Yellow Fever

UND yellow fever model

Lead modellers: Alex Perkins

Institution(s): University of Notre Dame (UND)

Brief description of model:

The model represents zoonotic spillover with a Poisson process driven by spatially variable rates accounting for population, coverage, and environmental factors. It represents an urban transmission cycle with a branching process driven by spillover and with SIR dynamics in the event of a large outbreak. The model previously focused on South America, and has recently been extended to focus on Africa too.

Figure 1.Modeling framework schematic.

Our modeling framework involves five sequential steps that result in a set of eight ensemble models of the force of infection (FOI) of yellow fever virus, and associated deaths, in each of 477 first administrative-level units (adm1s) across 34 countries in Africa. Each of these eight ensemble models corresponds to a different assumption about the interpretation of serological data in Step 1 (Table 1). Colors associated with the six steps are used in subsequent figures to refer to the step to which those results pertain.

Key publication(s):

Link to publicly available code (where available):

___________________________________________________________________________________________________________________________________________________________________

Imperial yellow fever model

Lead modeller: Katy Gaythorpe
Link to all modelling group members

Institution(s): Imperial

Brief description of model:

To model yellow fever (YF) outbreak occurrence and burden, we use a new stochastic, dynamic model of yellow fever virus (YFV) transmission which incorporates both sylvatic spillover and human-to-human (urban) transmission. The key parameters for this model are the force of infection for sylvatic spillover and the basic reproduction number of human-to-human transmission. These are calculated from environmental covariates (including non-human primate species richness, land cover and vector temperature suitability) using parameters estimated within a Bayesian framework from YF data from currently-affected regions of Africa and South America. This model has also been used to examine the risk of outbreak propagation in at-risk countries, as well as forming the basis for examining the impacts of climate change on YF risk..

Map of median projected R0 across 1st-level sub-national administrative regions in selected African countries.

Key publication(s):

Link to publicly available code (where available):

___________________________________________________________________________________________________________________________________________________________________

FIOCRUZ yellow fever model

Lead modeller: Douglas Adriano Augusto

Link to all modelling group members

Institution(s): Oswaldo Cruz Foundation (FIOCRUZ)

Brief description of model:

To be updated.

Key publication(s):

Link to publicly available code (where available):

___________________________________________________________________________________________________________________________________________________________________