December 4, 2020

expert reaction to latest ONS COVID-19 Infection Survey data

The Office for National Statistics (ONS) have released the latest data from their COVID-19 Infection Survey.

Dr Simon Clarke, Associate Professor in Cellular Microbiology, University of Reading, said:

“The latest data from the ONS show that while the effects of lockdown 2.0 has been to reduce infection rates, we are a long way from stamping out the disease quite yet.

“Those who argued against the stricter post-lockdown tiers, including many politicians, should look hard at this data and realise what it means. It means that, at the end of November, there were still half a million people in the UK with coronavirus. It meant that 1 in 50 secondary school children was carrying the disease. And it showed that some parts of the country still had stubbornly high infection and death rates.

“Scientists like me don’t want to come across as prophets of doom. But people should remember that while there is good news about vaccines and infection rates reducing, hundreds of people are still dying daily, and the virus is far from reaching the lower levels it reached as a result of the first lockdown.

“A contributory factor behind this is likely to be because schools and universities have remained open; this is reflected in the high infection rates among secondary school-aged children in particular. The government and Universities have taken good precautionary measures, in the form of mass testing and the ‘travel window’ for students, to reduce the risk of infections being taken home for Christmas.

“In this environment, people should think carefully about their plans for Christmas. Creating ‘Christmas bubbles’ containing three households may be what the law allows you to do, but I would be thinking very carefully about mixing children with their grandparents in particular this Christmas, especially considering how many under-16s are still silently carrying the virus.”

Prof James Naismith FRS FMedSci, Director of the Rosalind Franklin Institute, and University of Oxford, said:

“I think of the thousands of deaths from Covid 19 over the past 10 days and the heartbreak this has brought to so many. Each life cut short, no matter how old, represents a real loss.

“The ONS survey data up to 28^th November are in line with all other data, the lockdown has decreased the number of new infections (26, 000, 95 % CI 22, 000 to 29,000). The spread of the virus is slowing.

“It is clear that on a national level, the lockdown has had the predicted effect. I note that the number of positives detected in the testing programme was over 60% of that estimated by the ONS survey. This is encouraging, suggesting the testing program is once again better tracking the virus; we would expect this as the case numbers drop.  I and others became alarmed when the daily number of positive cases reported was less than half the ONS estimates (September and part of October). This is because the daily headline of positive cases drives the news agenda and often the political discussion, risking decisions based on flawed understanding. Reputable scientists predicted in late October that the death toll would hit 500 for several days in mid to late November. There is nothing magical about this: just divide the number of ONS infections by 250 for a low end estimate or multiply by 2 for a high end estimate of the number of deaths announced per day in three weeks’ time.  I hope by drawing attention to this, it helps the public understand better what has happened, why scientists were alarmed in September / October and how to make sense of the future data releases.

“We are currently on the downward slope of the second wave. There can be no celebration, too many have died; but we have managed the second wave somewhat better than the first.  The lower we get the daily number of infections, the less risk the Christmas relaxation poses.

“It is important for us to think about what will happen to the numbers over Christmas. 

“We do not need a crystal ball, the US Thanksgiving holiday provides a guide. The number of cases and deaths will drop very sharply for the days of the holidays themselves. As people return to work, cases and deaths will spike as the system catches up. Neither the dip nor the spike tell us anything. The Christmas disruption in the numbers will take around 1 week to work through. The UK new year holiday will again distort the numbers. It seems very likely that we not be able to make sense of the daily results until early (possibly mid)  January, We should be prepared for this. Hospital admissions will be a reliable guide to what is happening, but they won’t begin to tell us about any Christmas induced rise in infection until around two weeks later, (around 8 to 12th January). 

“By the middle to end of next week, US data will tell us what happened to infections over thanksgiving. The US went into this holiday with a high number of cases. 

“In a democracy, politicians are elected to take decisions affecting our lives. With so much at stake, lives, livelihoods and quality of life, our politicians will need the most reliable data. It would be good if we have ended the period of magical thinking about the virus. The virus can spread quickly when social restrictions are relaxed. Testing data can miss sudden increases and survey data report much later, thus the virus can gain a foothold before we realise it.  When cases rise, two weeks later hospitals admissions rise, a week or so later deaths rise. Thus, waiting to be sure is deciding to risk the virus infecting thousands a day and filling the hospitals. If we fill hospitals (not just beds but staff), the death toll will sharply increase. We can decide this is a risk we are prepared to take, but what should not be allowed is to pretend the risk does not exist.

“I am certain the roll of the Pfizer vaccine and new vaccines early next year will bring an end to the pandemic in the UK.”

All our previous output on this subject can be seen at this weblink:

www.sciencemediacentre.org/tag/covid-19

Declared interests

None received.