OUP user menu

Journal of Crohn's and Colitis: 10 (10)


Laurence J. Egan, Ireland

Associate Editors

Shomron Ben-Horin, IsraelSilvio Danese, ItalyPeter Lakatos, HungaryMiles Parkes, UKJesús Rivera-Nieves, USABritta Siegmund, GermanyGijs van den Brink, NLSéverine Vermeire, Belgium


Published on behalf of

An economic evaluation comparing concomitant oral and topical mesalazine versus oral mesalazine alone in mild-to-moderately active ulcerative colitis based on results from randomised controlled trial

Mark P. Connolly, Sandy K. Nielsen, Craig J. Currie, Philippe Marteau, Chris S.J. Probert, Simon P.L. Travis
DOI: http://dx.doi.org/10.1016/j.crohns.2009.02.005 168-174 First published online: 1 September 2009


Introduction: A previous randomised controlled trial has demonstrated that oral plus topical mesalazine enema is more effective than oral mesalazine alone for achieving clinical remission in mild-to-moderately active extensive ulcerative colitis (UC). To evaluate whether this strategy is cost-effective we conducted an economic evaluation comparing 1 g topical mesalazine in combination with 4 g oral mesalazine compared to 4 g mesalazine monotherapy in mild-to-moderately active UC.

Methods: The economic evaluation was based on the ability to achieve remission using changes from baseline in the ulcerative colitis disease activity instrument (UCDAI). A cost-utility analysis was used where the main outcome was quality-adjusted life years to reflect improved quality of life associated with achieving remission compared with active disease. A simulated Markov model with five health states was constructed to model cost and outcome changes over time: (1) active UC; (2) mesalazine-refractory active UC; (3) steroid-refractory active UC; (4) infliximab-responsive active UC; and (5) remission. To reflect parameter uncertainty in the cost-effectiveness analysis probabilistic sensitivity analysis (PSA) was conducted by varying relevant clinical parameters.

Results: Average treatment costs required to transition a patient from active UC to remission using oral and topical mesalazine compared with oral alone were £1812 and £2390, respectively. Improved remission rates attributed to oral and topical mesalazine resulted in moderate improvements in quality-adjusted life years (QALYs) compared to oral mesalazine alone. Disaggregation of medical costs indicated that medical consultations and diagnostic costs were similar for both treatment arms. An abbreviated analysis which considered costs up to steroid-refractory patients in subacute UC indicated that combination therapy offered a cost-savings of £285 over 16 weeks of therapy compared with monotherapy.

Conclusions: The results indicate that the addition of 1 g topical mesalazine results in significant cost-savings and moderate quality of life improvements. We have also shown that irrespective of which treatment modality is used in steroid-refractory patients (eg, infliximab, azathioprine, ciclosporine) that topical mesalazine is cost-saving.

  • Cost
  • Economic evaluation
  • Ulcerative colitis
  • Cost-effectiveness analysis
  • Mesalazine
  • Cost-utility analysis

1 Background

Ulcerative colitis (UC) is an inflammatory condition of the colon characterised by periods of remission and acute relapses causing bloody diarrhoea, abdominal discomfort and fatigue.1,2 Symptoms associated with active UC significantly affect a person's quality of life (QoL) when compared to remission.3,4 UC can influence social activities and work productivity, which often results in lost earnings and in some cases influences choices that people make regarding their profession.46

Several studies have explored the costs associated with managing UC during remission and active disease. Annual treatment costs vary by disease severity and variations in treatment practice between countries.5,7,8 Inpatient costs account for the majority of expenditure in patients with UC and any hospitalisation for acute severe colitis (affecting 25% of all those with UC) is strongly associated with colectomy.5,8,9 Nevertheless, the large majority of patients with UC who relapse have mild or moderately active disease that requires outpatient therapy. Because active disease is associated with increased healthcare use and puts a considerable burden on the quality of life, maintaining patients in remission or the rapid transition from active UC to remission would be expected to improve QoL and save healthcare resources.10

One approach to enhancing outpatient remission rates is the addition of 1 g topical mesalazine enema to oral 4 g mesalazine for mild-to-moderately active UC patients.11 The question is whether this is a cost-effective strategy. Consequently we conducted an economic evaluation that considered a broad range of costs and outcomes for the management of active UC. To reflect the quality of life impact attributed to acute UC we then used a cost-utility analysis to assess the quality of life impact attributed to active UC, taking account of health-related quality of life (HRQoL) for time spent in different UC health states in order to derive quality-adjusted life years (QALYs).

