COVID-19 Epidemiology, Part 5

COVID-19 Epidemiology, Part 5

(All my data are from the government site: Coronavirus (COVID-19) cases in the UK ; Last updated 14 Apr 2020, 3:44 pm GMT.)

     The news for the United Kingdom of GB continues to be mildly good. The 'doubling-time', measured in days, which is my chosen index of the spread of the virus through the population, was steady at 3.5 - 4 for the 13 days prior to the 24th March (see Fig. 1). 

     On the 24th March the government instigated what we call 'Lockdown' which closed schools, and many shops, cancelled many trains and busses, instituted restricted numbers inside food-shops, and many other curtailments. 

     It took some time for the public to get behind these restriction, but they are now very widely observed. Since the 24th of March the doubling-time of the total case-count has started to climb, and now climbs persistently. (See Fig. 1) On 14th April the doubling time was 12 days. 

Fig. 1  depicts the un-smoothed data as the increment in the total case-count, converted to 'doubling-time' in days. The horizontal axis is 'days' before and after 31st March

     Another very good measure of success in curbing the spread of this (or any other virus), discussed by Professor Trisha Greenhalgh and Jeremy Howard (1), is the transmission rate, i.e. the number of new infections generated by an infected person. If that number is greater than 1.0 the virus spreads; if less than 1.0 the epidemic dies out. COVID-19  started out with a transmission rate of 2.4. The paper by Greenhalgh and Howard (1) goes on to show how masks reduced the COVID-19 transmission rate in several countries. 

Reference:

(1. https://www.fast.ai/2020/04/13/masks-summary/)

COVID-19 Epidemiology, Part 4

COVID-19 Epidemiology, Part 4

     The good news this morning, for the UK, is that the country seems at last to be learning how not to catch the new Corona-virus. The doubling-time (in days) for 'total reported cases' has climbed steadily from the 2nd of April and is now (midnight 8th April) at 9 days. (See graph)
     The bad news (for those living in the rural county of Northamptonshire) is that Northamptonshire seems to have learnt nothing to date; no doubt living in the fool's paradise of thinking that "there are no case here". Well, there are cases in Northants.; and they are doubling in number every 41/2 days. 

Graph: Doubling-time against date in April 2020

For doubling times and other terms and calculations see previous entries in this blog.
COVID-19 Epidemiology, Part 1
COVID-19 Epidemiology, Part 2
COVID-19 Epidemiology, Part 3

COVID-19 Epidemiology, Part 3

COVID-19 Epidemiology, Part 3

I continue to plot the Log of UK cases (and deaths) each day (See COVID-19 Epidemiology, Part 1 and COVID-19 Epidemiology, Part 2.) However, the 'cases' data no longer fit a logarithmic plot; the growth is slowing. If I take the daily increase in the logged data for the period 16th - 24th March and plot that against the date (day 0 is 4th March), the data are (of course) rather wild. But the average daily rise is around 0.086 corresponding to a doubling-time of 3.5 days (0.3010/0.086 =3.5). Steady!

On 24th March lockdown in the UK was instituted in earnest. There is a very noticable change in the slope of this plot of 'Daily increment of log cases' against date. Over the 11 days 24th March - 4th April the calculated doubling-time has gradually (albeit erratically) risen and is now around 7 days. That is good news. 

No similar improvement is yet visible in the log(deaths)/day curve; doubling-time remains around 3.24 days. Presumably that is because there is a time-lapse of around a week between showing symptoms and dying. We must hope to see an improvement in the next few days regarding the data on deaths from this troublesome virus.

I understand that the 'cases' data from Mexico show an increase of around 10.5%  per day. That is to say, it is still logarithmic, with a doubling-time of close to 7 days. (0.301/Log(1.105)=6.942)