|
Cost-effectiveness analysis of transjugular intrahepatic portasystemic shunt (TIPS) versus endoscopic therapy for the prevention of recurrent esophageal variceal bleeding |
Russo M W, Zacks S L, Sandler R S, Brown R S |
|
|
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 Comparing transjugular intrahepatic portasystemic shunt (TIPS) with endoscopic sclerotherapy and endoscopic ligation for the prevention of recurrent esophageal variceal bleeding in cirrhotic patients successfully treated for esophageal variceal bleeding with endoscopic sclerotherapy.
Economic study type Cost-effectiveness analysis.
Study population A hypothetical cohort of cirrhotic patients successfully treated for esophageal variceal bleeding with endoscopic sclerotherapy who received prophylactic sclerotherapy, ligation, or TIPS over 1 year.
Setting Hospital. The economic analysis was carried out in the USA.
Dates to which data relate Effectiveness data were obtained from the literature published between 1992 and 1998. Resource use data corresponded to patients treated in two health institutions between 1995 and 1996. The price year was 1996.
Source of effectiveness data The evidence for the final outcomes was based on a review of the literature.
Modelling A decision analytic model was developed to estimate the costs and effects associated with the preventive strategy. The study model was similar to the design of clinical trials comparing TIPS with sclerotherapy. A hypothetical cohort of patients with variceal bleeding would be treated with endoscopic sclerotherapy; subjects who survived and whose variceal bleeding was controlled with endoscopic sclerotherapy would be randomised to sequential endoscopic sclerotherapy, sequential endoscopic ligation, or TIPS and followed for 1 year. The major complications of these procedures would be stricture formation, perforation, and portasystemic encephalopathy, respectively. Subjects with recurrent variceal bleeding despite repeat endoscopic therapy would cross over to the TIPS arm of the study.
Outcomes assessed in the review The following outcomes were assessed; rates of recurrent bleeding, encephalopathy, 1-year survival (against death from all causes), cross over to TIPS, control of bleeding, stricture requiring dilatation, perforation, and restenosis.
Study designs and other criteria for inclusion in the review Data were obtained primarily from randomised clinical trials comparing TIPS with endoscopic sclerotherapy. The study groups in these trials contained patients with liver disease predominantly from alcohol and viral hepatitis and were, on average, 50 years of age. In these studies approximately 70% of the patients had mild to moderate liver disease characterised by Child-Pugh class A or B and the remaining patients were Child-Pugh class C. The authors included endoscopic ligation in their model because many endoscopists at the time of the study preferred it to sclerotherapy for variceal bleeding. To address the uncertainty about outcomes associated with ligation, plausible ranges in sensitivity analyses were used to assess the robustness of the study results.
Sources searched to identify primary studies MEDLINE was searched from 1990 onwards and the bibliographies of the studies found by the MEDLINE search were used for additional references.
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 A total of 15 studies were directly reported as the references for the clinical probabilities incorporated in the model.
Methods of combining primary studies Midpoint values of the range of values reported in the literature were used for the base case analysis. It was reported that the values used in the base case analysis approximated the estimates reported in a published meta-analysis of trials comparing TIPS with endoscopic sclerotherapy. In sensitivity analyses the authors tested the robustness of the base case results using a wide range of reported values for key variables.
Investigation of differences between primary studies Results of the review The outcome results were as follows:
recurrent bleeding, 0.40 (range: 0.30 - 0.57) for endoscopic sclerotherapy, 0.30 (range: 0.15 - 0.36) for endoscopic ligation, and 0.20 (range: 0.12 - 0.35) for TIPS;
encephalopathy, 0.13 for endoscopic sclerotherapy, 0.13 for endoscopic ligation, and 0.35 for TIPS;
1-year survival (against death from all causes), 0.80 for all three procedures;
cross over to TIPS, 0.20 (range: 0.10 - 0.28) for endoscopic sclerotherapy, and 0.20 (range: 0.10 - 0.36) for endoscopic ligation;
control of bleeding, 0.80 (range: 0.75 - 0.85) and 0.85 (range: 0.80 - 0.90), respectively;
stricture requiring dilatation, 0.07 (range: 0.07 - 0.13) for endoscopic sclerotherapy;
perforation, 0.007 for endoscopic ligation; and
restenosis, 0.40 (range: 0.17 - 0.56) for TIPS.
Measure of benefits used in the economic analysis Rebleeding episodes per patient was the main measure of benefit adopted in the economic study. Mortality rates and the number of encephalopathy cases requiring hospitalisation per patient were also reported.
