Predictive Modeling Guides Plague Prevention and Planning

Plague, a severe, primarily flea-borne bacterial zoonosis cause by Yersinia pestis, is characterized by long inactive periods punctuated by rapidly-spreading outbreaks. Humans are at greatest risk of exposure to Y. pestis during epizoonotic periods when infectious fleas abandon their dying rodent hosts and occasionally bite humans. The disease is often fatal is appropriate antibiotic treatment is delayed or inadequate. Improved assessments of when and where humans are at highest risk of exposure to Y. pestis may aid in targeting limited public health resources and ultimately reduce the burden of disease. Within the United States, the majority of human infections are reported from New Mexico. Following reports of human and veterinary cases, the New Mexico Department of Health (NM DOH) conducts field investigations to determine locations of probably exposure. Capitalizing on more than 40 years of epidemiological surveillance data, researchers in DVBID and the NM DOH developed a spatial risk model that identified approximately 17% of the state as high risk (see figure), suggesting that resource requirements for regular surveillance and control could be effectively targeted on less than 20% of the state. Similar spatial and temporal predictive models are currently under development for plague in East Africa, a region of the world that in recent decades has reported the largest number of human plague cases world-wide. Although our example focuses on plague, these predictive tools are applicable to other disease systems.