2 Methods

2.1 Clinical data

Clinical data used to develop the model were derived from a previously reported randomised controlled trial.11 The primary outcome measure in the study reported by Marteau et al. was to establish whether adding topical 1 g/100 ml mesalazine enema to 4 g oral mesalazine was more effective for inducing remission compared to 4 g mesalazine alone. Remission was defined by a score of less than 2 on the Ulcerative Colitis Disease Activity Index (UCDAI) and was evaluated at 2, 4, and 6 weeks. This is a robust definition of remission and in practical terms means that at least three of the four measures of stool frequency, rectal bleeding, endoscopy and physician's global assessment must be normal and only one of the four may be mildly abnormal.12,13 Patients with mild-to-moderate exacerbations (UCDAI score of 3–8) of extensive UC were recruited and subjects were required not to have been on oral or topical steroids for four weeks prior to enrolment. Patient demographics were previously reported and there were no observed differences between the treatment groups.11 The randomised clinical study also collected health-related quality of life (HRQoL) assessment data using the EQ-5D instrument to measure patient preference for different UC health states.14 The clinical study was conducted according to good clinical practice (GCP) standards and the protocol was approved by an Independent Ethics Committee (IEC) or Institutional Review Board (IRB) at each centre.

2.2 Model design

A simulated Markov model was constructed to perform a cost-utility analysis comparing oral plus topical mesalazine, with oral mesalazine alone.15 The model was constructed using Treeage Pro 2008 (Treeage Software, Inc. Williamstown, MA, USA). The model consisted of five health states: (1) active UC; (2) mesalazine-refractory active UC; (3) steroid-refractory active UC; (4) infliximab-responsive active UC; and (5) remission. The cycle length used to represent the amount of time spent in each health state was 8 weeks. The 8 week cycle duration is implicit in the transition probabilities between health states based on reported treatment success from clinical trials between health states. Furthermore, within the modeled cohort, patients did not progress to surgery which reflects the mild-to-moderate nature of the population of outpatients with active UC recruited in the randomised study. The 8 week Markov cycle length is consistent with European guidelines for managing acute UC.16 The transition probabilities between health states and supporting references are provided in Table 1.

View this table:
Table 1

Treatment success rates and mean health state utility values with variances and distributions applied in probabilistic sensitivity analysis (PSA).

Model parameterTransition probability and varianceaDistributionReference
Remission with oral 4 g mesalazine and mesalazine enemab0.64 (0.50–0.76)Beta 11
Remission with oral 4 g mesalazine and placebo enemab0.43 (0.28–0.58)Beta 11
Probability success with prednisoloneb0.68 (0.43–0.87)Beta 18
Probability success with infliximabb0.39 (0.30–0.48)Beta 19
GP consultation rates in remission2.2 per year (+/− 20%)Uniform 5
Utility values applied in health states
Health state utility active UC0.81 (0.79–0.84)Beta 20
Health state utility remission0.94 (0.92–0.97)Beta 20
  • a Variance defined by 95% confidence interval (CI).

  • b Based on reported outcomes at 8 weeks.

Recent cost-effectiveness guidance from the National Institute for Health and Clinical Excellence (NICE) in the UK currently does not recommend infliximab for the treatment of subacute manifestations of moderately to severely active UC.17 To reflect cost-effectiveness guidelines recommended by NICE we developed an abbreviated model that excluded infliximab and evaluated costs and outcomes up to the steroid-refractory health state. This approach allows us to compare the immediate cost and outcome effects of oral plus topical mesalazine irrespective of which treatment is adopted in steroid-refractory patients. The abbreviated model is labeled as ‘steroid model’ in the Results section.

