Model of Ebola outbreak in Kikwit 1995

Model of Ebola outbreak in Kikwit 1995

This slider modifies the number of people involved in the outbreak. If more people enter the area then it will increase

Population at risk of contact

This slider modifies the intrinsic transmissibility of the pathogen. If there are more contacts between people or the contacts are more likely to result in transmission then it will increase

Beta

This slider modifies the duration of the infective period. Complete isolation of the sick will decrease the value, as would effective treatment of the illness.

Days infectious (1/gamma)

Mortality: % infected that die:

% of population that are naturally immune or vaccinated:

Model output

Blue line is a smooth fit to the data. Red line is the model output

Total cumulative deaths

This is the raw data that was used to fit the model.

Your task is to use the model as a guide to the possible changes to the outbreak that might have resulted if you had implemented interventions that were not actually used at the time. Interventions that move people into the area affected by the outbreak will increase the population at risk (slider 1). Interventions that affect transmission will change the beta parameter (slider 2). If you suggest measures that result in changes in the time that a person is infectious then this will affect gamma (slider 3). Treatements that result in improved outcomes after infection will lower mortality (slider 4). At this time there was no effective vaccine. However you can use slider 5 to model the possible impact of effective vaccination that leads to immunity.

Note that any combination of parameters that lowers R_0 to below 1 will lead to the outbreak being controlled from the start. Such a combination would be the most successful outcome. However this assumes that the model assumptions are met.

Some combinations may lead to the curve of the outbreak continuing to rise after the time that the actual outbreak ended. The final outcome in these cases is not shown.

When using the model for the assignment don't worry too much about precise measurements. These can never be known. Move the sliders by small ammounts in order to look at the combined impact of the suggested interventions. Most will affect more than one model parameter.

All the software you are using was designed and written by me for this course using the R programming language. You do not need to reference my work on this, although I hope you find it helpful. The data are from A.S. Khan, et al. 1999. The reemergence of Ebola hemorrhagic fever, Democratic Republic of the Congo, 1995. J Infect Dis 179:S76-S86.