Analytical approach:
A probabilistic Markov model with a 10-year horizon was constructed to model the ongoing risk of adverse events, relapse, and developing diabetes. The authors reported that the perspective of the UK National Health Service (NHS) was adopted.
Effectiveness data:
The clinical data were mainly from the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) and these were augmented by data from other randomised controlled trials (RCTs), non-randomised studies, and administrative databases. The main clinical parameters included the treatment discontinuation rates, relapse rates, and adverse events, such as diabetes, extrapyramidal symptoms, weight gain, and hyperprolactinaemia.
Monetary benefit and utility valuations:
The utility values for each health state were from a UK study.
Measure of benefit:
Quality-adjusted life-years were the measure of benefit and benefits were discounted at an annual rate of 3.5%.
Cost data:
The economic analysis included the costs of medications, stable schizophrenia, general practitioner and psychiatrist visits, relapse (including days in hospital), diabetes, and medications for extrapyramidal symptoms, hyperprolactinaemia, and weight gain. The unit costs and resource use data were reported separately and they were derived from published sources. All costs were reported in UK pounds sterling (£) for the price year 2006. They were discounted at an annual rate of 3.5%.
Analysis of uncertainty:
A probabilistic sensitivity analysis was undertaken, by assigning probability distributions to the model inputs. Cost-effectiveness acceptability curves and frontiers were generated, using the net benefit method. Deterministic one-way sensitivity analyses were performed on: the risk of extrapyramidal symptoms, the relapse risk, the discontinuation rate, the risk of weight gain and diabetes, the time horizon, and the length of hospital stay after relapse.