How to find natural reservoir hosts from endemic prevalence in a multi-host population: A case study of influenza in waterfowl

Abstract

The transmission dynamics of infectious diseases critically depend on reservoir hosts, which can sustain the pathogen (or maintain the transmission) in the population even in the absence of other hosts. Although a theoretical foundation of the transmission dynamics in a multi-host population has been established, no quantitative methods exist for the identification of natural reservoir hosts. For a host to maintain the transmission alone, the host-specific reproduction number (U), interpreted as the average number of secondary transmissions caused by a single primary case in the host(s) of interest in the absence of all other hosts, must be greater than unity. If the host-excluded reproduction number (Q), representing the average number of secondary transmissions per single primary case in other hosts in the absence of the host(s) of interest, is below unity, transmission cannot be maintained in the multi-host population in the absence of the focal host(s). The present study proposes a simple method for the identification of reservoir host(s) from observed endemic prevalence data across a range of host species. As an example, we analyze an aggregated surveillance dataset of influenza A virus in wild birds among which dabbling ducks exhibit higher prevalence compared to other bird species. Since the heterogeneous contact patterns between different host species are not directly observable, we test four different contact structures to account for the uncertainty. Meeting the requirements of U > 1 and Q < 1 for all four different contact structures, mallards and other dabbling ducks most likely constitute the reservoir community which plays a predominant role in maintaining the transmission of influenza A virus in the water bird population. We further discuss epidemiological issues which are concerned with the interpretation of influenza prevalence data, identifying key features to be fully clarified in the future. © 2009 Elsevier Inc. All rights reserved.