initiated in 2008. Hence, less expensive biosimilars (eg, Remsima, Inflectra or Flixabi) were not yet available. Recently, a budget impact model for the introduction of the biosimilar Remsima (assumption of 10%–30% lower list price than the originator) in five European countries with a 1-year horizon predicted cumulative cost savings of €26–77 million.31 In the present study, a sensitivity analysis with biosimilars—with a considerable lower market price of 40%–60%—showed that the difference in societal costs turned around, with resection now being more costly than the biosimilar, however not significantly. The break-even point, however, after which the biosimilar would become rapidly more expensive, was only 2 months following the first year of follow-up. As discussed previously, treatment with biologics is commonly maintained over a long-term horizon. Therefore, we expect that resection leads to considerable long-term cost savings. Finally, this economic evaluation was performed using Dutch standard pricing to calculate costs. Even though the Dutch healthcare system is comparable to most Western-European countries, infliximab reimbursement criteria and costs may vary considerably.32 Hence, obtained results might not be generalisable to all countries.
In conclusion, this multicentre randomised trial showed that in patients with ileocaecal Crohn’s disease in whom conventional therapy has failed, laparoscopic ileocaecal resection is associated with significantly more QALYs and higher IBDQ scores than infliximab at substantially lower costs. Therefore, laparoscopic ileocaecal resection is considered to be more cost-effective as compared with infliximab treatment in this group of patients.
EJG and TWS contributed equally.
WAB and CYP contributed equally.
Contributors Guarantor of the article: EJdG and TS. Development of study concept and design: EJE, PCFS, WAB and CYP. Acquisition of data: EJdG, EJE, TJG and BM collected the data. Analysis and interpretation of data: EJdG, TS, JEB, HD, CB, WAB and CYP. Statistical analysis: JEB and HD. Drafting of manuscript: EJdG, TS, JEB, HD, WAB and CYP. Critical revision of the manuscript for important intellectual content: all authors.
Funding This study was financially supported by a grant from The Netherlands Organisation for Health Research and Development. The funder had no role in the conduct and interpretation of the trial.
Competing interests EJdG, TS, JEB, HvD, EJE, TJG, BM, CB, PCFS and WAB declare no conflict of interest. CYP declares a grant from Takeda, Dr Falk Pharma and Abbvie, advisory board fees from Takeda and GSK, as well as speaker’s fees from Takeda, Abbvie, Ferring, MSD and Dr. Falk Pharma. GRAMD’H has served as advisor for Abbvie, Ablynx, Amakem, AM Pharma, Avaxia, Biogen, Bristol Meiers Squibb, Boerhinger Ingelheim, Celgene, Celltrion, Cosmo, Covidien/Medtronics, Ferring, DrFalk Pharma, Engene, Galapagos, Genentech/Roche, Gilead, Glaxo Smith Kline, Hospira, Immunic, Johnson and Johnson, Lycera, Medimetrics, Millenium/Takeda, Mitsubishi Pharma, Merck Sharp Dome, Mundipharma, Novonordisk, Otsuka, Pfizer, Prometheus laboratories/Nestle, Protagonist, Receptos, Robarts Clinical Trials, Salix, Sandoz, Setpoint, Shire, Teva, Tigenix, Tillotts, Topivert, Versant and Vifor and received speaker fees from Abbvie, Biogen, Ferring, Johnson and Johnson, Merck Sharp Dome, Mundipharma, Norgine, Pfizer, Shire, Millenium/Takeda, Tillotts and Vifor. AH declares an advisory board fee from MSD.
Patient consent Not required.
Ethics approval The medical ethics committee at the Amsterdam UMC, location Meibergdreef 9, Amsterdam, Netherlands and the corresponding committees in all participating centres in the Netherlands and the UK.
Provenance and peer review Not commissioned; externally peer reviewed.
Collaborators LIRIC study group: Janindra Warusavitarne; Ad A Van Bodegraven; Menno A Brink; Esther CJ Consten; Bart A Van Wagensveld; Marno CM Rijk; Rogier MPH Crolla; Casper G Noomen; Alexander PJ Houdijk; Rosalie C Mallant; Maarten Boom; Willem A Marsman; Hein B Stockmann; A Jeroen De Groof; Karlien Bruin; John Maring; Theo van Ditzhuijsen; Hubert Prins; Jan van den Brande; Paul Kingma; Anna van Geloven; Nanne de Boer; Donald van der Peet; Jeroen Jansen; Michael Gerhards; Janneke van der Woude; Ruud Schouten; Bas Oldenburg; Richard van Hillegersberg; Rachel West; Guido Mannaerts; Marcel Spanier; Ernst Jan Spillenaar Bilgen; Rob Lieverse; Edwin van der Zaag; Annekatrien Depla; Arnold van de Laar; Clemens Bolwerk; Hans Brouwer; Nofel Mahmmod; Eric Hazebroek; Juda Vecht; Robert Pierik; Gerard Dijkstra; Sijbrand Hofker; Tessa Uiterwaal; Quirijn Eijsbouts; Liekele Oostenbrug; Meindert Sosef; Djuna Cahen; Sjoerd van der Werff; Andreas Marinelli; Job Peters; Huib Cense; Nynke Talstra; Pritesh Morar.
Correction notice This article has been corrected since it published Online First. The author names have been corrected.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.