Ebola has also appeared in the Democratic Republic of the Congo. The WHO report from 21 August reports 15 cases and 10 deaths, mostly in the town of Isiro. This outbreak is in no way related to the Ugandan outbreak from late July which we previously mapped; the virus is a different strain.

We again mapped the Congo, highlighting Isiro. This map shows all towns/villages as black dots, which clearly follow the road and river networks — not surprising.

Just as in Uganda, the road network (as a graph) has more of a lattice than a power-law structure. The highest degree node has degree 8, mean degree is 2.

We could condense the nodes around population centers. This would make a more interesting network. But it is not evident that this would provide a better model of disease spread. Ebola transmits via contact with contaminated bodily fluids. An infected person in a crowd is unlikely to expose the rest of the crowd (unlike measles or influenza). Thus a high-population node is in fact less likely to transmit the disease further, as the probability of direct contact with an infected person decreases when more people are in the mix.

Ultimately, it is the nature of the disease which dictates that it is an unlikely candidate for a national epidemic. It kills too quickly and transmits too poorly to establish a permanent presence in a human population. Note that the WHO

does not recommend that any travel or trade restrictions be applied to Democratic Republic of Congo. (Aug 21 report)

It isn’t just the disease and network structure which makes Ebola a minimal threat. The real heros are the brave staff of Doctors without Borders (MSF), who are again on the scene. Note that of the 10 reported deaths, 3 are health care workers. The MSF people are again taking a very real risk for very little reward. And not just in the DRC, they are everywhere, caring for war refugees in Sudan, providing one of the only hospitals in norther Afghanistan, helping to treat the more than 13,000 cases of cholera in Sierra Leone, …

Data for the map comes from Humanitarian Response. Code for the map is largely the same we used to map ebola in Uganda.

drcroad <- readShapeLines('DRC/cod_transportation/cod_rdsl_rgc')
drcwater <- readShapeLines('DRC/cod_transportation/cod_wtw_rgc')
drcnat<-readShapePoly('DRC/cod_polbnda_adm0_rgc')
drcvil <- readShapePoints('DRC/cod_location/cod_pplp1_rgc')

png(“DRC_roads.png”,width=1000,height=1000)
plot(drcnat)
plot(drcroad[drcroad$CATEGORIE==”Locale”,],add=TRUE,lwd=0.5,col=’lightgray’)
plot(drcroad[drcroad$CATEGORIE==”Regionale secondaire”,],add=TRUE,lwd=0.5,col=’burlywood3′)
plot(drcroad[drcroad$CATEGORIE==”Regionale principale”,],add=TRUE,lwd=1,col=’burlywood3′)
plot(drcroad[drcroad$CATEGORIE==”Nationale”,],add=TRUE,lwd=2,col=’burlywood1′)
plot(drcroad[is.na(drcroad$CATEGORIE),],add=TRUE,lwd=1,col=’burlywood3′)
plot(drcwater,add=TRUE,col=’steelblue’)
Isiro <- grep('Isiro',drcvil$NOM1,ignore.case=TRUE)
points(drcvil[Isiro,], pch=16,col=’red’,cex=0.75)
plot(drcvil,pch=16,cex=0.01,add=TRUE)
points(drcvil[Isiro,], pch=1,col=’red’,cex=1.5)
points(drcvil[Isiro,], pch=1,col=’red’,cex=2.5)
points(drcvil[Isiro,], pch=1,col=’red’,cex=3.5,lwd=0.75)
points(drcvil[Isiro,], pch=1,col=’red’,cex=4.5,lwd=0.5)
dev.off()