2.3 Outcome measure

The outcome of interest in the economic evaluation was the difference in quality-adjusted life year (QALY) changes between the two treatment groups, based on the intention-to-treat population. QALYs were derived from the EQ-5D used to evaluate HRQoL for active UC and remission health states, which were weighted according to the period of time spent in each health state. Previous studies have demonstrated the EQ-5D to be a valid and reliable instrument for use in IBD patients.21 The health state utility scores derived using the EQ-5D from the PINCE study for active UC and remission were 0.81 and 0.94, respectively.20 The utility value for active UC was applied in health states 1, 2, 3, and 4 and the utility value for remission was applied in health state 5.

2.4 Cost data

The cost data included in the model were derived from published sources. Costs included in the economic evaluation are those related to pharmacological treatment, clinical consultations, and diagnostic examinations (eg, blood tests, stool samples, sigmoidoscopy). Consistent with clinical success probabilities the costs are based on 8 week treatment cycles. Costs related to scheduled follow up visits and protocol-driven procedures are not considered in the evaluation. Mesalazine costs were based on oral mesalazine (PENTASA, Ferring Pharmaceuticals, St Prex, Switzerland) and topical mesalazine (PENTASA, Ferring Pharmaceuticals, St Prex, Switzerland) obtained from the British National Formulary (May 2008).

Costs attributed to active UC treatment are based on treatment guidelines published by the British Society for Gastroenterology and the European Crohn's and Colitis Organisation (ECCO).16,22 In brief, it assumed that patients unresponsive to oral and topical mesalazine were treated for 8 weeks with a course of 40 mg tapered prednisolone in combination with 20 mg prednisolone enema. Steroid-refractory patients progressed to infliximab administered in an outpatient setting at weeks 0, 2, and 6 and then every 8 weeks until model termination. Costs attributable to colectomy were not considered because the study cohort had mild-to-moderate activity and no surgery was reported in the clinical population upon which the model was based. The perspective in the analysis was the UK National Health Service (NHS) and all costs are presented in Pounds Sterling (£; indexed to 2008 values). A summary of the relevant cost variables included in the model is provided in Table 2.

View this table:
Table 2

Healthcare resource items included in economic evaluation.

Resource itemCostReference and notes
Mesalazine 1 g retention enemas in 100 mL pack£18.09BNF, net price 7 enemas
Mesalazine£2.40 per dayBNF, 1 g, 50-sachet pack costs £30.02
Prednisolone oral 40 mg per day (tapering course over 8 weeks)£4.79BNF, 5 mg, 28-tablets, 60p.
Prednisolone enema 20 mg per day£0.71BNF, 20 mg, 100-mL single-dose (standard tube)
Infliximab£419.62BNF, price per vial. Dosed at 5 mg/kg based on 75 kg individual. Administered as day case.
Gastroenterologist consultation£132Cost per visit. 23
GP consultations£34Cost per visit. 23
Flexible sigmoidoscopy£297UK Department of Health, NHS National Tariff (Payment by Results)
Stool sample£51.15 http://www.doctornow.org/
C-Reactive Protein (CRP)£9.00Plymouth Hospital NHS Trust private medical tariff
Full blood count£3.00NHS National Tariff
Microbiological testing£7.00NHS National Tariff

BNF, British National Formulary.

    2.5 Sensitivity analysis

    To reflect variation in model parameters we performed an analysis using probabilistic sensitivity analysis (PSA). The PSA approach is recommended by national technology appraisal organizations such as NICE because it allows for simultaneous sampling with a defined distribution for those parameters where uncertainty is likely to exist.24,25 Parameters included in the PSA included efficacy values most likely to influence outcomes and costs in the model (Table 1). The variance and distribution for each parameter are described.

