A number of determinants underlie the choice of software to develop an economic model. A modeller’s prior experience and skill in a given software and the need for the model to be intelligible to another party are often key reasons to select one software over another, rather than selecting the software on the merits of its functionality. There is limited reporting on the methods and benefits of combining the output of different software packages within a single model to take advantage of the merits of different software.

Two software packages were used to develop a large diagnostic decision-tree model that fed into a series of Markov models to estimate costs and effects over a lifetime horizon. A key objective of using both Microsoft Excel (Excel) and TreeAge Pro (TreeAge) software was to expedite model development.

All Markov models were programmed in TreeAge, whilst the inputs and decision-trees were developed in Excel. The calculations from TreeAge were exported into Excel and linked to Excel input sheets.

Modelling across two software platforms allowed for a large model to be worked on simultaneously and independently by two modellers experienced in each software. The Markov model calculations were easily transported into Excel and input references were quickly updated due to agreed naming conventions for TreeAge variables and named cells in Excel. Formula checks in the Markov model were not required due to the accuracy of the imported calculations.

Through combining software, we could exploit the technical validity of the TreeAge calculations whilst allowing for a user interface in Excel. We also believe the overall time of model build was reduced (both in terms of overall hours spent and project time frame). A combination of software programmes should be considered within model development where data and calculations can easily be transferred between platforms.