Implications of the cattle trade network in Cameroon for regional disease prevention and control

Abstract

Movement of live animals is a major risk factor for the spread of livestock diseases and zoonotic infections. Understanding contact patterns is key to informing cost-effective surveillance and control strategies. In West and Central Africa some of the most rapid urbanization globally is expected to increase the demand for animal-source foods and the need for safer and more efficient animal production. Livestock trading points represent a strategic contact node in the dissemination of multiple pathogens. From October 2014 to May 2015 official transaction records were collected and a questionnaire-based survey was carried out in cattle markets throughout Western and Central-Northern Cameroon. The data were used to analyse the cattle trade network including a total of 127 livestock markets within Cameroon and five neighboring countries. This study explores for the first time the influence of animal trade on infectious disease spread in the region. The investigations showed that national borders do not present a barrier against pathogen dissemination and that non-neighbouring countries are epidemiologically connected, highlighting the importance of a regional approach to disease surveillance, prevention and control. Furthermore, these findings provide evidence for the benefit of strategic risk-based approaches for disease monitoring, surveillance and control, as well as for communication and training purposes through targeting key regions, highly connected livestock markets and central trading links.

Figure 1. Study area and locations of the livestock markets included in the analysis.

The Regions in yellow highlight the areas where data collection was carried out. Regions in gold are the areas of Cameroon that were not visited during the study but for which were identified trading links with livestock markets localized in the sampling areas. In grey are highlighted Regions of neighboring countries where livestock markets outside Cameroon are connected with the trade network (Generated using R statistical software (version 3.2.3) using the raster, rgdal and ggplot2 packages, and shp files obtained from the GADM database of Global Administrative Areas (source: www.gadm.org).

Figure 2. Cattle flow in the livestock market system in Cameroon.

The origin and destination of traded cattle are represented in this circular map. Each sector of the circle represents a Region of Cameroon or a neighboring country involved by the cattle flow in the region. Outgoing animal flow starts from the base of each sector while received movements do not reach the base of the relevant sector. For instance, the North-West Region, in red, trades most of its animals within the same Region and only very small proportion of them are moved to the Littoral, Central, West and South-West Regions of Cameroon or are introduced from the Adamawa Region.

Figure 3. Cattle trading network in Cameroon and neighbouring areas.

Node sizes in the map are weighted by the relative value of eigenvector centrality. Ties show the direction of the cattle movement as indicated by the arrows and the proportional volume of traded animals is indicated by the thickness of the arrow (Generated using R statistical software (version 3.2.3) using the raster, rgdal and ggplot2 packages, and shp files obtained from the GADM database of Global Administrative Areas (source: www.gadm.org).

Figure 4. Volume of cattle traded through the network by month over a 12 month period.

Blue bars refer to the rainy season (May to September) while yellow bars refer to the dry (October to April). Between September 2013 and August 2014, a total of 252,831 cattle were moved through the network.

Figure 5. Key-actors analysis on the markets network.

Correlation between eigenvector centrality and betweenness centrality is identifying markets with different roles within the trading network. “Gate-keepers”, in the bottom-right quadrant, are central entities in terms of their ability to bridge between the functional nodes of the network and wider community of nodes. In the top-left quadrant, “pulse-takers” are the nodes with the shortest paths to all other nodes having easy access to the other central markets as well as to the rest of the network. Nodes in the top-right quadrant have both abilities. Markets in the bottom left quadrant tend to have no particular role. The size of the labels is relative to the value of their residuals obtained through linear regression of betweenness over eigenvector centrality, indicating the extent of their deviation from a linear relationship.

Figure 6

Effectiveness of targeted removal of nodes and connections over the GSCC (a) and the GWCC (b) in the cattle trade network. The y axis shows the size of the largest strong (a) and weak (b) component expressed as a percentage and the x axis shows the percentage of nodes, or connections, removal from the network. The effect of node removal, driven by the different centrality measures on the fragmentation of the components, is shown by the different colors: betweenness centrality in red, eigenvector in dark green, in-degree in light blue, out-degree in purple and the residuals from the regression of betweenness over eigenvector centrality in light green. Effect of link removal depending on their edge betweenness scores is showed by the light orange line. Random removal of nodes over 1000 simulations is showed with the median value (pink line) and its 95% range (gray shaded area).

Figure 7

Effectiveness of targeted removal of nodes over the size of the biggest communities (a) and over the number of communities (b) in the cattle trade network. The y axis shows the percentage variation in the size of the biggest communities (a) and in the number of communities (b) of nodes within the network and the x axis shows the percentage of node removal. The effect of node discharge, driven by the different centrality measures on the number of communities present in the network, is shown by different colors: betweenness centrality in red, eigenvector in light orange, in-degree in light blue, out-degree in purple and the residuals from the regression of betweenness over eigenvector centrality in light green. Random removal of nodes over 1000 simulations is shown by the median value (pink line) and its 95% range (gray shaded area).