Global sensitivity analysis for stochastic models of infectious diseases: Re-emergence of yellow fever virus in the Americas as a case study.

Uncertainty in patch connectivity and non-human primate latency most strongly drives variation in cumulative yellow fever virus incidence across patches in a stochastic spillover model for the Americas.

• Why it matters: Outbreaks begin with few infections, so stochastic effects and sampling noise can obscure inference, limiting deterministic approaches and making parameter uncertainty hard to quantify efficiently.
• What they did: The authors apply uncertainty quantification tools to stochastic Markov chain–type disease models to separate stochastic process noise from sampling noise, then analyze which parameters most affect cumulative incidence.
• The result: In the Americas spillover case, only a handful of parameters’ uncertainties substantially shape cumulative disease incidence, with patch connectivity and non-human primate latency having the greatest impact across patches.

AI summary of the abstract · may contain errors · not medical advice · read the paper.