This has been a decisive week in the life of the pandemic. The total number of Covid cases in the UK has started to fall and we may, finally, be past the worst.
But the way data is presented in the media does not make this as clear as it could be. The headline figures, no matter what data source is being quoted, almost invariably refer to the daily number of new cases. These are still rising, but for several days the pace of increase has slowed.
The Office for National Statistics does not always help matters either. For example, the ONS’ Covid bulletin on Wednesday said infections were ‘continuing to rise across the UK’, and on Friday that the percentage of people testing positive ‘continued to increase in all regions of England, except London and the East of England’.
Now, that may be true of new cases, but for policy purposes the most important statistic is the total current number of cases at any point in time. It is this total current number which we need to estimate likely hospitalisations, intensive care occupancy and deaths.
The data from the ZOE Covid study by King’s College, London has proven to be a reliable indicator of Covid cases in the UK. It is based on self-reporting through an app, and so has the great advantage of providing an earlier indication of trends than the gold standard of the ONS data. The data from ZOE and the ONS data move together very closely, but ZOE is released first.
The total number of people estimated to have Covid peaked in the data reported on Tuesday January 11 at 2.744 million. Now, just three days later, the number is 2.697 million, a fall of nearly 2%. And the fall is accelerating, around 5,000 on Wednesday, 15,000 yesterday and 25,000 today.
Even better, the reported data refer to the situation two days’ previously, so we can say with some confidence that Omicron infections peaked on Sunday January 9. The data also suggest that the lag between infection and hospitalisation averages around 8-10 days, not the 17 days cited by the UK Health Security Agency. On this basis, by the end of January the current wave of Covid will be well and truly on its way out.
Bear in mind too that the total number today is the total number yesterday, plus new cases today, minus the ‘removed’ cases today. ‘Removed” is almost entirely those who have had Covid and now no longer have it, although it also includes, unfortunately, a number of deaths every day. Put simply, although record numbers of new cases were being reported around Christmas and the New Year, the very short duration of the current variant means that more cases were becoming ‘removed’ as people recovered from the disease – a fact largely overlooked by the media.
So, the growth in the total current number of cases has already begun to slow dramatically. It seems to have peaked on Christmas Day and then begun to fall. By January 4 it was at only half this level and falling increasingly rapidly.
Understanding how the total number of cases changes over time suggests it starts slowly, accelerates sharply, and then slows down before beginning to fall. In the case of Omicron, by December 17 it was already clear that the pace of growth of the new variant had moved into the slowing down phase; with the peak of the increase not far away. You don’t need a sophisticated mathematical model to appreciate this, just a knowledge of the pattern of how case numbers change over time.
Yet, at that time, there were many still calling for tougher restrictions, even though it was clear – looking at this as economists – that at that time the UK was already close to the peak. The issue here was not the science, but the way the statistical data was being interpreted.
This confusion over what the data is saying also strengthens the case for including social scientists in the decision-making process. That argument is still relevant, given the risk of future variants or another pandemic. Involving social scientists isn’t just about weighing up the costs of lockdowns – although that is important – but about properly understanding the role of people’s behaviour.
Early in the pandemic we jointly carried out work on epidemiological models – and at one stage our ‘traffic light approach’ informed the policy debate. One lesson from the spring of 2020 was that behaviour is the key to understanding how the pandemic will evolve.
The emergence of Omicron is a reminder too of the importance of global trends and lessons. In recent weeks it was not just how the data was unfolding in the UK, but the experience of South Africa that reinforced this point. There, Omicron had risen rapidly and was already falling away by the time we had begun implementing restrictions.
South African virologists, who stated clearly that the new variant was mild. were largely ignored on these shores. Their work did, however, provide evidence of a clear pattern that we are now seeing here, with the total number of cases starting to fall ever more rapidly. That is consistent not only with the evidence from South Africa, but with the UK data as well.
In responding to the Omicron variant, the Government’s strategy of minimal restrictions has been proven correct. Based on the latest data, the Government should feel empowered to ease restrictions and return to normal. In fact, it could remove restrictions now, rather than wait until January 26. The evidence is already staring us in the face.
Paul Ormerod is a visiting professor in computer science at University College, London and Dr Gerard Lyons is chief economic strategist at Netwealth
Click here to subscribe to our daily briefing – the best pieces from CapX and across the web.
CapX depends on the generosity of its readers. If you value what we do, please consider making a donation.