|
The cost-effectiveness of infliximab for severe treatment resistant rheumatoid arthritis in the UK |
Barbieri M, Wong J B, Drummond M |
|
|
Record Status This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn. Health technology The use of infliximab (INF) plus methotrexate (MTX) for the treatment of patients with rheumatoid arthritis (RA). INF (Remicade) is a new anti-tumour necrosis factor-alpha therapy that has recently been approved for the treatment of severe treatment-resistant RA. The dosages of INF used were 3 or 10 mg/kg for 4 or 8 weeks.
Study population The study population comprised a hypothetical cohort of patients with RA who were not adequately controlled on disease-modifying antirheumatic drugs (DMARDs) and who were resistant to MTX.
Setting The setting was secondary care. The economic study was carried out in the UK.
Dates to which data relate The effectiveness and resource use data were derived from studies published between 1988 and 2001. The costs came from sources published in 1998, 1999, and 2000. The price year was 2000.
Source of effectiveness data The effectiveness evidence was derived from a synthesis of published studies.
Modelling A stationary Markov model was constructed to examine the long-term clinical and economic impact of RA treatment. The time horizon was lifetime and the cycle length was 6 months. Health states were defined according to a pairwise combination of five treatments and four disability categories. The five treatment options were INF+MTX, MTX, DMARD, DMARD+MTX, and non-steroidal anti-inflammatory drugs (NSAIDs). The four disability categories were based on the Health Assessment Questionnaire (HAQ) score, a disease-specific instrument used to assess the overall physical and mental health of patients on a scale of 0 (best) to 3 (worst): HAQ=0; 0 < HAQ =/< 1; 1 < HAQ =/< 2; and 2 < HAQ =/< 3. The primary analysis assumed that treatment was for 1 year only. In the absence of maintenance of INF treatment after 54 weeks, the model predicted that the clinical benefit in terms of reduced HAQ score diminished over time, so that one third was lost by 2 years, three quarters by 5 years, and nearly all was lost by 10 years.
Outcomes assessed in the review The outcomes estimated from the literature were:
the 1-year effectiveness results (percentage of patients with improvements in American College of Rheumatology (ACR) 20 and ACR 50),
the proportions of patients in the different HAQ groups at 54 weeks,
the utility values associated with different HAQ scores,
mortality associated with each increase in HAQ disability, and
6-month transition probabilities across health states.
The ACR 20 was defined as a 20% decrease in the number of swollen and tender joints; a 20% improvement in the patient's assessment of pain; a 20% improvement in the patient's global assessment of disease status; and a 20% improvement in the HAQ score.
Study designs and other criteria for inclusion in the review A systematic review of the literature was not undertaken and the primary studies were identified selectively. The evidence came from a clinical trial (the Anti-Tumour Necrosis Factor Trial in Rheumatoid Arthritis with Concomitant Therapy, ATTRACT), the ARAMIS (Arthritis, Rheumatism, and Aging Medical Information System) cohort and another published study. Some information on the study sample and other characteristics of the primary studies was provided. The ATTRACT included 428 patients receiving different dosages of INF+MTX or MTX alone followed up for 54 weeks, while the ARAMIS involved 4,258 patients followed up to 17,085 patient-years. Data from the ATTRACT were used to derive clinical values for the first year of treatment, while transition probabilities for the subsequent years came from the ARAMIS cohort.
Sources searched to identify primary studies Criteria used to ensure the validity of primary studies No explicit criteria were used to ensure the validity of the primary studies. However, much of the data came from a multi-centre, randomised, double-blind, placebo-controlled clinical trial, which usually has a robust design.
Methods used to judge relevance and validity, and for extracting data Number of primary studies included Three primary studies provided evidence.
Methods of combining primary studies The primary estimates were not combined. In fact, the results of ATTRACT were used to compare the efficacy of INF versus INF+MTX during the first year, while the ARAMIS database was used to estimate disease progression after the first year.
Investigation of differences between primary studies Results of the review During the first year, the ACR 20 improvement was 17% with MTX alone, 42% with INF 3 mg/kg for 8 weeks, 48% with INF 3 mg/kg for 4 weeks, and 59% with INF 10 mg/kg both 8 weeks and 4 weeks.
The ACR 50 improvement was 8% with MTX alone, 21% with INF 3 mg/kg for 8 weeks, 34% with INF 3 mg/kg for 4 weeks, 39% with INF 10 mg/kg for 8 weeks, and 38% with INF 10 mg/kg for 4 weeks. All INF strategies were significantly more effective than the MTX alone option, (p<0.001). No statistically significant differences were observed in the frequency of serious adverse events and serious infections.
