Analytical approach:
The analysis was based on a decision-tree model, with a one-year horizon. The author stated that the perspective was that of the third-party payer, which was the UK NHS.
Effectiveness data:
The clinical data were from a literature review in PubMed and the Cochrane Library. The numbers of studies identified and selected were reported in an online appendix, with the key characteristics of each study. The treatment effect (baseline risk and relative risk) was mainly from randomised controlled trials. A meta-analysis was performed, using the inverse variance method, to pool the evidence from the selected studies. No head-to-head studies were found and an indirect comparison was needed. The adverse event information was from clinical trials. The response rates were the key input for the model.
Monetary benefit and utility valuations:
The utility values were from studies found in the PubMed database. The estimates for patients with actinic keratosis were from a study that used the standard gamble technique and a study that used the time trade-off method. The disutilities associated with adverse events were from studies of patients with conditions of similar severity.
Measure of benefit:
Quality-adjusted life-years (QALYs) were the summary benefit measure.
Cost data:
The economic analysis included the costs of imiquimod, PDT, and out-patient consultations (initial and follow-up visits and treatment for adverse events). The unit costs were from NHS reference costs and the British National Formulary. The resource quantities were based on a conventional treatment protocol. A specific cost for PDT was not identified and the cost of laser destruction of skin lesions was used. All costs were in UK pounds sterling (£) and the price year was 2006.
Analysis of uncertainty:
A probabilistic sensitivity analysis was undertaken, using 5,000 Monte Carlo simulations and conventional probability distributions for the model inputs. Cost-effectiveness acceptability curves were created. One-way sensitivity analyses were conducted on selected inputs.