Sensitivity analyses
The robustness was evaluated with one-way sensitivity analyses (SA) of base case parameters with the greatest level of uncertainty such as utilities, hospital-related costs, and the probability of hospitalization. Tornado diagrams were used as graphical method for displaying one-way sensitivity analyses. Probabilistic sensitivity analysis was made using the Monte Carlo technique with a simulation of a hypothetical cohort of 10 000 patients in which each parameter varied randomly according to certain distributions (beta distribution in the case of probabilities, and gamma distribution in the case of costs); to generate expected cost utilities with a 95% confidence intervals (95% CI). To evaluate the uncertainty surrounding the cost-effectiveness of MS a cost-effectiveness acceptability curve was used. We estimated the population expected value of perfect information to inform the expected cost of uncertainty (expected opportunity loss surrounding the decision) (25). Microsoft Exel ®was used in all analyzes.