Analytical approach:
The analysis used a published Markov model with a 10-year horizon. This published model for a severe rheumatoid arthritis population was adapted to represent an early rheumatoid arthritis population. It had five health states based on functional capacity. The transitions between states occurred every six months. A reduction in the dose of etanercept was possible for patients with a disease activity score (DAS28) of less than 2.6 for two consecutive six-month cycles. The authors stated that the perspective of society was adopted.
Effectiveness data:
The clinical data came from various sources, including a randomised controlled trial comparing the combination against methotrexate alone for early rheumatoid arthritis. This Combination of Methotrexate and Etanercept in active early rheumatoid arthritis (COMET) trial supplied the treatment effectiveness. The South Swedish Arthritis Treatment Group (SSATG) registry provided the effectiveness of the second biologics and their impact on functional capacity (Health Assessment Questionnaire, HAQ, score) and DAS28. The impact of a reduced dose of etanercept came from a cohort study (PADOVA). The key model inputs were the change in HAQ score, the change in DAS28, the efficacy of dose reduction, and the percentage of patients switching at discontinuation.
Monetary benefit and utility valuations:
The utility values were derived from a survey of the population in Southern Sweden, using the European Quality of life (EQ-5D) questionnaire.
Measure of benefit:
Quality-adjusted life-years (QALYs) were the summary benefit measure and they were discounted at 3% per annum.
Cost data:
The analysis considered the health care costs, community services, patient costs, and productivity losses. The resource data came from the population survey that provided the utility data. All costs were presented in Euros (EUR) for the year 2008 and a 3% annual discount rate was applied.
Analysis of uncertainty:
A probabilistic sensitivity analysis, using 300,000 simulations, was carried out. The key inputs that were varied in the deterministic sensitivity analysis were the time horizon, the perspective, the discontinuation rate, the proportion of patients switching or returning to full dose, and the utility score.