Using Networks to track Ebola
After the lecture on epidemics I was interested in looking into how real life public health experts might use the concepts we learned about in class to map the spread of epidemics. Ebola is an epidemic that has come into the news a lot with its major 2015 breakout and the smaller, more recent one. Additionally, because it has no airborne transmission vector it seemed particularly relevant to our model that involves contacts between nodes. Sure enough a study was done on this matter using a similar model:
https://medicalxpress.com/news/2015-11-reveals-deadly-route-ebola-outbreak.html
The study was done by the Columbia Mailman school of public health and was published in the Journal of the Royal Society Interface. Kenema and Port Loko, two cities noted in the report functioned as major points in the outbreak because of their major amount of connections to other cites. By our model, Ro = pk. As K, the number of contacts increases, the basic reproductive number increases. The report acknowledges that while too late to help with this outbreak, understanding Kenema and Port Lokos strategic location could help with stopping future outbreaks. As they have multiple edges to other communities, they would be ideal points at which to focus efforts on stopping the disease and it would make sense to allocate most resources to that area. Report also stresses the importance of maintain proper records on medical data. In our model p standards for probability of transmission and each node carries a probability of P. Thus when calculating final totals for reproductive numbers, its important to know how many nodes are there. This is done by something called contact tracing where you go through records painstakingly to see whom everyone has been in contact with recently and interviewing them about where they went. At the time of the 2015 outbreak, Sierra Leone’s medical system was in collapse and records were scrambled. This lack of data bout nodes made containing the outbreak even harder than it already was.