Online, semi-parametric estimation of optimal treatment allocations for the control of emerging epidemics

A key component in controlling the spread of an epidemic is deciding where, when, and to whom to apply an intervention. Here, we conceptualize the epidemic as spreading across nodes in an network. A treatment allocation strategy formalizes this process as a sequence of functions, one per treatment period, that map up-to-date information on the epidemic to a subset of nodes to receive treatment. An optimal treatment allocation strategy minimizes the expectation of some cumulative measure of harm, e.g., the number of infected individuals, the geographic footprint of the disease, the estimated total cost of the disease, or a composite outcome weighing several important measures. One approach to estimating an optimal allocation strategy is to model the underlying disease dynamics and then use to simulation—optimization. However, constructing a high-quality estimator of the complete system dynamics is difficult especially in the context of emerging epidemics where there is little scientific theory to inform a class of models. We derive estimating equations for the optimal allocation strategy that does not require a model the system dynamics. Furthermore, because this estimator does not require simulation of the disease process it is computationally tractable even for very large problems. We demonstrate the proposed methodology using data on the spread of white-nose syndrome in bats