Eliminating Cholera in Haiti: Combining Dynamic Models with Data to Inform Vaccination Policy

• Activity Number: 523 - Real-World Epidemiological Studies
• Type: Contributed
• Date/Time: Thursday, August 11, 2022 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Jesse Wheeler* and Anna Rosengart and Josh Jiang and Edward Ionides
• Companies: University of Michigan and University of Michigan and University of Michigan and University of Michigan
• Keywords: Partially Observed Markov Process; Disease Modelling; Haiti; Cholera; Forecasting; Vaccinations

Abstract

Public health decisions must be made about when and how to implement interventions to control an infectious disease epidemic. These decisions should be informed by data on the epidemic as well as current understanding about the transmission dynamics. Such decisions can be posed as statistical questions about scientifically motivated dynamic models. As a case study, we consider a cholera epidemic in Haiti. The 2010 introduction of cholera to Haiti led to an extensive outbreak and sustained transmission, eliminated in 2019 with the help of vaccination and other public health measures. We study four models developed by expert teams to advise on vaccination policies. We assess methods used for developing, fitting, and evaluating these models, leading to recommendations for future studies. Methods that lead to better statistical fit include (i) benchmarking mechanistic models against associative statistical models; (ii) ensuring models have sufficient stochasticity to represent uncertainty in the system; (iii) additional attention to numerical difficulties related to the challenge of investigating the likelihood surface.