Summary.
This is the third and final post that will look at the impact of the ‘Stay at Home’ lockdown measures implemented in the UK during the period October 2021 to May 2022. A period where the UK home nations saw two separate period of lockdowns.
The two previous posts can be found here .
The aim of these posts is twofold. First does the data tell us anything about the impact these lockdowns had on deaths in the period after they were implemented. And secondly, can we learn anything about the importance of the timing and duration of lockdowns
This article looks at the impact of the Winter lockdown implemented on January 6, 2021 in the English regions. It tells a story of how regional prior to lockdown changed it’s impact on deaths. More importantly, it also shows that earlier lockdowns saved lives.
Winter is Coming.
Let’s start our story by looking at the big picture and the following graph plots the number of deaths (per 100,000) for England for the period after the end of the Autumn lockdown until March 2021. The grey shading shows when the Winter lockdown was in place together with key dates when restrictions were changed. The two blue diamonds show the range of dates after the start of lockdown when we would expect to see a decrease in the number of deaths. Appendix 1 gives a full technical explanation for choosing this range of delay.
The chart shows that the death rate for England was high when the Autumn lock down ended on Dec 3. The death rate continued to rise and did not peak until Jan 19 2021 just 13 days after Winter lockdown was announced. This is two days ahead of the earliest day we would expect to see a decrease based on the lead-time between infections and death. Critics of lockdowns have argued that this is evidence that infections were falling before lockdowns were implemented and, by inference, they were not needed.
As we will see shortly this is not the case, but first we need to look at actions taken ahead of the Winter lockdown to control the level on Covid infections.
Autumn Control Measures.
After the Autumn lockdown the three tier system had been re-introduced. The aim of these tiers was to progressively reduce the level of physical contacts by limiting the opportunities to mix socially. Tier 1 had the lowest level of restrictions and tier three had the highest although these were well short of a lockdown. The Tier levels were implemented at the local rather than the national level.
In response to the rapid rise in infections the government were increasing the number of local areas placed in to the higher tiers but to little effect, and on Dec 20 implemented a more restrictive Tier 4 ‘Stay at Home’ level for many areas. The following series of maps show how quickly the tier levels changed from Dec 2, 2020 to Jan 5, 2021. The change was rapid and by Dec 31 the majority of areas were in the most stringent Tier 4.
Clearly these local restrictions will have a different effect on the local outcomes of the Winter lockdown so let’s now look at the regional trends for deaths across England starting this time with the three southern most regions.
The Southern Regions.
The southern regions are London, the South East and East of England. The following chart plots the number of daily deaths per 100,000 people over time for each region and shows when they entered Tier 4 ‘Stay at Home’ restrictions. The two blue diamonds show the range of dates after the start of lockdown when we would expect to see a decrease in the number of deaths.
All three regions peak ahead of the date when we would expect to see an impact (the blue diamond at 15 days). But does this mean that the lockdown was not needed? Clearly this is not the case as many of the regions in the south entered the ‘Stay at Home’ Tier 4 ahead of the Winter lockdown.
All of the London region entered Tier 4 on Dec 20 together with most areas in the other southern regions. The remaining areas in the southern region went fully into Tier 4 on Dec 26.
As all of the London region entered Tier 4 first it would be expected that their death rate would peak first and this is the case. The death rate peaked on Jan 16 which was 26 days after entering Tier 4 and is slightly later than the expected lag between infections and death. The South East (23 days) and East of England (25 Days) were also slightly later than expected. Clearly using the start of full lockdown as a date from which to start the clock cannot be used to to assess the impact these regions.
Middle England Regions.
The ‘Middle’ England regions are the West Midlands, East Midlands and South West. Again, the following chart charts the number of daily deaths per 100,000 people over time for each region and shows when they entered Tier 4 ‘Stay at Home’ restrictions on Dec 31 – New Years Eve.
It’s difficult to assess the impact of the Tier 4 restriction implemented just 6 days ahead of lockdown and in the middle of the holiday season. Some may have moderated their behaviour and some may have taken the opportunity to meet ahead of the anticipated lockdown.
Visually, you can see that death rates for West Midlands peaked on Jan 21 within the expected range and just 21 days after entering Tier 4. The South West, which had less areas in Tier 4, peaked on Jan 23 which was 23 days after entering Tier 4. However, East Midlands peaked slightly earlier than would be expected on Jan 18, just 18 days after Tier 4 restrictions were introduced in some local areas.
