Deriving input parameters for cost-effectiveness modeling: taxonomy of data types and approaches to their statistical synthesis

Background: The evidence base informing economic evaluation models is rarely derived from a single source. Researchers are typically expected to identify and combine available data to inform the estimation of model parameters for a particular decision problem. The absence of clear guidelines on what data can be used and how to effectively synthesize this evidence base under different scenarios inevitably leads to different approaches being used by different modelers.

Objectives: The aim of this article is to produce a taxonomy that can help modelers identify the most appropriate methods to use when synthesizing the available data for a given model parameter.

Methods: This article developed a taxonomy based on possible scenarios faced by the analyst when dealing with the available evidence. While mainly focusing on clinical effectiveness parameters, this article also discusses strategies relevant to other key input parameters in any economic model (i.e., disease natural history, resource use/costs, and preferences).

Results: The taxonomy categorizes the evidence base for health economic modeling according to whether 1) single or multiple data sources are available, 2) individual or aggregate data are available (or both), or 3) individual or multiple decision model parameters are to be estimated from the data. References to examples of the key methodological developments for each entry in the taxonomy together with citations to where such methods have been used in practice are provided throughout.

Conclusions: The use of the taxonomy developed in this article hopes to improve the quality of the synthesis of evidence informing decision models by bringing to the attention of health economics modelers recent methodological developments in this field.