Analytical approach:
The analysis was based on a two-part decision model. The first part was a decision tree that covered the first 15 months of therapy, which used a Monte Carlo simulation to sample age and initial intraocular pressure of the patient cohort. The second part was a Markov model with six-month cycles that was applied to cover the rest of patients’ lives (given the overall lifetime horizon of the analysis). The authors stated that the perspective of the health care system was adopted.
Effectiveness data:
A selective approach appeared to have been used to identify relevant sources of data. Published systematic reviews were used whenever possible. Meta-analyses were used for treatment effect of all four drugs under analysis. Side effects were from several published studies not described. Patient characteristics came from a database of 1,000 charts of patients visiting a non-referral general ophthalmic practice in Maastricht. Changes in intraocular pressure values were the key inputs of the model; these were from the meta-analyses.
Monetary benefit and utility valuations:
Not considered.
Measure of benefit:
Years of blindness were used as the summary benefit measure and were defined as the mean expected time spent in blindness per person within 18.7 years of life expectancy. A 4% annual discount rate was applied.
Cost data:
The analysis included the costs of drugs, outpatient visits, perimetry, laser, transportation, and glaucoma therapy. Economic data were from various sources including official price lists, hospital databases, and previous published studies. Patterns of resource consumption were based on specialists’ opinion and the recommendations of the American Academy of Ophthalmology. Costs were in Euros (EUR). The price year was 2003. A 4% annual discount rate was applied.
Analysis of uncertainty:
One-way sensitivity analyses were carried out to vary key inputs of the model such as side effects of medication, cost of glaucoma therapy, and cost of latanoprost. In a two-way sensitivity analysis, the intraocular pressure-lowering effect of timolol and latanoprost was investigated.