The economic evaluation used a Markov model, namely the quit benefits model (QBM), which predicted the benefits of smoking cessation in quitters taking account of four main smoking-associated diseases. The time horizon of the model was a lifetime (up to the age of 85 years). The authors stated that the perspective of the health care system was adopted.
The clinical data came from a selection of known, relevant sources, including published studies and the report of the programme implementation which gathered data by means of household telephone surveys. Around 2,000 Australians aged 18 and over were interviewed in May 1997 and around 4,200 were interviewed in November 1997. The bulk of the evidence came from Australian sources and was supplemented by data from international studies when required. The primary clinical outcome was the reduction in smoking prevalence due to the anti-smoking campaign.
Monetary benefit and utility valuations:
The utility estimates came from two published sources, which were a meta-regression of 20 studies and an international registry. No other details were given.
Measure of benefit:
The two summary benefit measures were life-years (LYs) and quality-adjusted life-years (QALYs). Both were calculated using the QBM and were discounted at an annual rate of 3%. The number of avoided cases of the four smoking-associated diseases, which were lung cancer (LC), acute myocardial infarction (AMI), stroke, and chronic obstructive pulmonary disease (COPD), were also reported.
The economic analysis considered two main categories of costs, which were the cost of the programme implementations, and the cost saved due to the reduced incidence of the four smoking-associated diseases. A breakdown of cost items was not provided. All the costs and quantities were based on Australian data, except for the cost of COPD which was derived from a Canadian study. All costs were in Australian dollars (AUD) and were discounted at an annual rate of 3%. The price year was 2001.
Analysis of uncertainty:
In a deterministic analysis, different time horizons were considered. In an alternative scenario, it was assumed that only half the reduction in smoking prevalence was due to the NTC. A first-order Monte Carlo simulation was used to generate confidence intervals around the mean outcomes of the model.