Recent advances in health economic modelling: moment-generating functions, fast Fourier transforms and the Delta-PSA method

Summary: Dr Snowsill will present three recent advances in health economic modelling. The moment-generating function method [Snowsill TM. Medical Decision Making 2019; 39(5):523–539] facilitates the calculation of discounted lifetime outcomes (e.g., costs and QALYs) allowing the times of transitions between health states to be modelled flexibly (related to semi-Markov modelling). The fast Fourier transform method [Snowsill TM. Submitted (9 April 2020) to Medical Decision Making] takes an alternative approach to solving a similar problem, and additionally allows the flexibility to incorporate mortality based on time since the model origin (e.g., age-dependent mortality). The Delta-PSA method [Snowsill TM. Submitted (28 May 2020) to SMDM 42nd Annual North American Meeting] is an alternative to the standard Monte Carlo approach for probabilistic sensitivity analysis based on Taylor series expansion, and has been demonstrated with a moment-generating function model and a Markov model.

This talk is similar to (but not exactly like) the presentation given at the above TEEHTA seminar.