Mike Gill: Covid-19 Christmas restrictions—wrong decision, wrong time, wrong reason

The wide-ranging harms caused by unnecessarily delaying the decision on Christmas covid-19 restrictions are being felt by the whole country, says Mike Gill

On 19 December, Boris Johnson, the UK’s prime minister radically changed the covid-19 restrictions and Christmas arrangements for England. In my view he took the wrong decision, at the wrong time, and for the wrong reason. 

He introduced a new, tougher Tier 4 for London and much of the south east of England, with restrictions similar to the last lockdown. The Christmas relaxation of restrictions for those in other Tiers has been reduced to only Christmas Day itself. The reason, he said, was that on 18 December, he learned that the virus was spreading more rapidly in London, the south east and the east of England than would be expected given the tough restrictions which were already in place. [1] He also expressed concerns that the spread appeared to be driven by the possibly more transmissible strain of the virus, mentioned by Matt Hancock in his House of Commons statement on 14 December, when he also put London and adjacent areas to the north into Tier 3 to mitigate that risk. The justification for taking further action four days after announcing a Tier change was that there was now “proof” that the new variant was driving the spread of the virus. [2]

In truth, the situation in the south east had been concerning since before the end of lockdown on 4 December. Elsewhere in England, lockdown had achieved significant falls in new cases and in test positivity rates, but by 25 November, London, south east and eastern regions were trending in the wrong direction. Initially, this was especially in north Kent, but soon across a larger area, covering east and north east London, and adjacent parts of Essex and Hertfordshire. The increase in prevalence was confirmed in these areas.

Hospital data are hard to interpret: “patients admitted” for example is a simple count of people that are admitted who had a positive covid test within the preceding 14 days, or who tested positive on admission, irrespective of their reason for admission. This will increase proportional to the amount of testing being done, and this has increased significantly in recent weeks in the relevant regions. The numbers on ventilators (actually strictly those in beds which are capable of delivering mechanical ventilation) are arguably more robust indicators of viral prevalence three weeks or so earlier. Between 4 and 15 December in the three regions, South East, East, and London, the 7-day averages all increased by 32% (110 to 145), 25% (84 to 105), and 13% (249 to 281) respectively. In the East Kent Hospitals NHS Trust alone they increased by 53% (17 to 26). Put together these data confirmed a steadily growing problem, apparent well before 18 December. 

In other words, it was already clear that releasing the south east from lockdown was going to be risky. Close attention should have quickly revealed that Tier 3 was insufficient to gain control in Kent, and Tier 2 insufficient for London and the adjacent areas to the north and east, hence the change to tier 3 for London on 16 December. Kent’s status was unaccountably left as it was despite “very sharp exponential rises in the previous week” and its hospitals already being “under pressure.” [3] 

What apparently forced the prime minister’s hand was the confirmation from Public Health England (PHE) that much of the rise in transmission rates was driven by the new variant (VUI-202012/01). This is hard to understand. Whether or not the variant was responsible should have made no difference to the need for the decision to take public health action that had been obvious for well over a week. To have delayed such action, pending arrival from PHE of this “proof,” is culpable. How is it to be explained that Boris Johnson appears to have required this confirmation, and presented the re-imposition of restrictions as a necessary and appropriate response to the news? 

There are three sorts of explanation. The first explanation is based on the prime minister’s enduring record of delaying taking difficult decisions as long as possible. The timing of the first lockdown is possibly the most egregious example. The second explanation is political: by delaying the decision until parliament had started its Christmas break prevented scrutiny of a decision likely to be unpopular in certain sections of his own party. [4] The third explanation is a reflection of the assumptions many of us hold about the different scientific approaches to proof of causation. The fact that the UK has a “world-class genomic capability” does not in itself make the “proved” implication of the variant as a cause of transmission any more important to the process of public health decision making than the epidemiological data that have been telling the story of growing viral prevalence for some time. [3] Laboratory-based science, especially modern and “world class” somehow carries more weight when it determines that “x” is a cause of “y”, which in this case seems to have taken over our focus, rather than the epidemiological approach, which has made plain what should be done since soon after lockdown was lifted.

After all, NERVTAG (New and Emerging Respiratory Virus Threats Advisory Group) had expressed “moderate confidence that VUI-202012/01 demonstrates a substantial increase in transmissibility compared to other variants.” [5] This view is consistent with the new strain being coincident with, rather than a cause of, the rise in SARS-CoV-2 prevalence in a limited number of areas. Prevalence was falling in other areas, including some where the new strain was present. It is of note that even in mid November, this variant seems to have had the same level of representation in three different regions. If it had a rapid spread advantage, we might expect it to have been ahead in one region, and spread to others later. [6] 

This is a good example of misplaced faith in reductionist science, in this case mistakenly used to justify late, extreme, and destructive action. John Snow would have had to wait 30 years to use the identification of vibrio cholerae as the justification for removal of the Broad street pump. In 1985, within 24 hours of completion of a case control study of families affected by a rare salmonella, the formula milk factory identified as the source of contaminated milk powder was closed without any microbiological evidence. This only appeared six months later when a hairline crack was found in the factory’s spray drier. [7] Both these examples illustrate the value of pre-emptive, but proportionate action, of the sort needed to optimally tackle this pandemic. Both relied on the careful interpretation of epidemiological data. There have already been too many preventable cases and deaths. The wide-ranging harms caused by unnecessarily delaying the decision on Christmas covid-19 restrictions are being felt by the whole country. 

Mike Gill, former Regional Director of Public Health, South East England.

Competing interests: None declared.


  1. Prime Minister’s statement https://www.gov.uk/government/speeches/prime-ministers-statement-on-coronavirus-covid-19-19-d ecember-2020 (accessed 20/11/20) 
  2. Matt Hancock on Andrew Marr show 20th December
  3. Oral statement to parliament https://www.gov.uk/government/speeches/sharp-rise-in-coronavirus-numbers-and-a-new-variant (accessed 20/12/20) 
  4. Sir Charles Walker BBC The World this Weekend 20/12/20 
  5. https://khub.net/documents/135939561/338928724/SARS-CoV-2+variant+under+investigation%2C+ meeting+minutes.pdf (accessed 21/12/20) 
  6. https://www.gov.uk/government/publications/slides-to-accompany-coronavirus-press-conference-19 -december-2020 (accessed 21/12/20) 
  7. Rowe B et al Salmonella Ealing infections associated with consumption of infant dried milk https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(87)91384-5/fulltext (accessed 20/12/20)