With the declaration that COVID-19 was a pandemic disease, countries all over the world began to appraise possible approaches to mitigate its severity. One is imposing a national lockdown, while another is allowing herd immunity to build. A recent study published on the preprint server medRxiv* in August 2020 shows that lockdown is an effective way to reduce the extent and speed of infections in a country.
Herd Immunity Vs. Lockdown
When faced with the pandemic, Sweden aimed at achieving conferring herd immunity. It was hoped this would indirectly protect the rest of the population by ensuring that the risk of being exposed to the virus was reduced since infection chains cannot propagate among those who are immune to the virus. Proponents of the herd immunity approach say that it is a preferred strategy since lockdown cannot prevent viral spread but only delays it, at an unnecessary cost both economically and socially.
The lockdown approach is recommended, on the other hand, because it does decrease the effective reproduction number – the average number of people infected by each infected individual. When fewer people are infected in this way, the transmission chain is broken, and there are fewer cases to treat as well. Despite much media debate about which measures are better, lockdown or herd immunity, and plenty of research using predictive methods or tools to measure the change in the reproduction number, there was a lack of definitive data on the effectiveness of lockdown measures.
The current study is based on a quasi-experimental technique called interrupted time series (ITS) analysis. Here, the outcome of interest is collected at multiple points in time, both before and after a specific intervention. The trend of the outcome over time is thus observed, to find the effect of the intervention. In studies where randomization cannot or should not be done, this is an alternative practice.
The History of the UK Lockdown
By June 3, 2020, the World Health Organization confirmed about 279,000 and 38,500 cases of COVID-19 in the United Kingdom (UK) and Sweden, respectively, with the number of deaths crossing 39,000 and 4,400 in these countries. England had its first two cases on January 31, 2020, with the first COVID-linked death being reported on March 2. This was followed by escalation, with the lockdown put in place on March 23.
Both England and Sweden initially opted for herd immunity, but England switched to the lockdown strategy later, partly due to a model which showed that without any intervention, about 510,000 people in the UK would die of COVID-19. The two countries were chosen for the head-to-head comparison of daily case and death counts because the delay in declaring a lockdown in the UK allowed the pre-lockdown picture in both countries to be highly similar – a prerequisite for making inferences based on this data.
With the lockdown, people in England were not able to travel out of their homes freely to work, to stores, or for social activities. This was intended to keep the virus from spreading rapidly between people in close contact.
The History of the Herd Immunity Approach in Sweden
In Sweden, the first case was reported on the same day as in the UK, and the first death on March 11. However, Sweden did not adopt the lockdown strategy, instead choosing to make the public responsible for their health by stringent recommendations to stay at home for work as well as avoid traveling, as far as possible, to maintain social distancing, and if above 70 or showing the least signs of COVID-19, to stay at home all the time. However, preschools and elementary schools remained open.
The researchers used the effect of lockdown in England on daily cases and the rate of new cases and deaths in the first 100 days of lockdown, relative to Sweden. They found that the incidence of cases in England over 30 days was over 20,000 cases per 10 million people, vs. 15,000 in Sweden. Thus, England had a relative case incidence of 1.32.
In England, the incidence rate was approximately 3,400 deaths per 10 million people, vs. 2,400 in Sweden. Thus, the relative death incidence was 1.42. The case fatality in England and Sweden was 17% and 16% respectively.
Finding the Effect of Intervention
The study analyzed any possible effect of lockdown, by calculating the adjusted daily incidence of cases and deaths in England and Sweden, relative to the population of these countries. They observed that pre-lockdown, the adjusted daily case incidence in England exceeded that of Sweden by 8 per 10 million people. On the day of lockdown, there were +693 cases per 10 million people in England relative to Sweden.
COVID-19 deaths in England were in excess of those in Sweden by 1.5 daily deaths per 10 million people before the lockdown. As a result, the day of lockdown saw 50 more deaths per 10 million people in England compared to Sweden.
After the lockdown began, the daily case incidence dropped by 19 cases per 10 million people in England, compared to Sweden. The daily death rate went down by two deaths per 10 million people in England compared to Sweden.
What about after the restrictions were lifted? In England, the lockdown was lifted in a phased manner, but neither the daily case incidence nor the death rate increased in either phase. In phase 1, beginning May 13, more people were permitted to travel outdoors and to go to work, while on June 1, with phase 2, restrictions on children were lited in a phased manner. More local transport systems in cities, and some more stores, were allowed to reopen. Two households could meet together regularly rather than only those in a single household.
All these relaxations were accompanied by enhanced education about observing social distancing, wearing masks, and hand hygiene.
Potential limitations and Implications
The use of interrupted time-series analysis allowed the atypical and dynamic growth of the pandemic to be modeled using comparisons. This allowed both countries to be compared even though it was unclear if the same trends would continue, and because more complex patterns could be modeled in a nonlinear way.
Of course, other confounding factors may have affected the result, since these are countries at some distance from each other, with different climates and geography. However, the use of similar tools, the presence of a similar life expectancy, and the onset of the pandemic at the same time reduce the probability of significant confounding, according to the authors.
The researchers conclude, "The estimated causal effect is the result of net anti-COVID-19 measurements implemented by England but not Sweden." Whereas the higher rate of spread in England before the lockdown vs. Sweden led to a higher mortality rate, once the lockdown was in place, the daily cases went down, accompanied by the daily death rate.
The study does not look at long-term outcomes of the outbreak, including a fresh winter outbreak or a second wave, occurring after phase 3 relaxation of the lockdown in England, which has already started. However, it is indisputable that short-term favorable outcomes were observable after the UK lockdown, due to a break in transmission. The researchers point out, "Future research is needed to understand the long-term net outcome of the anti-COVID-19 policy implemented in both countries."
medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.