Evidence of lockdown efficiency against COVID

Today I’ve read a paper about the effectiveness of lockdown policies against Covid. The paper’s source can be found at the bottom of this article.

Most countries have responded to the Covid crisis with lockdown policies that restrict people’s mobility to limit the spread of the virus. The impact of such policies on the economical world is huge which should incentivize politics to study and quantify the risk-benefit ratio of lockdowns.

The authors of today’s paper provide a quick analysis that seems to strongly indicate that lockdowns are, indeed, effective against Covid. They took the daily death data of many countries and noted when these countries started to implement restriction measures. From these inputs, they looked at the normalized daily deaths (normalized over the total number of deaths) from the start of the lockdown for each country.

This kind of plot represents how does the number of deaths evolves after lockdown policies begin to be effective in each country. If the lockdown wasn’t effective at all, we could expect the number of deaths to keep increasing steadily with little to no change to its rate. On the opposite, if the lockdown was effective we could expect to see a collapse of the curve at some point.

The author’s results are pretty clear for all countries (except Sweden): the normalized daily deaths ratio keeps increasing for the first 20 days of lockdown and then collapse. Interestingly, 20 days correspond to the typical time delay between infection and death. The first 20 days of the curve when the normalized daily deaths keep increasing corresponds to people previously infected who keep dying since the lockdown doesn’t have any effect on their outcome. After 20 days, since fewer people have been infected, the normalized daily deaths logically decrease.

This pattern is present for all countries except Sweden, which includes countries with both quick and slow responses to Covid, countries which low to high mortality, and countries with moderate to strict lockdown policies.

The authors conclude their papers by stating that “even if correlation does not prove causality, the quality of the curve collapse […] provide strong evidence of a positive effect of lockdown measures in curbing the number of fatalities”.

Source: Bradde, S., Cerruti, B. and Bouchaud, J.P., 2020. Did lockdowns serve their purpose?. arXiv preprint arXiv:2006.09829.