‘Our estimate which is that the risk of death increases by 30 per cent is itself uncertain. We think it could be anywhere between 10 per cent and 50 per cent according to our analyses,’ said Dr Nick Davies, the author of one of the studies referenced in Friday’s Downing Street press conference, during an interview on Radio 4.
Between listening to the press conference last night and Dr Nick Davies, you could be forgiven for thinking that this latest development is all rather tentative and possibly merely a quirk of statistics that is - as yet - no cause for alarm. The controversial way the figures were presented; briefed to The Sun first, then announced at a press conference with the actual study released only later - may give the impression of a government spin operation design.
This, however, might be wishful thinking. There is plenty of doubt, to be sure, but it’s also possible that we have just heard an understatement of what could potentially be a dreadful situation. But does that mean it is necessarily a clear situation? No, far from it.
As we have seen many times during the course of the pandemic, it is a fluid - and quickly moving situation, one filled with lots of noisy data.
Why the uncertainty from Sir Patrick Vallance?
Sir Patrick Vallance, did not report the claim about a higher case fatality rate (CFR) as irrefutable fact. Instead, the chief scientific adviser emphasised that it was the opinion of Nervtag, the government committee which monitors risks associated with respiratory tract viruses. ‘The evidence is not yet strong,’ he said. Sir Patrick went on to explain that it had been pieced together from a ‘series of different bits of information’ and that we do not ‘have a very good estimate of its precise nature’.
So why the cautious tone? One explanation could be that he wanted to give himself some cover in case the studies don’t bear their findings out later when they are more developed. It is also possible that Sir Patrick wanted to create some distance between himself and Nervtag, for this body has come under a lot of fire throughout this crisis. A common refrain has been that Nervtag is too pessimistic and overly keen on lockdowns.
How did Nervtag come to the conclusion that the virus could be more deadly?
The Nervtag review was a summary of four studies (below). Some used the same data set, to compare the risk of death from the UK ‘variant of concern’ (the Kent strain) to that of other Covid-19 variants.
Their best assessments of what they think is truly happening is that there has been an increased risk of dying of between 35 and 91 per cent. The central estimates of these studies - which are not published, but are referenced in the Nervtag report - are broken down as follows:
- London School of Hygiene and Tropical Medicine - 35 per cent
- Imperial College - 36 per cent
- PHE - 65 per cent (it is worth noting that in an earlier study PHE did not believe it was more lethal.)
- University of Exeter - 91 per cent
This is a huge range, of course. So what explains it? For starters, the different data sets: Imperial, the London School of Hygiene and Tropical Medicine and Exeter used one set; PHE used another one. This would suggest, therefore, that something is clearly most likely going on, and that it is not simply a case of noisy and unreliable data.
Why did Nervtag go for the lower end of the range?
Having gathered these four studies, Nervtag concluded that, on balance, the Kent strain could be 30 to 40 per cent more lethal.
How have they arrived at this figure when the four studies said 35 - 91 per cent? It is not clear in the Nervtag paper, but it would make sense that they selected such a figure on the basis of likelihood, as it is at the conservative end of the range.
But, how likely do they think it is that the virus is actually that much more lethal? This is where it gets somewhat more complicated. For Nervtag state in their paper on the four studies that it is a ‘realistic possibility’.
What do they mean by this? Essentially, this term is civil service speak for there being a 40 to 50 per cent chance of it actually being the case. (This is to do with how they express probabilities in government, without resorting to numbers.)
It is, in other words, a civil service method of expressing likelihood: they judge the current odds of the CFR effect being real as no more than evens. Or put another way: Nervtag have made it clear that they think there is a 50/50 chance the new Kent strain is 30 to 40 per cent more deadly.
What are the limitations of the studies?
As with any study like this, the data used is only partial: the sample size was only 8 per cent of total Covid-19 deaths, which is not as bad as it sounds, but could be much better. To really draw a sound conclusion, in science, we would ideally want significantly larger numbers than these (there is no magic number here, but it would not be unreasonable to expect to have at least 20 per cent to start to draw significant conclusions). Members of Nervtag readily concede that other data sets need to be examined.
What can we draw from yesterday’s conference?
When deciding whether to go public with these still quite uncertain observations, the government had a really hard task. Go in too hard and they would be accused of scaremongering. If they play it too cool, then they would be accused of hiding things from the public and causing extra deaths. Sir Patrick chose to put it in terms of 60 year old men and the rise of a CFR from 1 per cent to 1.3 or 1.4 per cent with the Kent variant. ‘Out of 1,000 people aged in their 60s you would expect 10 patients with Covid-19 to die,’ he said. ‘With the new variant 13 or 14 would be expected to die.’ This might sound like a small change in absolute terms. But when scaled up, and the impact on older age groups who suffer a higher CFR already is factored in, then it is significant. Indeed, Nervtag states explicitly that the risk ‘appears to be apparent across age groups’. The fact that the UK is currently experiencing a death rate of around 1,100 per day (on a seven day average) means that this increased risk can account for hundreds of extra deaths every day. It is not trivial.
It is clear that there is still a lot more work to be done in order to properly verify the severity of the situation. The observations made in the four studies and the subsequent Nervtag report, however, do probably represent a kernel of truth. If we know one thing from the last year, it is that the virus is unpredictable and has side-stepped us at virtually every turn, so there is no reason to doubt that the Covid-19 landscape could yet change again.
Of course it might be some time before the more complex picture than the one presented yesterday is fully revealed.
However, it is clear in the meantime that Nervtag have not rowed back on the belief that the Kent variant is more transmissible (the original 70 per cent figure has been scaled down to a range of 30 to 50 per cent - but this is still very significant). Boris Johnson, in fact, stated this explicitly at the press conference, meaning a bigger burden on public health will continue if the virus is allowed to spread unchecked. This, unfortunately, poses even more serious questions over how we open up society again.