epiforecasts / ringbp

Simulate infectious disease transmission with contact tracing
http://epiforecasts.io/ringbp/
Other
15 stars 10 forks source link

Alter R0 of initial cases #10

Closed jhellewell14 closed 4 years ago

jhellewell14 commented 4 years ago

The R0 of initial "within cluster" cases is set to R0within which is, by definition, low. This means that the initial 40 cases only cause very few offspring which means there is a large dip in weekly incidence before the outbreak gets going again. Really these cases should have the missed cases R0 when deciding their secondary cases because we are assuming that the transmission has already happened unobserved in the community.

seabbs commented 4 years ago

Can you initialise the model with all of the first 40 cases being in their own cluster (and therefore being undetected)?

jhellewell14 commented 4 years ago

image

This is what I get if I run 40 clusters with 1 case in each, is that dip now due to incubation period of offspring from initial 40 cases?