PHRM 231 Lecture Notes - Lecture 10: Monte Carlo Method, Sensitivity Analysis
Document Summary
Find the next move with the lower icer. Stop when there is no strategy, or next jump"s icer > wtp. Usual care must be an option and should be defined. Evidence is always accompanied by uncertainty (e. g. sd, se, 95% ci). The uncertainty in evidence needs to be propagated to the uncertainty in the outcomes. Change the value of parameters across a defined range and see how icer changes. Often done for one parameter at a time (one-way sensitivity analysis) Also two-way sensitivity analysis (changing two parameters simultaneously) A model is a mathematical function on model parameters that returns payoffs as outputs. Assign probability distribution to each of the input parameters. Repeatedly run the model, each time with a random draw from the distribution of input parameters. Calculate icer as average difference in costs over average difference in effectiveness. Cheers checklist : recommendations to optimize reporting of health economic evaluations.