    To supplement the incremental cost-effectiveness analysis and better reflect uncertainty in the stochastic cost-effectiveness analysis, we calculated the net health benefit statistic. Estimation of the net health benefit statistic permits graphical presentation of the proportion of iterations in which each product is more cost-effective relative to the other.26 Results are presented as a cost-effectiveness acceptability curve (CEAC), which expresses the proportion of simulations for each strategy that generates the maximum net health benefit across a range of willingness-to-pay thresholds. The product with the highest net benefit at any willingness-to-pay threshold would be considered the optimal treatment strategy. UK norms were applied for maximum willingness-to-pay per QALY of £20,000.27,28

    3 Results

    Two sets of cost-effectiveness results are presented: (1) the base model where infliximab is used to treat steroid-refractory patients and (2) an abbreviated model which excludes infliximab costs and only considers costs up to the point of patients failing steroid therapy for induction of active UC.

    Cost and outcome results from the base economic evaluation based on the expected-value analysis and probabilistic analysis comparing oral and topical mesalazine against oral mesalazine alone are presented in Table 3. Average treatment costs required to transition a patient with mild-to-moderately active UC into remission using 4 g oral mesalazine in combination with 1 g topical mesalazine compared with 4 g oral mesalazine monotherapy were £1812 and £2390, respectively (Table 3). The addition of 1 g topical mesalazine for treating patients with active UC resulted in a cost-saving of £578 per patient treated, in spite of the higher drug costs. There was also a modest improvement in QALYs during the treatment period. Similar treatment costs and outcome differences were found in the expected-value analysis. The incremental analysis indicated that combination of oral and topical mesalazine was the dominant treatment strategy, because of observed cost-savings and improved patient-relevant outcomes.

    View this table:
    Table 3

    Average cost and QALYs for subjects treated with oral and topical mesalazine or oral mesalazine alone after 32 weeks of treatment (4 cycles) based on probabilistic sensitivity analysis (PSA) and expected-value analyses.

    Oral mesalazine with mesalazine enemaOral mesalazine with placebo enema
    PSA output a
    Average treatment costs£1812£2390
    95% CI£1226–£2614£1501–£3557
    Average cost-effectiveness ratios£3245£4320
    95% CI£2174–£4727£2694–£6534
    Expected-value output
    Average treatment costs£1813£2388
    Incremental cost (mesalazine–placebo)(£575)
    Incremental QALYs (mesalazine–placebo)0.01
    Average cost-effectiveness ratios£3245£4328
    Incremental cost-effectiveness ratioDominant

    SD, standard deviation.

    • a Based on Monte Carlo simulation derived from 10,000 samplings.

    Results found using the ‘steroid model’ based on 16 weeks treatment which excludes infliximab costs are presented in Table 4. Average treatment costs required to transition a patient with mild-to-moderately active UC into remission using 4 g oral mesalazine in combination with 1 g topical mesalazine compared with 4 g oral mesalazine alone were £1114 and £1399, respectively. In addition to the small cost-savings observed with oral and topical mesalazine of £285 over 16 weeks, there was a small improvement in QALYs indicating this was the dominant treatment strategy. Similar results were obtained using the expected-value analysis and the PSA.

    View this table:
    Table 4

    Average cost and QALYs for subjects treated with oral and topical mesalazine or oral mesalazine alone in ‘steroid model’ after 16 weeks of treatment (2 cycles) based on probabilistic sensitivity analysis (PSA) and expected-value analyses.

    Oral mesalazine with mesalazine enemaOral mesalazine with placebo enema
    PSA output a
    Average treatment costs£1114£1399
    95% CI£862–£1458£1033–£1875
    95% CI0.266–0.2770.260–0.273
    Average cost-effectiveness ratios£4106£5251
    95% CI£3141–£5435£3827–£7106
    Expected-value output
    Average treatment costs£1114£1399
    Incremental cost (mesalazine–placebo)(£285)
    Incremental QALYs (mesalazine–placebo)0.01
    Average cost-effectiveness ratios£4102£5245
    Incremental cost-effectiveness ratioDominant

    SD, standard deviation.

    • a Based on Monte Carlo simulation derived from 10,000 samplings.

