Analytical approach:
The analysis was based on a discrete-event simulation with a five-year time horizon. The authors stated that the analysis took the perspective of the UK health care payer.
Effectiveness data:
The sources included various double-blind, randomised, placebo-controlled trials, published meta-analyses, and other studies. In particular, baseline risk was based on the placebo arms of various quetiapine clinical trials, while relative risk was taken from two published meta-analyses. Given the lack of head-to-head comparisons for some drugs, indirect comparisons were made using lithium as common comparator. Side-effects were from published trials. The primary end point was the efficacy of treatment, which was modelled based on the relative risk of events with active pharmacological therapy compared with placebo.
Monetary benefit and utility valuations:
The utility values were from published studies that were identified through a literature review. One of the key sources elicited the utility estimates from patients using the standard gamble method. Another source used the European Quality of life (EQ-5D) instrument and UK tariffs.
Measure of benefit:
Quality-adjusted life-years (QALYs) were the summary benefit measure and were discounted at an annual rate of 3.5%.
Cost data:
The economic analysis included the costs of drugs and the direct medical costs associated with the treatment of mood events (mania, depression, and remission) and the management of selected adverse events (weight gain and extrapyramidal symptoms). The quantities of resources used were based on clinical trials and a National Institute for Health and Clinical Excellence (NICE) guideline on bipolar disorder. This latter source was used to derive some of the unit costs. Drug costs were taken from the UK NHS Electronic Drug Tariff. Costs were in UK £. The price year was 2011. A 3.5% annual discount rate was applied.
Analysis of uncertainty:
One-way sensitivity analyses were carried out to see how robust the model outcomes were to variations in selected inputs such as costs, start age, time horizon, treatment duration, disutilities and costs of discontinuation and adverse events, dosage of quetiapine and quetiapine as additional therapy. Indirect costs associated with sick leave were included in a sensitivity analysis, using data from a published study. A probabilistic sensitivity analysis was carried out by assigning distributions to the model inputs, excluding unit costs, which were generally based on fixed estimates from price lists.