search

charlesmilk / covid19

Forecast of the developments of the COVID19 in Portugal
13 stars 4 forks source link

New variable: Lockdown dates #1

Closed isabelportugal closed 4 years ago

isabelportugal commented 4 years ago

A city or a country lockdown should influence the rate at which the virus is transmitted, therefore changing the distribution. I didn't find a clean dataset with lockdown date for each {city, country}, but it would be interesting to explore the growth diff when a country is on lockdown (? growth) vs. when it isn't (exponential growth). Italy is already on lockdown nation-wise, so you probably will start to notice a difference in Italy's growth rate. By adding this variable, it would be possible to quantify the importance of lockdown as a preventive measure in highly infectious diseases.

charlesmilk commented 4 years ago

That is spot on. Actually in the analysis provided I commented out that we could definitely stop the propagation. If we look at Macau or China today for instance. I will do something to show the difference in growth when the aforementioned countries went to lockdown. This will be a very good way to spread the word of how important is the quarantine. Thank you for the feedback @isabelportugal :)

isabelportugal commented 4 years ago

Keep up the excellent work @charlesmilk :)

I haven't seen many studies that try to predict the development of the European COVID-19 case. Even if you intend to focus on how Portugal is being/will be affected by COVID-19, your work will help people understand what to expect during this outbreak -- which is an excellent initiative.

Fig.1 shows you the hypothesis that if we prevent a massive spike of active cases, our health care system will be capable of mitigating this infection much faster. This spike represents a sudden explosion of illness that requires more people to be hospitalized than what our health care system can handle. Hence why I expect to see a slowdown in the growth rate of the confirmed cases once countries implement preventive measures.

Another interesting variable for this would be # active cases (or under treatment) = #confirmed cases - #recovered - #deaths . By adding this second variable you would have the demand for medical assistance while being able to also analyze the slowdown and mitigation of the infection over time (Fig.2 - yellow bars).

c84de7f7-8eb3-462a-bc0f-f2ea14768dcf (Fig.1) https://www.vox.com/2020/3/10/21171481/coronavirus-us-cases-quarantine-cancellation

image (Fig.2) Original data provided by China’s National Health Commission (http://www.nhc.gov.cn/xcs/yqtb/list_gzbd.shtml)