Accounting for both heterogeneity and biases in economic evaluations presents difficult challenges, which can be addressed by simulation models as discussed in Fernando Alarid-Escudero’s recent CHDS seminar. Alarid-Escudero is an Assistant Professor of Health Policy at Stanford University School of Medicine. Alarid-Escudero’s research uses discrete-event simulation modeling techniques to quantify the opportunity cost of study biases stemming from non-rigorous or poorly transferable research.
Alarid-Escudero explained how the different estimates of the value of information (VOI) can be used to assess the value of eliminating or reducing the uncertainty of all or some parameters included in a decision model. He highlighted the potential presence of bias in VOI estimates that come from using data from non-rigorous or poorly transferable research studies for population health policymaking. These studies provide average treatment effect estimates from an average patient. However, these estimates might not be ideal for population-level decision-making because of bias due to the heterogeneity of preferences and characteristics between individuals and groups, as well as contextual differences between the empirical studies and the health-policy setting. If these estimates are used, one-size-fits-all policies that are not suitable for everyone might be favored. To address this, he proposes a framework to address uncertainty, heterogeneity, and bias using causal inference methods and discrete event simulation models to account for both heterogeneity and biases in economic evaluations and quantify the opportunity cost of study biases.
Learn more: Read the publication, Value of Information: Success Stories. In Value of Information for Healthcare Decision-Making
Learn more: Read the publication, Computing the Expected Value of Sample Information Efficiently: Practical Guidance and Recommendations for Four Model-Based Methods
Learn more: Read the publication, The Curve of Optimal Sample Size (COSS): A Graphical Representation of the Optimal Sample Size from a Value of Information Analysis
Learn more: Read the publication, A Value of Information Analysis of Research on the 21-Gene Assay for Breast Cancer Management
Related news: Breast Cancer Management Seminar