Title: Value of information for HIV evidence synthesis
Authors: Daniela De Angelis - University of Cambridge (United Kingdom) [presenting]
Abstract: Annual national estimates of the number of people living with HIV in England, particularly those who are unaware of their infection, have, for several years, been based on a Bayesian model that combines evidence from multiple sources of surveillance, survey data and prior beliefs. In such an evidence synthesis model it is important to know which parameters most affect the estimates and, therefore, the decision from the model; which of the parameter uncertainties drive the decision uncertainty; and what further data should be collected to reduce such uncertainty. These questions can be addressed by Value of Information (VoI) analysis, allowing estimation of the expected gain from learning specific parameters or collecting data of a given design. We introduce the concept of VoI for Bayesian evidence synthesis, using and extending ideas from health economics, computer modelling and Bayesian design. We then apply it to our HIV prevalence model. Results show which parameters contribute most to the uncertainty about each prevalence estimate, and the expected improvements in precision from specific amounts of additional data. These benefits can be traded with the costs of sampling to determine an optimal sample size.