Source of effectiveness data The effectiveness data were derived from a review and synthesis of completed studies.

Transcription

1 Economic evaluation of maintenance therapies for reflux esophagitis: comparison between step-up therapies and step-down therapies Sugano K, Kobayashi 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 Maintenance therapies for reflux oesophagitis, namely H2-receptor antagonists (H2RA) using ranitidine (RAN) and proton-pump inhibitors (PPIs), were examined. The two PPIs used were lansoprazole (LPZ) and omeprazole (OPZ). The dosage levels varied according to a number of treatment strategies, which are fully described in the 'Modelling' and 'Results' sections. Type of intervention Treatment. Economic study type Cost-effectiveness analysis. Study population The target population was adults who were examined by endoscopy and found to be suffering from gastro-oesophageal reflux disease or reflux oesophagitis. Setting The setting was a hospital. The economic study was carried out in Japan. Dates to which data relate The effectiveness data were obtained from papers published between 1987 and The cost data relate to Source of effectiveness data The effectiveness data were derived from a review and synthesis of completed studies. Modelling Markov models were used to compare the therapy processes. Therapy A was the step-up therapy that started with H2RA (RAN), then changed to LPZ for non-cured cases. Therapies B and C were the step-up therapies that started with H2RA, then changed to OPZ for non-cured cases. Therapy D was step-down therapy with LPZ alone. Therapies E and F were step-down therapy with OPZ alone. While E started with 10 mg/day, F started with 20 mg/day. For all the therapy groups, if the PPI treatment was not effective, the dosage was doubled. If the condition was cured after the first maintenance therapy and the condition-free period lasted for 6 months, the medication was stopped. If recurrence cases were cured of the condition after maintenance therapy, medication was reduced to half for the next 6 months (if the full amount was given during the maintenance therapy) or stopped (if half the amount was given during the maintenance therapy). Page: 1 / 5

2 Outcomes assessed in the review The outcomes assessed were the cure rates and recurrence rates after the first-stage treatments. Study designs and other criteria for inclusion in the review The inclusion criteria specified the following: randomised controlled studies (single- and double-blinded); studies with oral or written consent from all of the patients; studies with more than 20 cases; studies that confirmed cure and recurrence of oesophagitis by endoscopy; studies that confirmed cure of oesophagitis on the 8th week after the start of medication; studies in which the therapy was maintained for 6 or 12 months; studies that examined erosive or ulcerative oesophagitis (Savary and Miller categorisation, grade II to IV); when studies categorised according to mild, moderate and severe oesophagitis, only studies that clarified the cure and recurrence rates of moderate and severe cases; and if studies treated all grades of oesophagitis, only those that clarified cure and recurrence rates of Savary and Miller grades II to IV. Sources searched to identify primary studies MEDLINE and the Cochrane Library were searched. Criteria used to ensure the validity of primary studies Methods used to judge relevance and validity, and for extracting data Number of primary studies included Fifty studies were included in the review. Methods of combining primary studies The data in the model were synthesised using a meta-analysis, according to the techniques adopted by Einarson et al. (see Other Publications of Related Interest). Investigation of differences between primary studies Results of the review The cure rates on the 8th week after the start of the therapy were for LPZ (30 mg/day), for OPZ (20 mg/day), and for RAN (300 mg/day). Page: 2 / 5

3 In the per protocol analysis, the recurrence rates within 6 months of maintenance therapy were (0.038/month) for LPZ (15 mg/day), (0.025/month) for LPZ (30 mg/day), (0.065/month) for OPZ (10 mg/day), (0.033/month) for OPZ (20 mg/day), and (0.072/month) for RAN (300 mg/day). In the intention to treat analysis, the recurrence rates within 6 months of maintenance therapy were (0.033/month) for LPZ (15 mg/day), (0.017/month) for LPZ (30 mg/day), (0.069/month) for OPZ (10 mg/day), (0.032/month) for OPZ (20 mg/day), and (0.067/month) for RAN (300 mg/day). In the per protocol analysis, the recurrence rates within 12 months of maintenance therapy were (0.027/month) for LPZ (15 mg/day), (0.019/month) for LPZ (30 mg/day), (0.046/month) for OPZ (10 mg/day), (0.018/month) for OPZ (20 mg/day), and (0.063/month) for RAN (300 mg/day). In the intention to treat analysis, the recurrence rates within 12 months of maintenance therapy were (0.026/month) for LPZ (15 mg/day), (0.014/month) for LPZ (30 mg/day), (0.043/month) for OPZ (10 mg/day), (0.015/month) for OPZ (20 mg/day), (0.063/month) for RAN (300 mg/day), and (0.101/month) for no medication. Measure of benefits used in the economic analysis The benefit measure used in the economic analysis was the number of days in which patients were free from reflux oesophagitis, in other words, the disease-free days (DFDs). Direct costs The direct costs included the primary outpatient examination costs, medication and endoscopy costs. The examination costs were for repeated examination, administration and prescription costs. Endoscopy costs were for medication and film costs. The costs were derived from the medication costs and doctors' examination fee reported for April The costs and the quantities were reported separately. Discounting was not conducted, but was irrelevant due to the short period of analysis (1 year). Statistical analysis of costs Not carried out. Indirect Costs The indirect costs were not included. Currency Japanese Yen (Y). Sensitivity analysis A one-way sensitivity analysis was carried out to examine the cost-effectiveness variability in relation to the cure and recurrence rates of various medication methods. First, the rates were changed within the 95% confidence intervals. Second, the cure rates after recurrence were changed between 0 and 100%. Variability in the cost-effectiveness was also examined at 6 months by per protocol analysis, and at 12 months by both intention to treat and per protocol analyses. Estimated benefits used in the economic analysis Therapy A gained 230 DFDs, therapy B 227 days, therapy C 229 days, therapy D 266 days, therapy E 248 days, and therapy F 263 days. Page: 3 / 5

