Probability model on the use of sentinel animal monitoring for arbovirus.

We specify the relations among the major parameters involved in using sentinel animals to monitor arbovirus transmission. By using probability models, first we demonstrate that vector density, infection rate, incubation period of sentinel animals, the number of sentinel animals used, and blood sampling frequency could directly affect the observed seroconversion rates and the comparison of these rates. Second, we evaluate the reliability of using sentinel animals to monitor arbovirus transmission under various conditions. On the basis of the model, we propose some measures for enhancing the reliability of using sentinel animals to monitor arbovirus. Third, we present a formula for estimating the number of sentinel animals used for a study designed to compare seroconversion rates. Fourth, we derive a necessary condition for deciding the blood sample frequency of sentinel animals. Fifth, we develop a formula that could be used to obtain the vector infection rate by using seroconversion data. Finally, we evaluate the epidemiology consequences of heterogeneous exposure of human hosts to vectors by theoretical modeling.