|
Cost-effectiveness of treatment with etanercept for psoriasis in Sweden |
Knight C, Mauskopf J, Ekelund M, Singh A, Yang S, Boggs R |
|
|
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. CRD summary This study estimated the cost-effectiveness of intermittent treatment with etanercept (Enbrel) in patients with moderate to severe psoriasis. The authors concluded that etanercept was cost-effective compared to both non-systemic treatment and adalimumab, assuming a EUR 50,000 willingness-to-pay threshold. The study was reasonably well conducted and reported, but it was not clear that the treatment effect data were derived appropriately, which makes the results uncertain. Type of economic evaluation Study objective The study estimated the cost-effectiveness of intermittent treatment with etanercept (Enbrel) in patients with moderate to severe psoriasis. Interventions Three treatment regimes for psoriasis were assessed: continuous indefinite dosing with adalimumab (40mg every other week; 80mg in week 0); intermittent dosing with etanercept (50mg once weekly for up to 24 weeks followed by treatment interruption, with subsequent intermittent dosing of etanercept 25mg twice weekly for up to a further 24 weeks); and non-systemic treatment, such as topical steroid creams. All patients who received etanercept or adalimumab also received non-systemic therapy. After 12-week intervals, patients on adalimumab or etanercept were assessed for their response to treatment, with subsequent treatment assigned according to the level of efficacy observed. Methods Analytical approach:A Markov model was constructed to extrapolate one year trial results over a 10-year time horizon. A cycle length was 12 weeks. The authors stated that the perspective was societal. Effectiveness data:For etanercept, effectiveness data came from two double-blind randomised controlled trials (RCTs) which tracked 1,283 patients with chronic plaque psoriasis. The two RCTs administered etanercept in different dosages. The authors claimed that it had previously been shown that the steady-state serum concentrations of etanercept were very similar at the different dosage regimens, so it could be assumed that the two regimens would have comparable efficacy and safety profiles in the clinical setting. The placebo groups in these trials were used as the non-systemic therapy data source. For adalimumab, effectiveness data mainly came from a 12-week clinical controlled trial that treated patients with moderate to severe psoriasis; an extended, 120-weeks open-label study from the clinical trial; and two phase III trials. The primary endpoint considered was a 75% improvement from baseline in the Psoriasis Area and Severity Index (PASI 75 response) over 12 weeks. Monetary benefit and utility valuations:Utility scores for PASI response ranges were calculated from mean Dermatology Life Quality Index (DLQI) scores, which were taken from the Wyeth submission to the National Institute for Health and Clinical Excellence (NICE). The DLQI scores were mapped to EQ-5D utility using a published formula. The DLQI scores came from the results of a prospective survey of UK patients. Measure of benefit:The measure of benefit was quality-adjusted life-years (QALYs). The authors did not report at what rate health effects were discounted. The discount rate was altered between 0 and 5% within sensitivity analysis, but the base case value was not reported. Cost data:The costs included were those of drugs, drug administration, patient loss of earnings, and patient loss of productivity. Costs came from national or regional official sources specific to Sweden. The incidence of sick leave was estimated from a cross-sectional observational study (conducted in the UK, Spain, France, Italy, and Germany), but costs for lost work days were calculated from Swedish sources. Costs were reported in Swedish kronor (SEK) and Euros (EUR) using an exchange rate of SEK 9.43 to EUR 1. The discount rate was altered from 0 and 5% within sensitivity analysis. Analysis of uncertainty:Uncertainty was assessed by one-way sensitivity analysis, which evaluated the sensitivity of the results over a range of plausible values for specific model parameters. Probabilistic sensitivity analyses using Monte Carlo simulation (10,000 simulations) were also conducted, which accounted for uncertainty in all model parameters simultaneously, excluding drug costs. The results of the one-way sensitivity analysis were selectively reported within the narrative. The results of the probabilistic analysis were reported as cost-effectiveness acceptability curves (CEACs) in which the probability of cost-effectiveness was plotted against a range willingness-to-pay thresholds. The cost-effectiveness was assessed using both total costs and direct costs alone. Results For adalimumab, the total costs were SEK 1,116,336 (EUR 118,332) and the associated QALYs were 6.74. For etanercept, the total costs were SEK 836,464 (EUR 88,665) and the associated QALYs were 6.56. For non-systemic therapy, the total costs were SEK 781,062 (EUR 82,793) and the associated QALYs were 5.97. The incremental cost effectiveness ratios (ICERs) for the analysis using total costs were SEK 93,629 (EUR 