OBJECTIVES: Whilst the use of R in health decision sciences has recently been increasing, Excel is still being used very commonly for development of health economic (HE) models. In the HE literature and textbooks, there is a paucity of hands‐on, practical guidance for the construction of efficient Excel models. In this study, we wanted to: 1) highlight strengths and limitations of Excel for development of HE models, 2) provide hands‐on practical tips to improve Excel’s efficiency, 3) consider when it may be better to use alternative software.

METHODS: A targeted search was conducted to identify literature on the use / comparison of software for development of HE models. The Parexel HEOR modelling team convened a workshop to discuss findings of the literature review and share experience in April 2023.

RESULTS: Excel has many built-in statistical and econometric functions that can be further extended using VBA. Excel can, however, become slow when complex HE models or computationally demanding analyses must be run. A list of tips for structuring, formatting, and coding efficient HE models in Excel has been put together and will be presented. Modern programming languages are better suited to conduct complex, computationally demanding, and/or real time analysis. Clear examples include value of information analysis, model calibration in dynamic transmission models, and making real-time updates to statistical analyses of patient data and the resulting HE outcomes.

CONCLUSIONS: HE models are often built in Excel, owing to its wide-spread accessibility, user familiarity, and perceived transparency. Methodological and computational advances have allowed HE models to become more sophisticated and to better reflect clinical reality over time. Although modern programming languages are better suited to conduct complex, computationally demanding, and/or real-time analysis, ambiguity over the acceptance of such HE models by health technology assessment bodies remains a barrier to their adoption.