Modeling the effect of public health resources and alerting on the dynamics of pertussis spread

We consider the response of a local health department (LHD) to a pertussis outbreak using a composite discrete event simulation model with a stochastic branching process. The model captures the effect of epidemiologic spread of disease as a function of the health alert levels and the resource availability of the LHD. The primary response mode in the model is contact tracing that is assumed to be a resource-based delay with an iterative tracing policy. The effect of the threshold for initiating contact tracing and its relationship with the resource availability of the LHD is explored. The model parameters associated with contact tracing are estimated using North Carolina (NC), U.S.A. pertussis case data and data from the NC Public Health Information Network. The infectivity parameters are derived from literature. The results suggest that the time to initiate contact tracing significantly affects the magnitude and duration of the outbreak. The resource levels for contact tracing have less significant impact on the outbreak outcomes. However, when the nurse schedule is constrained, that is, if the total hours devoted to contact tracing a week is restricted, the effect of the resource level becomes significant. In fact, some outbreaks could not be controlled within the 1-year time limit of simulation.