Evaluating empirical contact networks as potential transmission pathways for infectious diseases

Abstract

Networks are often used to incorporate heterogeneity in contact patterns in mathematical models of pathogen spread. However, few tools exist to evaluate whether potential transmission pathways in a population are adequately represented by an observed contact network. Here, we describe a novel permutation-based approach, the network k-test, to determine whether the pattern of cases within the observed contact network are likely to have resulted from transmission processes in the network, indicating that the network represents potential transmission pathways between nodes. Using simulated data of pathogen spread, we compare the power of this approach to other commonly used analytical methods. We test the robustness of this technique across common sampling constraints, including undetected cases, unobserved individuals and missing interaction data. We also demonstrate the application of this technique in two case studies of livestock and wildlife networks. We show that the power of the k-test to correctly identify the epidemiologic relevance of contact networks is substantially greater than other methods, even when 50% of contact or case data are missing. We further demonstrate that the impact of missing data on network analysis depends on the structure of the network and the type of missing data.

Role of Live-Duck Movement Networks in Transmission of Avian Influenza, France, 2016–2017

“…A new aspect of this study is the inclusion of a permutation-based approach based on the dates of holding infection to evaluate whether the acquisition of holding infection was consistent with virus transmission through the network. This approach has been limitedly applied in the epidemiology of infectious diseases (23)(24)(25), although it outperforms other degreebased statistical methods, such as logistic regression and nonparametric tests. Outcomes about the relative contribution of movement and proximity networks represent a required basis on which predictive models of HPAI spread could be developed.…”

“…Next, we assessed the likely contribution of liveduck movements in the distribution of H5N8 outbreaks in the network using a permutation-based approach (23)(24)(25). The rationale behind this approach was that if the outbreaks resulted from disease spread through the movement networks, the mean number of infected holdings in contact with an infected holding in the network would be significantly greater than expected if infected holdings were randomly distributed in the network.…”

Burkholderia mallei : The dynamics of networks and disease transmission

“…To test the hypothesis of direct association between animal movement and B. mallei, the k-test was used (VanderWaal, Enns, Picasso, Packer, & Craft, 2016). This method calculates the mean number of possible outbreaks that may occur within n steps from an infected node.…”

A multi-species repository of social networks

“…Network analysis is rapidly becoming a central approach in several basic and applied research areas of ecology and evolutionary biology, including behavioral ecology, epidemiology, spatial ecology, and social evolution [14]. Recently, researchers have demonstrated the utility of network analysis in explaining the transmission of social information in animal groups [13,4], evolution of cultural behavior [17], epidemiological consequences of group substructure [19], mechanisms of infectious disease transmission in animal groups [22,19,8], and emergence of collective behavior [5]. Network analysis leverages detailed social interaction and movement data and allows for the incorporation of heterogeneity at the individual scale to explain population level processes, as well as the ability to objectively quantify the organization of social interactions and dynamics of group behavior.…”