EU Referendum


Coronavirus: a fog of ignorance


12/04/2020




Something of a glutton for punishment, I watched the BBC TV evening news again yesterday, and there was no doubt about it. The bulletin clearly stated that 917 people had died from Covid-19 in the previous 24 hours.

Yet, this is manifestly not true. It is not even close to true. When we refer to the NHS site, and specifically the spreadsheet breaking down the 823 people who died in English hospitals, we see that only 115 occurred in the previous 24 hours.

A further 325 are recorded on 9 April – the day before that – and 161 are allocated to 8 April, while another 209 are split between 1-7 April. The remaining 12 deaths occurred in March, with the earliest new death taking place on 25 March.

There is no secret about this – the NHS is openly publishing the daily breakdowns, starting from 2 April. This means that there is now a far better opportunity to build a picture of Covid-19 mortality, albeit that the data are still fragile, for the reasons pointed out yesterday.

What is inexplicable, therefore, is the willingness of ministers at the daily No 10 pressers, to allow the interpretation that we are being offered clean figures for each of the preceding days. That this is then compounded by the media adds further to the mystification, as the data continue to be misrepresented.

Bringing it almost to the level of high farce, we then have the BBC's "Head of statistics", Robert Cuffe, solemnly offer us an analysis, telling us that "the growth in the total number of new deaths has stalled in the last four days".

Cuffe then goes on to tell us that in other some other countries that implemented lockdown, the numbers of reported deaths stopped growing about three weeks into lockdown, but his only concession to reality is his statement that it is too soon to know for sure whether we have reached that point.

"There have", he says, "been reporting lags at weekends and it is possible that a bank holiday weekend will include deaths that go unreported until next week" – an almost comic understatement. It is a matter of certainty, borne out by the NHS daily breakdowns, that it is taking well over two weeks to pull in the figures for any one day.

Effectively, therefore, we won't know the true picture for the current daily total until the end of the month, and only from then on will we be able to ascertain the hospital position. And that, of course, leaves out deaths which are recorded outside hospitals, in people's homes, in care homes and the like.

No one is suggesting that there is any conspiracy in the way that the official daily figures are being presented but, as I observed yesterday, it would be far better if the ministers stressed that the figures were provisional, with updates issued as a matter of routine.

Of course, it is extremely frustrating that we can't have instant figures but it must be recognised that it takes time for deaths to be recorded officially on medical certificates, for testing to be confirmed, for post-mortem reports to be processed, and data from any further tests to be validated.

It can't be emphasised enough though that accurate and timely information is the key to managing an epidemic. The case and death rates provide important information, and the former can be a relatively sensitive indicator of the success of control measures taken.

However, because so many cases go untested and the number of tests recorded changes through the epidemic (which in turn influences the number of positives), the data can be extremely unreliable. On the other hand, the death rate – although a trailing indicator – is regarded as more reliable.

From looking at the spreadsheets, it seems to take well over a week for the bulk of death reports for any one day to trickle into the system, which means that there is probably a lag of about a week before there can be any confidence that changes in incidence can be detected.

Ostensibly, this information is needed to guide ministers in whether to lift lockdown restrictions, but this is not all they will need. To get some idea of what might happen, it would be a good idea to have some indication of how many people had been exposed to infection, area by area, together with some further indications of whether they have any acquired immunity.

This will enable estimates to be made as to whether rapid spread is likely once restrictions are lifted. But, before relaxation is even considered, it would be wise to have in place a community test and trace system, so that the inevitable recrudescence can be quickly detected, allowing timely controls to be re-imposed.

Given the fragility of the data, however, and the absence of any effective community monitoring, ministers are flying blind. Not only will it be difficult to judge when to lift restrictions, it will also be difficult to judge the effects of the action taken, with the risk of new hotspots breaking out and getting out of control.

Clearly, though, something has to give. The lockdown, patchy at best, is already starting to fray. In my own little corner, over the weekend, I observed a boozy street party in one street and, in the next, a house party in a front garden. People are getting bored, they need to socialise and no amounts of threats of police intimidation is going to damp down that need.

For all that there are some encouraging articles about a vaccine being ready by September, even if that turns out to be true, there are substantial problems in manufacturing sufficient quantities to sustain a national vaccination programme.

Then, given that the R0 might be higher than some have forecast, with the US CDC recently suggesting that it could be as high as 5.7, that means we will have to achieve a herd immunity in the region of 85 percent in order to suppress any further infection.

Unknown at this time is the longevity of any conferred immunity, and whether booster shots will be needed to complete a programme. If the period is relatively short-term, it may mean that stocks of vaccine will have to be built up before the programme can start, with resources mobilised to enable vaccination to be completed to the desired level, before herd immunity begins to wane.

Even at best, we are still looking at about a year before any vaccination can take effect and 18 months to two years is still looking more realistic. With our ability to monitor the outbreak reliant on fragile data, and with the further lack of any substantial community testing and contact tracing, ministers are running out of options.

There are some suggestions that "big data" could come to the rescue, with mobile phone apps being developed that would enable an infected person's contacts to be traced electronically. In the system as planned, contacts could be alerted and directed to a testing centre.

But even if that worked and could be developed in time, it would require personnel to administer and police the system, especially as there would be a default rate, where contacts did not report for testing. The necessary people are not in place. Recruitment, training and their organisation could hardly be a speedy process.

This, therefore, has to smell of one of those panacea schemes, dragged out of nowhere as a quick fix – only for the complications then to slow the whole thing down to the extent that it never fulfils its original promise.

It still doesn't seem to have dawned on either government or those proposing their grand solutions that this epidemic is comprised of a series of outbreaks. Many of these are small-scale and, at a local level, can be managed. Still there is the tendency to treat it as a single event, with ministers attempting to micromanage from their London offices.

That much is evident from the growing controversy over the provision of PPE in NHS hospitals. If purchasing was decentralised, and individual hospitals were able to source (and pay for) their own supplies, the problem would vanish in a heartbeat. Instead, we have a feline Priti Patel telling us she's "sorry if people feel that there have been failings", as hospital and social care staff go short.

Eventually, this government will have to rediscover the value of localism, using the untapped skills and resources of local communities. Until then, we will be seeing ministers flounder in their fog of ignorance, while the epidemic continues and the economy turns from recession to depression.