Analytical approach:
A Markov model was used to combine data from a single study with data from published evidence. This model followed the psoriasis assumptions from the health technology assessment (HTA) report undertaken by the Centre for Reviews and Dissemination (CRD), University of York, UK. The time horizon of the study was 10 years. The perspective was that of the Ontario Ministry of Health, Canada.
Effectiveness data:
Effectiveness data was from the 12-week Active Comparator Psoriasis Trial (ACCEPT). This was a phase three multicentre randomised controlled trial. The study included 903 adult patients who were randomised between etanercept (347 patients), ustekinumab 45mg (209 patients) and ustekinumab 90mg (347 patients). Only the ustekinumab 45mg group was used in the model. The study had a follow-up period of 12 weeks. The main clinical effectiveness estimate was the Psoriasis Area and Severity Index (PASI) 75 score collected from patients at week 12.
The trial results were extrapolated for the model using published literature and a Delphi panel of Canadian experts.
Monetary benefit and utility valuations:
Utility gains were applied to patients regardless of treatment arm in accordance with the PASI change from baseline achieved in the ACCEPT study. PASI data did not provide utility scores, so the Dermatology Life Quality Index (DLQI) index was transformed to utility values through linear regression. The regression analysis used data from the Health Outcomes Data Repository where both EuroQol-5D (EQ-5D) and DLQI were collected from patients. ACCEPT did not collect DLQI data, so the scores were obtained from the two pivotal Phase 3 randomised double-blind placebo-controlled trials of ustekinumab in similar patients (PHOENIX 1 and PHOENIX 2) to provide patient-level data for the regression analysis.
Measure of benefit:
The benefit measure was Quality Adjusted Life Years (QALYs) discounted at an annual rate of 5%.
Cost data:
The main cost categories in the study were outpatient visits (dermatology visit), monitoring (lab tests for full blood count, liver functioning and urea and electrolyte test) and drugs. Costs came from the Ontario Health Insurance Plan. Resource use was estimated by a panel of experts and validated using a previous study. Costs were presented in Canadian dollars (CAD) and discounted at an annual rate of 5%.
Analysis of uncertainty:
One-way sensitivity analyses were conducted to examine the influence of uncertainty on the incremental cost-effectiveness ratio. Probabilistic sensitivity analysis was conducted by varying model parameters across their potential distributions over 10,000 iterations. A scatter plot was used to presents the results.