Wednesday, March 11, 2020

Modeling the Coronavirus Contagion



Without a vaccine, the protection to at-risk populations is a matter of not getting the disease in the first place. So, the question of the pervasiveness of the spread of the disease is of greatest importance. The answer to this question depends almost entirely on the immunity of those infected and recovered since the US is not able to institute draconian isolation measures over any large geographical area.


While there was an initial indication that recovered individuals may be contagious, this is likely a test failure. Let us therefore assume that recovered individuals have immunity from reinfection and are not contagious for at least six months. If so, then the protection of at-risk individuals is feasible if they can be locally isolated when the wave of infections sweeps through an area. The recent success of the Chinese to stem the rate of spread of the disease should give us hope even though we do not weld doors shut on apartment complexes.


Remember, there is an approximately two-week lag time. So, when you see a bump up in the local rate of infection, act quickly and decisively to isolate your at-risk populations. Be prepared for 6-8 weeks of isolation per wave of infection. Meanwhile, on the national level, the best policy is to isolate the nation and quickly saturate with testing and isolate any affected occurrences. This disease is not super-human, but it is no slacker either.


A more detailed nonlinear control model will be forthcoming from the Prof. Ellina Grigorieva when data is available to estimate the parameters.



Good Luck.