Week 10 - Where do we go from here? Should we be just a little flexible?

At present, we seem to be at a cross-roads. Either governments should have a single policy for their entire country,  or they might be some advantages in having a flexible policy, treating some parts of a country differently from others.

 

Have conditions been satisfied to begin to release the lockdown?

 

Most European countries (UK, Italy, Spain and France) are nervously emerging from lockdown after experiencing very high levels of Covid-19 cases. The reasons that politicians are keen to relax the lockdown as soon as possible are that the economic costs of the lockdown are huge, and we are likely to move into a recession.  They may possibly be hoping, as ours seems to be, that just bringing their R number below 1 will somehow, magically, make the virus disappear. But in countries that have experienced such high prevalence, if a vaccine proves impossible, and if anti-viral treatments are only partially effective, it seems more likely that there will always be a residuum of active cases hiding in the community. An editorial in Nature (6^th May 2020) considers that conditions have not yet been met:  the rate for new sporadic cases is too high, one should minimize transmission in hospitals, and contact tracing should be underway. 

 

To recapitulate, a complete lockdown allows a government to buy time, and lower the R number. But this is all it will do – to achieve elimination of the virus, it must be accompanied by control of our borders, by repeated testing in order to identify positive cases in both arrivals, and in patients and staff of care homes and hospitals. (This is the exact reverse of what the government actually did. 

Those carrying the virus were not even screened on arrival). The UK government is talking about case identification and contact identification, but it appears to be being introduced in a careless and ambivalent manner (BBC news, this week).

 

What is our R at present, 19^th May?

 

There is some dispute about this. A paper by a distinguished group near Bristol (R. Challen, K. Tsaneva-Atanasova et al, April 2020) have calculated values for R in various UK locations, including all 4 countries in the UK. They find R consistently above 1, Within England currently the R is lowest in the Midlands at 1.11 (95% CI 1.07; 1.14) , and highest in the North East of England at 1.38 (95% CI 1.33-1.42). They estimated that by 10 April R was above unity in England, but had dropped to unity in the other three countries.  That was then, what about now?

 

On Friday 15^th May the ‘Government Office for Science’ (GOS) published the latest R number range for the UK, and gave a single value of  ‘0.7 to 1.00’ for the entire country. There is no necessary inconsistency – there is variation between estimates, and just over a month has passed between the two estimates.  Public Health England replied to my request for their latest figures which were dated 10^th May 2020, and arrived with confidence limits for each estimate of R.

 

 

                                                                       95% CI

            Region           Median           Lower             Upper

1.          ‘England’       0.75                0.72                0.77

2..        East England  0.71                0.68                0.74

3.         London           0.40                0.36                0.43

4.         Midlands        0.68                0.65                0.71

5.         NE & Yorksh.  0.80                0.76                0.83

6.         North West    0.73                0.70                0.76

7.         South East      0.71                0.68                0.74

8.         South West    0.76                0.72                0.80

 

The new numbers now given by the GOS are higher and by no means the same as those from  PHE,  but it is reassuring to see that the new 95% CIs seem reasonable.  The GOS figures are produced by a group that has senior politicians on it. One should at least consider the possibility that things are not quite as they are claimed to be.

 

Might a different policy have been approved?

 

The first thing to say is that new policies should not be introduced in places where it might be dangerous to do so. At present there are rates much too high in at least three areas: rates/100K of new cases are 43.5 in Gateshead, 41.6 in Herefordshire, and 41.3 in South Tyneside. At the other end of the scale, they’re 2.6 in Bath & E.Somerset, 2.3 in Bedford, and 0.7 in Torbay. At a time when things at the high end are calming down, it would be absurd to release the lockdown in the high incidence places, but non-controversial to do it in the quiet areas.

 

How do we compare the UK with the USA, if we use comparable measures of risk?

 

These are rates / million at risk, using the Financial Times’ new gadget. In covid cases, there is a definite advantage to the UK, but in deaths we are in a more precarious position than the USA. But these are early days for the USA!

 

                                    Covid Cases              Covid deaths

UK                               50/M                          6.0/M

USA                             75/M                          4.5/M

 

Expressing the morbidity in this way should surely wipe the grin off any British face. But poor Mr Trump and his partner-in-crime Mr Bolsonaro  must be hoping that their muddled and somewhat incoherent policies might somehow win through, and he will achieve a worthwhile economic advantage with an increased death rate that will be somehow acceptable to his voters. I think things can only get still worse for both of them, and well before the election!

 

David Goldberg

21 May 2020

70 days served (endless lockdown for old people stretches ahead).