The proportions of patients in the different HAQ groups at 54 weeks were as follows:
for MTX alone: HAQ 0, 0.062, 0 < HAQ =/< 1, 0.215; 1 < HAQ =/< 2, 0.523; 2 < HAQ =/< 3, 0.2; and
for MTX+INF: HAQ 0, 0.087; 0 <HAQ =/< 1, 0.366; 1 < HAQ =/< 2, 0.44; 2 < HAQ =/< 3, 0.107.
The utility value in the first year was 0.509 for MTX and 0.595 for INF+MTX, (p<0.001).
The utility values associated with NSAIDs, DMARDs, MTX and DMARD+MTX were, respectively:
0.891, 0.868, 0.838 and 0.836 in HAQ 0;
0.738, 0.735, 0.738 and 0.714 in 0 <HAQ =/< 1;
0.539, 0.547, 0.574 and 0.547 in 1 < HAQ =/< 2;
0.400, 0.413, 0.434 and 0.416 in 2 < HAQ =/< 3.
The mortality associated with each increase in HAQ disability was 1.77.
Transition probabilities were also reported.
Measure of benefits used in the economic analysis The summary benefit measure was the estimated number of quality-adjusted life-years (QALYs). These were estimated using the decision model. Utility values were obtained both from the ATTRACT study (first year) and from the ARAMIS cohort (subsequent years). Patients in both the ATTRACT and ARAMIS completed a questionnaire, based on a visual analogue scale, every 6 months together with an assessment of HAQ. An annual discount rate of 1.5% was applied.
Direct costs An annual discount rate of 6% was applied because of the long timeframe of the model. The unit costs were reported separately from the resources used, both for the first year and for the subsequent years. The health services included in the economic evaluation were drugs (including their administration) and the costs of monitoring and treating adverse events. A detailed breakdown of the cost items was provided. Other key components were outpatient visits, hospitalisations and joint replacements. The cost/resource boundary of the NHS was adopted in the study. The costs were estimated from the Personal Social Services research unit, the Chartered Institute of Public Finance and Accountancy, the Healthcare Resource Groups, the British National Formulary and the literature. Resource use during the first year was estimated from the ATTRACT, while resource consumption beyond the first year came from the Norfolk Arthritis Register (NOAR) cohort, where 1,236 patients were followed for 5 years. The price year was 2000.
Statistical analysis of costs No statistical analyses of the costs were carried out.
Indirect Costs The indirect costs were not included in the economic evaluation.
Sensitivity analysis Univariate sensitivity analyses were carried out to examine the impact of model inputs on the model results. In particular, different discount rates were used, items of resource use not measured in the NOAR database (non-medical direct costs) were considered, and alternative UK resource use rates were used. Additional sensitivity analyses were also performed, changing the mortality rate for RA patients, increasing or decreasing the odds likelihood ratio of HAQ progression for non-INF-treated patients, and varying the increase in quality of life due to INF treatment.
Three alternative scenarios were considered:
the analysis assumed treatment for 1 year, but with the assumption of lasting benefit in reducing HAQ score (radiographic progression analysis);
an intent to treat (ITT) analysis was undertaken over 102 weeks, assuming that continued INF therapy beyond 54 weeks would maintain the patient's HAQ status (ITT analysis);
the ITT analysis was extended to lifetime, assuming that patients would maintain, but not improve, their HAQ status (lifetime). In the lifetime analysis patients could receive INF during the rest of their life, keeping the discontinuation rate found for the first year constant for the following years.
Estimated benefits used in the economic analysis The incremental QALYs with INF+MTX over MTX alone were 0.26 in the primary analysis, 1.53 in the radiographic progression analysis, 0.40 in the ITT analysis, and 1.26 in the lifetime analysis.
Cost results The average direct costs per patient after one year were 4,980.84 for MTX alone and 3,880.99 for INF+MTX. However, the costs of INF were not included in this cost calculation, and these appear to be the highest category of cost (approximately 8,900 during the first year).
The incremental costs with INF+MTX over MTX alone were 8,576 in the primary analysis, 7,835 in the radiographic progression analysis, 14,635 in the ITT analysis, and 30,147 in the lifetime analysis.
Synthesis of costs and benefits An incremental cost-utility ratio was calculated to combine the costs and QALYs of INF+MTX over MTX alone. The incremental cost per QALY was 33,618 in the primary analysis, 5,111 in the radiographic progression analysis, 36,616 in the ITT analysis, and 23,936 in the lifetime analysis.
The sensitivity analysis showed that the base-case results were quite robust to variations in model inputs, The cost per QALY varied from 17,252 to 40,453 for the primary analysis, from 3,832 to 9,693 for the radiographic progression analysis, from 34,182 to 42,634 for the ITT analysis, and from 20,477 to 34,680 for the lifetime analysis.
