search

thibautjombart / covid19_bed_occupancy

Shiny app providing estimates of future bed occupancy given recent admissions
https://cmmid-lshtm.shinyapps.io/hospital_bed_occupancy_projections/
Other
12 stars 5 forks source link

Move to R0 and serial interval specification #74

Open samclifford opened 4 years ago

samclifford commented 4 years ago

Can ask user to specify either R0 and serial interval or doubling/halving time for exponential growth/decay. This gives us a way to look at steady state occupancy.

samclifford commented 4 years ago

Park et al (2020) have done a systematic review containing serial interval and R0. They cite Du et al. (2020) who studied 468 pairs and report a mean of 3.96 (SD 4.75) days, with negative serial intervals indicating presymptomatic transmission. So perhaps in this case a normal distribution for the serial interval is most appropriate.

samclifford commented 4 years ago

Implemented in d1345a02158baecf88b17f9361b6391fe4ac2ef7

samclifford commented 4 years ago

Above implementation not correct. Need to cannibalise the r_doubling code to pull out the generation of R0 and Tc values and then put a selector to determine whether to use doubling time or a branching process. Will also require a selector for the dispersion parameter of the negative binomial distribution. This should probably only take on published values for the time being: 0.1 (Endo et al.) and 0.54 (Riou and Althaus).

samclifford commented 4 years ago

Definitely done at least as late as b0e2b1bb3193493ca0ad1600266c21439bfabd5b