Analytical approach:
Two analyses of overall survival were presented. The main analysis used a decision tree to model the clinical events for one year after initial treatment. These outcomes were extended to a lifetime horizon (maximum 100 years) by a two-state (alive or dead) Markov model. The alternative analysis used a decision tree for three years, followed by the same Markov model. The perspective was stated to be that of the UK NHS.
Effectiveness data:
The key clinical data were survival and the risk of adverse events, including bleeding, ischaemic stroke, or repeat myocardial infarction. These data were from the HORIZONS-AMI trial, which was an international, prospective, randomised, open-label, phase III clinical trial. The authors stated that this was the only appropriate trial available in the literature. An intention-to-treat analysis was conducted. The prevalence of radial arterial access was higher in the UK, than in other countries, so the estimates from the HORIZONS-AMI trial were adjusted to reflect UK data. Long-term survival was based on data from the Nottingham Heart Attack Register. These were adjusted to the mean age of the HORIZONS-AMI patients, and for the increase in life expectancy, since the data were collected, using life-tables for England and Wales and the declining exponential approximation of life expectancy (DEALE) method.
Monetary benefit and utility valuations:
Two utility values were used one for the first year after an acute ST-segment elevation myocardial infarction, and the other for subsequent years. These were from a study of 229 consecutive myocardial infarction survivors at a UK centre. The EQ-5D was used, with UK tariffs and the time trade-off method.
Measure of benefit:
The measure of benefit was quality-adjusted life-years (QALYs). Benefits were discounted at a rate of 3.5% per annum.
Cost data:
The economic analysis included the costs of drugs, treatment procedures, adverse events (bleeding, ischaemic stroke, repeat myocardial infarction, and death), ward use, and long-term annual cardiovascular treatment. Most of the resource use was from the HORIZONS-AMI trial. The estimates for alternative glycoprotein IIb/IIIa inhibitors and hospital length of stay were adjusted to represent UK practice, using UK sources and authors' assumptions. Allowances were made for treatment wastage. The costs were from UK sources and were reported in UK £. The price year was 2009 to 2010, with costs inflated where necessary. Future costs were discounted at a rate of 3.5% per annum.
Analysis of uncertainty:
Deterministic sensitivity analysis was conducted by varying the key model parameters within given ranges. Probabilistic sensitivity analysis assessed the impact of overall uncertainty in the model inputs, by drawing parameter values from assigned distributions. Several scenario analyses were performed to assess the applicability of the main results to UK practice.