Authors' conclusions New therapies for rheumatoid arthritis (RA), like infliximab (INF), have the potential to be cost-effective in patients with severe active treatment-resistant RA. The budget impact on the National Health Service (NHS) for the acquisition of the drug would be around 50 million per year (drug treatment regimens for patients with severe disease that is resistant to disease-modifying antirheumatic drugs, DMARDs, cost between 428 and 3,417 per year).
CRD COMMENTARY - Selection of comparators The authors did not provide a clear justification for the choice of the comparator (MTX alone) although it was stated that this represented the comparator used in the clinical trial that was the main source of the evidence. It would appear that MTX was chosen because of the data available from the clinical trial. You should decide whether this is a valid comparator in your own setting.
Validity of estimate of measure of effectiveness The bulk of the effectiveness evidence came from two studies, both of which were extensively described. The first one, the ATTRACT, was a robust clinical trial on account of the characteristics of the design (multi-centre design, randomised and double-blind). The second study, the ARAMIS cohort, which was used to extrapolate 1-year data to a long-term perspective, was a longitudinal study involving a large sample of patients. The two studies were identified selectively and there was no review of the literature to identify clinical inputs. However, the use of the ARAMIS was justified by the large sample of patients included. Key model inputs were varied in the sensitivity analysis and alternative scenarios were also considered.
Validity of estimate of measure of benefit The summary benefit measure (QALYs) was appropriate because it considers the impact of the interventions on quality of life and survival. Discounting was applied, as UK guidelines recommend. The utility weights were derived from patients' self values. The approach used to calculate the QALYs was described.
Validity of estimate of costs The costs included were consistent with the perspective adopted in the analysis. Only direct medical costs were considered, although the inclusion of direct non-medical costs was investigated in the sensitivity analysis. Extensive information on the unit costs and quantities of resources used was provided, which enhances the possibility of replicating the cost analysis in other settings. The source of the data, which represented typical NHS sources, was reported. A UK cohort was used to estimate resource use associated with each treatment and HAQ state. The costs were treated deterministically. The price year was reported, which will facilitate reflation exercises in other time periods.
Other issues The authors compared their findings with those from a recently published Swedish study that reported similar results. The issue of the generalisability of the study results to other settings was not explicitly addressed, but several sensitivity analyses were carried out to examine different scenarios. The authors stated that their results were likely to be conservative because other benefits obtained from the use of INF were not considered, such as indirect the costs. The model was also validated using published data. The authors noted some limitations of their study. First, the use of a visual analogue scale to obtain QALYs. Second, the assumption made about the duration of the INF effect. Finally, there was no probabilistic approach to uncertainty in the data.
Implications of the study The study results suggested that INF+MTX might be cost-effective from the perspective of the NHS for the treatment of patients with RA with uncontrolled disease despite treatment with MTX.
Source of funding Supported by Schering-Plough Inc.
Bibliographic details Barbieri M, Wong J B, Drummond M. The cost-effectiveness of infliximab for severe treatment resistant rheumatoid arthritis in the UK. PharmacoEconomics 2005; 23(6): 607-618 Other publications of related interest Lipsky PE, van der Heijde DM, St Clair EW, et al. Infliximab and methotrexate in the treatment of rheumatoid arthritis. New England Journal of Medicine 2000;343:1594-602.
Maini R, St Clair EW, Breedveld F, et al. Infliximab (chimeric anti-tumour necrosis factor alpha monoclonal antibody) versus placebo in rheumatoid arthritis patients receiving concomitant methotrexate: a randomised phase III trial. ATTRACT Study Group. Lancet 1999;354:1932-9.
Wong J B, Gurkipal S, Kavanaugh A. Estimating the cost-effectiveness of 54 weeks of infliximab for rheumatoid arthritis. American Journal of Medicine 2002;113:400-8.
Kobelt G, Jonsson L, Young A, et al. The cost-effectiveness of infliximab (Remicade) in the treatment of rheumatoid arthritis in Sweden and the United Kingdom based on the ATTRACT study. Rheumatology 2003;42:326-35.
Indexing Status Subject indexing assigned by NLM MeSH Antibodies, Monoclonal /economics /therapeutic use; Antirheumatic Agents /economics /therapeutic use; Arthritis, Rheumatoid /drug therapy /economics; Cost of Illness; Cost-Benefit Analysis /methods; Drug Resistance; Great Britain; Humans; Infliximab; Markov Chains; Methotrexate /economics /therapeutic use; Quality-Adjusted Life Years AccessionNumber 22005008278 Date bibliographic record published 31/12/2005 Date abstract record published 31/12/2005 |
|
|
|