9,925) per QALY for etanercept compared with non-systemic therapy; SEK 434,782 (EUR 46,087) per QALY for adalimumab compared with non-systemic therapy; and SEK 1,559,939 (EUR 165,354) per QALY for adalimumab compared with etanercept. One-way sensitivity analysis showed that altering the discount rates between 0 and 5% resulted in the ICER for etanercept compared with non-systemic therapy to range from SEK 81,916 (EUR 8,683) to SEK 102,554 (EUR 10,871) per QALY; the ICER for etanercept compared with adalimumab to range from SEK 357,246 (EUR 37,868) to SEK 526,735 (EUR 55,834) per QALY. Changes in indirect costs, resource costs and efficacy rates were reported to have little effect on the ICERs. The results of the probabilistic sensitivity analysis showed that at at a low threshold of SEK 100,000 (EUR 10,600), the probability of non-systemic therapy being the most cost-effective alternative was virtually 100% for direct costs and 43% for total costs. Increasing the threshold to SEK 300,000 (EUR 32,000) resulted in etanercept becoming the preferred treatment with a probability that it was the most cost-effective option of 52% for direct costs only and 96% for total costs. At a threshold of SEK 470,000 (EUR 50,000), the probability of etanercept being the most cost-effective option was 89% for direct costs and 96% for total costs. Authors' conclusions The authors concluded that etanercept was cost effective compared to both non-systemic treatment and adalimumab, assuming a EUR 50,000 willingness-to-pay threshold. CRD commentary Interventions:The interventions adalimumab and etanercept were clearly reported but only limited details concerning non-systemic care were reported. The comparators appeared to be appropriate and included usual practice. Effectiveness/benefits:Effectiveness estimates and data sources were clearly reported and relevant, but the methods used to identify and select data sources was not discussed. It was not clear how the different treatment efficacy rates were derived. In particular, it was not clear whether randomisation was preserved when using relative treatment effects identified in different trials. The utility score derivation and values were clearly reported. There may have been additional uncertainty not captured in the model parameter estimates due to the use of different dosage regimens in the selected RCTs and mapping from DLQI utility scores to EQ-5D utility scores. The discount rate for future benefits was not reported. Costs:Costs were generally well reported. They mainly came from national and regional sources relevant to the setting of Sweden. A study based in other EU countries had to be used to estimate the incidence of sick leave, which was justified given the lack of data specific to the Swedish population. Whilst the authors stated that the incidence of sick leave would likely be higher in Sweden than in other EU countries, they did not discuss to what extent results might differ, or how much results varied within those countries assessed, or which country's result may have been most applicable to Sweden. Sensitivity analysis was conducted on costs but the range of values was not reported, so it was unclear whether uncertainty for the incidence of sick leave was included. The discount rate for future costs and the price year were not reported. Analysis and results:The model was appropriate and clearly described with a diagram provided. The results of the base-case analysis were clearly reported. The methodology and results for the one-way analyses were not clearly reported; in particular, the range of values selected for each parameter were not reported or justified and only selected results were reported. This made it difficult to assess the validity of the one-way analyses. Probabilistic sensitivity analysis was also conducted allowed all the known uncertainty in the model to be captured in the results. Each of the parameters was assigned appropriate distributions. The conclusions drawn from the probabilistic sensitivity analysis appeared to be valid.
Concluding remarks:The study was reasonably well conducted and reported, but it was not clear that the treatment effect data were derived appropriately, which makes the results uncertain. Bibliographic details Knight C, Mauskopf J, Ekelund M, Singh A, Yang S, Boggs R. Cost-effectiveness of treatment with etanercept for psoriasis in Sweden. European Journal of Health Economics 2012; 13(2): 145-156 Indexing Status Subject indexing assigned by NLM MeSH Adalimumab; Anti-Inflammatory Agents /administration & Antibodies, Monoclonal, Humanized /administration & Cost-Benefit Analysis; Etanercept; Health Care Costs; Humans; Immunoglobulin G /administration & Immunologic Factors /administration & Markov Chains; Psoriasis /drug therapy /economics; Quality-Adjusted Life Years; Randomized Controlled Trials as Topic; Receptors, Tumor Necrosis Factor /administration & Sweden; Treatment Outcome; dosage /economics /therapeutic use; dosage /economics /therapeutic use; dosage /economics /therapeutic use; dosage /economics /therapeutic use; dosage /therapeutic use AccessionNumber 22012014770 Date bibliographic record published 20/06/2012 Date abstract record published 12/02/2013 |
|
|
|