    The disaggregated average treatment costs observed in the base model are presented in Table 5. In both treatment arms, similar percentages of costs were allocated to medical consultations and diagnostic examinations, but there were small differences in the absolute costs observed for each treatment approach. The modeled cost data indicate that the majority of cost-savings achieved with combination of oral and topical mesalazine were achieved from reduced use of infliximab therapy for refractory disease average infliximab costs were estimated at £759 and £1201 for combination and oral monotherapy, respectively. Mesalazine costs were higher for patients on combination of oral and topical therapy due to higher mesalazine dosage and higher rates of remission that would result in increased use of mesalazine prophylactically while in remission.

    View this table:
    Table 5

    Disaggregated treatment costs required to treat extensive, mild-moderately active ulcerative colitis (32 weeks) in base model.

    Healthcare resourceOral mesalazine with mesalazine enemaOral mesalazine with placebo enemaIncremental cost difference
    Average cost% of total costsAverage cost% of total costs
    Consultation costs£35520£42918(£74)
    Diagnostic exams (sigmoidoscopy, blood tests, CRP)£39421£50921(£115)
    Mesalazine and prednisolone costs£30717£25211£55

    4 Sensitivity analysis

    Results from the net health benefits analysis comparing benefits for both treatments over a range of willingness-to-pay (WTP) thresholds are presented using a cost-effectiveness acceptability curve (Fig. 1). The maximum WTP evaluated in the analysis was based on accepted UK norms for willingness-to-pay per QALY.27 Over the range of WTP thresholds, from £0 to £20,000 per QALY, the addition of a mesalazine enema consistently demonstrated a higher probability of being more cost-effective than mesalazine monotherapy. This is noted by the constant proportion of subjects with improved cost-effectiveness for those treated using topical mesalazine over all different WTP thresholds compared with those using placebo as demonstrated in Fig. 1.

    Figure 1

    Cost-effectiveness acceptability curve for combined oral and topical mesalazine compared with oral mesalazine based on willingness-to-pay (WTP) threshold from £0 to £20,000 in base case.

    5 Discussion

    Investigations have shown that the addition of a 1 g mesalazine enema improves remission rates by 50% (from 40% to 64%) in patients with pancolitis and mild-to-moderately active disease.11 This has clear therapeutic benefit and reduces the chance of such patients needing ‘rescue’ therapy with conventional steroids, or even infliximab. The current study has shown that a 1 g mesalazine enema in combination with 4 g oral mesalazine achieves cost-savings compared to oral mesalazine monotherapy, in spite of higher drug costs. This is encouraging. The cost-saving is achieved by significantly higher remission rates with combination therapy, which prevents progression to costly interventions.

    To reflect UK cost-effectiveness guidelines issued by NICE we tested the impact of an abbreviated analysis which excluded expensive downstream interventions frequently used in subacute patients. To understand whether topical mesalazine was cost-effective over the short-term, prior to infliximab therapy as defined by our model, we conducted an abbreviated analysis which stopped at 16 weeks. In the abbreviated ‘steroid model’ we show that combination therapy with 1 g topical mesalazine is cost-saving compared with 4 g oral monotherapy at 16 weeks. This is encouraging as it indicates the model does not rely on off-setting expensive interventions for combination of oral and topical mesalazine to be cost-saving. This also suggests that irrespective of whether prednisolone failures progress to infliximab or azathioprine (2 mg/kg daily) costing £1678 per course or £1.50 per day, respectively, use of combination mesalazine would remain cost-saving in either treatment setting.

    In our evaluation we allowed steroid-refractory patients to progress to infliximab therapy. This may not reflect clinical practice in all settings and certainly not for all patients. However, we believe that the alternative for refractory disease is colectomy before or after a period with refractory symptoms despite further steroids or hospital admission that would only add to the costs in the model. The decision to exclude surgery in our model correctly reflects the mild-to-moderate characteristics of the study cohort, as well as patient preferences for medical intervention rather than surgery.29 Had we included colectomy and associated costs, the conclusion that there was a higher treatment failure rate with oral mesalazine monotherapy would not have changed, but increased surgery and additional costs would have applied to this population.

