Analytical approach:
A discrete event simulation model was developed in order to estimate, on the basis of individual simulations, the costs and benefits of the two strategies using published evidence. The time horizon of the analysis was 5 years. The authors stated that the perspective was that of the health care system.
Effectiveness data:
The clinical data were derived from two recent, relevant studies. The first one was a randomised clinical trial (RCT) of 2,521 patients receiving ICD, amiodarone or placebo. The second was a meta-analysis of 13 trials of amiodarone. Other clinical data, such as adverse events, were derived from manufacturers or other published studies. The case-fatality rate of a severe arrhythmia according to treatment was the main clinical input of the model. Several equations were used to estimate the individual risk of clinical events.
Monetary benefit and utility valuations:
The value of a life saved was calculated using the willingness-to-pay (WTP) approach. Sources for WTP for immediate loss of life were obtained from the Department of Transport for the UK, and from research commissioned by the French government for France.
Measure of benefit:
The benefit measure was the monetary value of the number of lives saved. An annual discount rate of 3.5% was used for the UK and 3% for France.
Cost data:
The main cost categories were hospitalisations (for initial and subsequent operations), ICD (lead and device) and medications (optimal therapy, amiodarone and treatment of toxicity). The costs in the two countries were derived from national health agencies and national drug databases. Resource consumption was derived from published evidence, but no details of this were provided. The costs were in euros (EUR) and the price year was 2004. The annual discount rate was 3.5% in the UK and 3% in France.
Analysis of uncertainty:
A univariate deterministic sensitivity analysis was undertaken on key model inputs such as age, case-fatality rates, arrhythmia rates and hospitalisation rates. The ranges of values under examination were presumably based on published data. A threshold analysis was conducted in order to estimate the WTP break-even point at which the cost-benefit ratio was equal to 1 (costs equal to benefits).