Earlier Lockdowns Save Lives.
Importantly, comparing death rates for the ‘Middle’ England regions with those for the Southern regions does tell us something significant about the effectiveness of the Winter lockdown.
We know that the Alpha variant wave started in the south of England. Consequently, on the day the national Winter lockdown started the progress of the Alpha wave in the southern regions was well ahead of the ‘Middle’ England regions. This can be clearly seen by comparing the death rates on the day lockdown was announced. On Jan 6 2021, southern regions had a daily death rate of about 2 deaths per 100,000 people, considerably higher than the ‘Middle’ England regions rates which ranged between 0.8 to 1.3 deaths per 100,000. This means that, in effect, the lockdown measures were implemented later in the south than for the other regions.
Comparing the two charts shows that ‘Middle’ England regions had lower peaks for their death rates, with less deaths, than the southern regions who entered their lockdowns later. This is strong evidence that earlier lockdowns saves lives.
The Northern Regions.
What about the evidence from the Northern regions? Does this support the claim that earlier lockdowns save lives?
The following chart shows the death rate for the the three regions in the north of England, comprising the North West, North East and Yorkshire and the Humber. All three regions had a small number of areas in Tier 4 ahead of the Winter lockdown.
Here we see that all three northern regions had a much lower death rate than all other regions, with the exception of the South West. What these regions have in common is that they entered the lockdown at lower death rates meaning that the lockdown was implemented earlier in the wave than was the case in the southern regions.
The evidence from the northern regions strongly support the claim that early lockdown saves lives.
In conclusion
I hope that I have shown that drawing conclusions on the merits or otherwise of lockdowns from simplistic analysis of charts can mislead. Only by fully exploring the context and understanding the data at a more granular (regional) level can you confidently determine the impacts of lockdowns.
Appendix 1. Technical note on the time lag between infection and death
The time between a COVID infection and death can vary depending on several factors such as the severity of the infection, age, and underlying health conditions of the infected person.
A review from the Office for National Statistics, based on early studies using data from China, reported that median times between symptom onset and death were between 16 to 19 days. However, a later study by CO-CIN found that the time from symptom onset until death in UK hospitalised patients was between 7 to 13 days.
These studies show a quite large range so I have taken the average of the two median times of symptom onset to death from each study to calculate a lower (10 days) and upper (18 days) value. Adding the average 5 days from infection to symptom onset gives an estimated range of 15 days to 23 days for the time between Covid infection and death. It is these values that are used in the analysis.
Adam Kucharski, from the London School of Hygiene and Tropical Medicine, highlights a subtle but significant point that needs careful consideration when ‘interpreting the timing of changes in infection (and transmission) from subsequent clinical outcomes.’ In a Sep 2022 tweet, he gives a worked example showing that a sharp fall in infections shortened the observed time between the peak of infections and the peak of reported outcomes (e.g. deaths) below what would be expected.
This rather counter-intuitive finding is illustrated in the two graphs shown below taken from his tweet. The left hand graph gives the distribution of the time between infection to outcome with an average of 25 days. Intuitively, we would expect to see a 25 day lag between a change in infections and a change in outcome, and this is what we normally see. However, a sharp abrupt reduction in infections, as we would expect when implementing a lockdown, changes this normal behaviour and reduces the time lag between the peak of infections and the peak of outcomes. This is shown in the right hand graph where a sharp decline in infections (black line) results in a decline in the reported outcome (red line) some days later. In this modelled example it is 17 days between the two peaks not the expected 25 days.
This is because infections drop off sharply in this scenario, the timing of the peak in clinical outcomes will be influenced by the cumulative shapes of all the delayed outcomes of the infections, the bulk of which occurred before the fall in infections.
In the worked example, we see a change in reported outcomes 8 days earlier than the average number of days between infection and outcome. This is quite a significant ‘pull-ahead’ but it’s important to note that the result is determined by the spread of the delay from infection to outcome - the wider the spread the longer the ‘pull-ahead’. Separately, I have modelled the ‘pull-ahead’ for a delay from infection to death of 25 days using a normal distribution with a standard deviation of 5 days. These gave a ‘pull-ahead’ of 2 to 3 days which is within the range of days I have used for my analysis.