    Increased rates of remission with combination of oral and topical mesalazine improves the quality of life measured by validated indices, which is reflected by the small increase in QALYs from oral and topical mesalazine compared with oral mesalazine alone. It is perhaps surprising that the relative crude measure of QALYs, designed to measure the impact of major therapeutic interventions, is sensitive to the simple addition of topical to oral therapy. The cost-saving and increased QALYs using 1 g topical indicates that mesalazine is the dominant treatment strategy.

    To allow for uncertainty in the cost-effectiveness analysis, we generated cost-effectiveness acceptability curves, using net health benefits.26 The results indicate that over a range of thresholds that UK decision-makers are willing-to-pay (WTP) per QALY, that combined oral and topical mesalazine for patients with mild or moderately active UC is the most cost-effective option. The probability of being cost-effective was continuously in favour of topical mesalazine over WTP thresholds ranging from £0 to £20,000.

    Consistent with guidelines for performing economic evaluations in the UK and internationally, we have constructed our economic model using results from a randomised controlled trial. Despite the strength of our clinical evidence on which to base an economic model, there can be uncertainty regarding the external validity of our results when viewed in the context of real world practice, where compliance can be problematic and treatment durations vary. We have not tested these known variants in the economic model because of lack of available data describing how poor compliance and varying treatment duration impacts outcomes. Without knowing how compliance and treatment duration impacts outcome, it is inappropriate to only adjust costs attributed to poor compliance and treatment duration because this would bias the analysis in favour of topical mesalazine (i.e. it would only reduce treatment costs without influencing the outcomes). Despite these potential weaknesses we are confident with our results, and emphasise that cost-effectiveness data is only one of many different types of data used by clinicians and decision-makers to guide treatment practices.30

    The costs presented here represent only those costs that arise from the point of treating active UC up to 32 weeks after treatment initiation or 16 weeks in the abbreviated model. Because our analysis is based on a cohort of subjects starting with active disease, the resulting costs are higher than previous studies reporting annual average treatment costs in patients with UC in all stages of activity, including remission. As a reference point, average annual treatment costs have been estimated at £1524 per patient-year.8 While it is not entirely possible to compare costing studies, the cohort described by Odes et al. is heterogeneous and includes subjects experiencing active disease and in remission over the course of one year. The cohort presented in our evaluation is more homogenous, since it only considers subjects with active disease which accounts for the increased costs observed in our study when compared to Odes et al.

    Our results indicate that concomitant treatment with 4 g oral mesalazine and 1 g topical mesalazine enemas results in significant cost-savings and moderate quality of life improvements compared with 4 g oral mesalazine alone. Cost-savings observed with oral and topical mesalazine were principally achieved by avoiding progression to more costly intervention. However, our abbreviated analysis which excluded expensive downstream costs in subacute UC still indicated concomitant mesalazine therapy is cost-saving. This work supports the adoption of topical mesalazine therapy as standard clinical practice for patients with mild or moderately active extensive UC and helps save scarce healthcare resources.


    • Disclosures. Sandy Nielsen is currently employed by Ferring Pharmaceuticals. Mark Connolly is a former employee of Ferring Pharmaceuticals. This study was funded by an independent grant from Ferring Pharmaceuticals St Prex, Switzerland. Professor Chris Probert, Dr Simon Travis, and Professor Marteau served as non-remunerated clinical advisors to the project and have received unrestricted educational grant support from, or acted as advisers to, or lecturers for Ferring Pharmaceuticals. Dr Craig Currie served as a non-remunerated health economics advisor for this project.


    1. [1].
    2. [2].
    3. [3].
    4. [4].
    5. [5].
    6. [6].
    7. [7].
    8. [8].
    9. [9].
    10. [10].
    11. [11].
    12. [12].
    13. [13].
    14. [14].
    15. [15].
    16. [16].
    17. [17].
    18. [18].
    19. [19].
    20. [20].
    21. [21].
    22. [22].
    23. [23].
    24. [24].
    25. [25].
    26. [26].
    27. [27].
    28. [28].
    29. [29].
    30. [30].
    View Abstract