4 Cost results The incremental direct costs per person for one year were Y95,858 for therapy A, Y95,944 for B, Y99,121 for C, Y90,768 for D, Y100,060 for E, and Y112,320 for F. Synthesis of costs and benefits The average costs per DFDs were also calculated. These were Y417 for therapy A, Y422 for B, Y433 for C, Y341 for D, Y404 for E, and Y428 for F. Sensitivity analyses confirmed the lowest cost per DFDs was obtained for therapy D in comparison with the other therapies. An incremental analysis was not performed. Authors' conclusions Step-down therapy starting with 30 mg/day lansoprazole (LPZ) was the most cost-effective therapy for the treatment of above-moderate reflux oesophagitis. CRD COMMENTARY - Selection of comparators The rationale and justification for the choice of the comparator (PPI) was clear and appropriately presented. You should consider if the two alternatives examined in the study are relevant to your own setting in the treatment of reflux oesophagitis. Validity of estimate of measure of effectiveness The effectiveness data used in the construction of the Markov models were derived from the literature using a systematic review, and were synthesised using a meta-analysis. Moreover, confidence intervals were used to examine the impact of variability in the estimates for the sensitivity analyses. The reporting was clear and transparent. The effectiveness results should, therefore, have high validity and can be independently verified by other studies. Validity of estimate of measure of benefit The measure of benefit (DFDs) was clearly appropriate for the patient domain and disease studied. DFDs were derived directly from the effectiveness results through modelling. The benefit measure was used, also appropriately, in a synthesis of the costs and benefits. Validity of estimate of costs All the relevant costs for the chosen perspective appear to have been included, along with the sources and price years. The indirect costs (such as productivity losses) would need to be included for a societal perspective, which would be more informative for the UK NHS. Other issues The authors compared their results with those from other studies. In relation to the generalisability of the results, they state their findings are applicable to both western countries, where H2RAs and PPIs have been evaluated, and to Japan, where PPIs have only recently been accepted for long-term therapy. However, more work in Japanese populations is necessary to determine if the response to treatment is consistent between western and Japanese populations. However, some of the studies used to derive the effectiveness data came from Japanese as well as western studies. Sensitivity analyses were not undertaken on the costs. This is often the case in Japanese studies, as prices for reimbursement are fixed centrally by the Japanese government. As such, the cost results are probably not generalisable to other countries such as the UK. Implications of the study The findings of this study suggest step-down therapy, starting with LPZ 30mg/day, is the most cost-effective therapy for the treatment of above-moderate reflux oesophagitis. To improve the reliability of the results for the Japanese Page: 4 / 5

5 Powered by TCPDF ( setting, more work with Japanese study populations is necessary. Source of funding None stated. Bibliographic details Sugano K, Kobayashi M. Economic evaluation of maintenance therapies for reflux esophagitis: comparison between step-up therapies and step-down therapies. Japanese Pharmacology and Therapeutics 2001; 29(6): Other publications of related interest Einarson TR, et al. Pharmacoeconomic applications of meta-analysis for single group using antifungal onychomycosis lacquers as an example. Clinical Therapeutics 1997;19: Indexing Status Subject indexing assigned by CRD MeSH Adult; Antacids /therapeutic use; Cost-Benefit Analysis; Gastroesophageal Reflux /drug therapy; Gastrointestinal Agents /therapeutic use; Histamine H2 Antagonists /therapeutic use; Japan; Omeprazole /therapeutic use; Proton Pumps /therapeutic use; Ranitidine /therapeutic use AccessionNumber Date bibliographic record published 30/06/2003 Date abstract record published 30/06/2003 Page: 5 / 5