Presentation (CTI)

OBJECTIVES: Health economic models play a critical role in informing decision-makers, making accuracy essential. Therefore, conducting a comprehensive quality check (QC) of these models is vital. A good approach for performing QC is the independent model reprogramming. However, due to its time and resource demands, it is rarely performed, and quality control is often limited to running predefined checklists. This study aimed to assess the potential of an automated model generator to enhance QC.
METHODS: An automated generator was created using Visual Basic for Applications (VBA) and R to build health economic models. Users input parameters into a predefined MS Excel workbook and run an R script to generate the model in Excel. The tool operates on a fully deterministic algorithm, ensuring consistent results for a given input. The generator was employed during the QC of a Markov model with five health states, three-month cycle length and lifetime time horizon, taking a healthcare payer perspective. An independent rebuild of the model was conducted as one of phases of the QC, utilizing the tool.
RESULTS: Following a series of manual adjustments, the assessed model was successfully replicated. Initially, the results obtained through the rebuild differed from the original model. However, a comprehensive analysis of these discrepancies revealed issues in the original model that had not been identified in earlier QC phases. Once the issues in the original model were resolved, the difference between the results of the models was reduced to less than 0.001%. The entire replication process took three days.
CONCLUSIONS: The cost-effectiveness model generator proved to be useful in model QC by facilitating a rapid and less error-prone reconstruction of the model. This process helped identify issues in the original model that had been overlooked during manual checks.