Analytical approach:
The analysis was based on a decision analytic model, with a lifetime horizon. The authors stated that a societal perspective was adopted, but not all sources of costs could be included.
Effectiveness data:
The clinical evidence over 30 days came from the Enoxaparin and Thrombolysis Reperfusion for Acute Myocardial Infarction Treatment - Thrombolysis in Myocardial Infarction (ExTRACT-TIMI) 25 trial, which was a prospective, randomised, double-blind, double-dummy, parallel-group, multinational trial, involving 20,506 patients at 674 sites in 48 countries. The key endpoint was the composite of death from any cause and non-fatal recurrence of myocardial infarction, in the first 30 days after randomisation. Long-term data were estimated using published Framingham equations that were adapted to the ExTRACT-TIMI 25 trial patient characteristics and using Canadian life tables.
Monetary benefit and utility valuations:
Not considered.
Measure of benefit:
Life-years (LYs) were the summary benefit measure and they were discounted at an annual rate of 5%. The mean survival was calculated using data from the clinical trial and the Framingham Heart Study. The impact of baseline patient characteristics was accounted for by regression analysis.
Cost data:
The economic analysis included the costs of treatment (enoxaparin and unfractionated heparin), first hospitalisation, discharge to 30 days (major procedures, interventions, and management of severe adverse events), and long-term management costs. Drug costs were based on resource consumption observed in the clinical trial and Canadian unit prices. All the resources used over the first 30 days were based on the clinical trial. Hospitalisation costs and other costs incurred in the 30 days after discharge were from the clinical trial, US diagnosis-related group data, and the Ontario Case Costing Initiative. The long-term costs came from the Canadian literature. All medical costs, besides those in the 30-day trial and those associated with years of life lost, were assumed to be the same in the two arms of the trial. All costs were in Canadian dollars (CAD) and the price year was 2004. A 5% annual discount rate was applied.
Analysis of uncertainty:
Incremental cost-effectiveness ratios were analysed, using bootstrapping with 5,000 replications, and cost-effectiveness acceptability curves were generated. A deterministic analysis considered the variability in length of hospitalisation, using registry data, and years of life lost, using arbitrary estimates.