Direct costs Costs were not discounted due to the short time frame of the cost analysis (one-year). Some quantities were reported separately from the costs. Cost items were reported separately. Cost analysis covered the costs of endoscopic sclerotherapy, endoscopic ligation, hospital stay for recurrent bleeding, encephalopathy, perforation, and complications from procedures (stricture, perforation, encephalopathy, and stenosis). The perspective adopted in the cost analysis was not explicitly specified. Cost data were obtained from two hospital centres. ICD-9 codes were used to identify the hospital admissions for esophageal variceal bleeding, hepatic encephalopathy, and esophageal perforation. The average cost of hospitalisation for each diagnosis was used. Direct costs were calculated using activity-based costing at one of the study hospitals, and costs were calculated with the TSI systems at the other study hospital. Charges and cost-to-charge ratios were not used for hospitalisations. The mean cost of the two medical centres was used for each procedure in the base case analysis whilst costs from each individual centre were used in the sensitivity analysis. The price year was 1996.
Sensitivity analysis A series of one-way sensitivity analyses was performed on input parameters to test the robustness of the study findings.
Estimated benefits used in the economic analysis In a hypothetical cohort of patients with variceal bleeding followed for 1 year there were 0.39 rebleeding episodes per patient with sclerotherapy and 0.32 rebleeding episodes per patient with ligation. In comparison, the model projected 0.07 rebleeding episodes per patient with TIPS. Mortality rates were 0.24 in the sclerotherapy group, 0.21 in the ligation group, and 0.20 in the TIPS group. The number of encephalopathy cases requiring hospitalisation per patient were 0.17 in the sclerotherapy group, 0.17 in the ligation group, and 0.26 in the TIPS group.
Cost results The total annual costs per patient were $23,459 for endoscopic sclerotherapy, $23,111 for endoscopic ligation, and $26,275 for TIPS.
Synthesis of costs and benefits The incremental cost to prevent one rebleeding episode was calculated as the measure of cost-effectiveness, yielding a value of $8,803 for TIPS compared with sclerotherapy and $12,660 for TIPS compared with ligation. Ligation was less expensive and more effective than sclerotherapy in the base case analysis. The sensitivity analyses showed that the study conclusions were relatively robust to changes in the parameter values, except at extreme values for stenosis and the cost of TIPS.
Authors' conclusions Compared with endoscopic therapy, TIPS leads to lower recurrent variceal bleeding rates and is more cost effective in the short term for the prevention of recurrent esophageal variceal bleeding.
CRD COMMENTARY - Selection of comparators No specific preventive strategy was explicitly regarded as the comparator since it appears that there is no consensus on the procedure of choice in the context in question. You, as a database user, should consider which preventive strategy is widely used in your own setting.
Validity of estimate of measure of effectiveness The effectiveness results are likely to be internally valid due to the comprehensive literature review performed which included, primarily, randomised trials; furthermore, uncertainties were addressed by sensitivity analysis. However, the criteria used to ensure the validity of primary studies (such as blinding methods adopted in the studies), and the methods used to judge relevance and validity, and for extracting data were not reported.
Validity of estimate of measure of benefit Estimation of benefits was modelled. The instrument used to derive a measure of health benefit, (decision analytic model), appears to be appropriate.
Validity of estimate of costs Some quantities were reported separately from the costs and adequate details of methods of cost estimation were given. It appears that the important direct cost elements were included in the cost analysis. The price year was specified. The cost calculations were, to some extent, based on true costs (in one medical centre) rather than charges. The perspective adopted in the cost analysis was not explicitly specified. The effects of alternative procedures on indirect costs were not addressed.
Other issues The authors' conclusions appear to be justified given the comprehensive literature review, the randomised trials included in the review, and the extensive sensitivity analysis performed. The issue of generalisability to other settings or countries was not explicitly addressed, however some comparisons were made with other studies. Regarding the degree to which the study cohort was representative of the study population, it was reported that the hypothetical cohort of patients in the study was similar to the patient populations in the clinical trials comparing TIPS with endoscopic sclerotherapy. A short-time follow-up and lack of quality of life analysis were mentioned as the major limitations of this study; it was acknowledged that quality of life studies may help determine patient preferences and which is "the lesser of two evils", recurrent bleeding or encephalopathy.
Implications of the study The study results apply to patients who require therapy for 1 year, such as patients with a history of variceal bleeding who are on a liver transplant waiting list and who are anticipated to undergo transplantation in the near future. Endoscopic therapy may be more appropriate for patients requiring long-term care for recurrent variceal haemorrhage. Further research might choose to investigate the issues of this study over a longer period of time, using state transition models.
Bibliographic details Russo M W, Zacks S L, Sandler R S, Brown R S. Cost-effectiveness analysis of transjugular intrahepatic portasystemic shunt (TIPS) versus endoscopic therapy for the prevention of recurrent esophageal variceal bleeding. Hepatology 2000; 31(2): 358-363 Indexing Status Subject indexing assigned by NLM MeSH Cost-Benefit Analysis; Endoscopy; Endoscopy, Digestive System; Esophageal and Gastric Varices /complications; Gastrointestinal Hemorrhage /prevention & Health Care Costs; Humans; Ligation; Portasystemic Shunt, Transjugular Intrahepatic /economics; Sclerotherapy; Secondary Prevention; Sensitivity and Specificity; control AccessionNumber 22000000350 Date bibliographic record published 31/12/2000 Date abstract record published 31/12/2000 |
|
|
|