Lesson From Covid Testing
The only metric available to the world to measure the severity of the Covid 19 outbreak is the number of confirmed cases. This of course, also relies on the testing rate/ method (random, group etc) adopted by the particular government, testing rate also in addition depends on the strictness of restrictions adopted by the government to curtail the spread. Therefore countries that managed to contain the outbreak, and were able to keep the number of infects low, had to perform fewer tests to monitor the outbreak adequately vis-a-vie countries where it spread more widely.
For this same reason the timing of testing (vigorous) matters, a vigorous testing model in time of low cases is crucial in curtailing the further spread of the virus ( that is at the beginning stage). Therefore the total number of confirmed cases only tells part of the story, additional question in terms of when were these tests conducted & what the outbreak scenario in the country was at that time; this helps in providing a better picture about the response by the various governments.
South Korea, UK, US and Italy all had discovered their first case in late January, however each one of them adopted a very different approach to testing and as a result experience very different outbreaks. Below, there is a comparison between the four mentioned countries and their testing strategies, subsequent growth in confirmed cases over time. There are three windows of time between mid-January and early May. Overall what we observe is that Korea tested early and widely to monitor the outbreak and bring it under control whereas the other three did not.
Below we show the same chart three times, for three different periods: firstly only up to 7th March, then up to 7th April and finally up to 7th May. It shows how the daily number of confirmed cases (horizontal axis) and the daily number of tests (vertical axis) in each country has changed over the course of their outbreak.
Part 1) Tests up to 7th of March:
South Korea: As visible they started with widespread testing early on, First few days of February they shifted its policy to open public testing including on asymptomatic people; they conducted contact tracing for all confirmed cases. From the chart, testing ramped up to more than 10,000 people per day. By mid — March, the cases in the country were pass the peak, the country never saw more than 600 new confirmed cases per day. Also, as the new cases per day fell the testing continued at the same rate.
United Kingdom: The country tested hundreds of people; they got a head start by testing every early case coupled with comprehensive contact tracing. But as visible from the charts by March the testing was unable to keep up with the new cases each day. Cases rise while testing stagnates.
United States: For this crucial period, no data of testing was published.
Italy: testing in the country was pretty slow, throughout early march the testing rate was stagnant while the number of cases continued to increase steadily. As the line moves horizontally, the number of tests per confirmed case falling from 10 to around 5 (grey dotted lines). The outbreak was far more widespread and growing at a faster rate, than indicated by the number of confirmed cases.
Part 2) Tests up to 7th of April:
United States: Testing only stated in Mid- march, it however increased rapidly, to more than 100,000 tests per day by the end of the month. However, as the number of tests outstrips other countries, so did the outbreak. By early April, the number of new confirmed cases passed 30,000 per day (50 times higher than South Korea’s peak).
Italy: It was pass its peak of 6000 new cases per day roughly ten times that of Korea, the outbreak gradually slowed, while continuing to accelerate.
South Korea: Similar level of testing continued in the country, despite the number of new cases falling. They tested 10,000 people each day in early April, fewer than 100 confirmed being infected.
The gap between Italy and Korea is more than what is shown in the chart because in case of Korea number represents the confirmed cases and in case of Italy number of people tested; since people can be tested more than once’ the real magnitude of the gap between the two is much larger.
United Kingdom: Early April saw a considerable increase in testing in the UK, in absolute terms. However the outbreak in the country had grown far more quickly. The policy of contact tracing ceased on 12th march, testing was never extended beyond symptomatic people. 10,000 people were tested each day in April, 4,000 of which were confirmed infected. Such a ratio is dangerous as it indicates clearly that many infections were going undetected (x2 grey line which is comparatively very low testing).
While the Four countries stated the same, with their first cases of Covid 19 being discovered in late January; three month later they were clearly following very different paths.
Part 3) Tests up to 7th of May:
United States v/s South Korea:
The outbreak in the US continued to accelerate until Mid- April, a month after Korea’s peak. By early may the confirmed cases in US passed one million (one hundred times more than total of Korean cases). Cases gradually started to decline mid- April, the US continued to increase its testing capacity; which now dwarfs other countries in absolute terms, however in relative terms the testing remains low by international standards. Late April saw the number of tests –per confirmed case rise, for the first time in the country.
United Kingdom v/s Italy:
New confirmed cases peaked in both of the countries roughly at the same level; 6000 per day. Italy the daily cases post that started to fall rapidly, while in the UK they remained at the same level for an additional month. Differences is testing may explain this difference.
In UK, levelling of new confirmed cases coincided with the dramatic acceleration in testing. After 2 months the people tested had jumped from 10,000 to 60,000 each day. Growth of the outbreak however may be understated due to limited testing; this rapid acceleration may have understated the pace with which the outbreak was receding.
What these charts show us is how closely testing is connected to the total number of confirmed cases in a country. It is not possible to learn about the outbreak from the number of confirmed cases, unless the exact extent of testing done by the country is known. In a country like UK where every other person tested is found to be infected with the disease, indicates that there are many undetected infected people. In such a case the number of confirmed cases gives an unreliable indication on the extent of the outbreak. On the other hand where only 100 of the total tested population is found to be infected, as in South Korea- the case numbers gives a more reliable picture of the outbreak in the country.
Its only relative to the size of the outbreak- as indicated by the confirmed cases- that the countries adequacy of testing response can be judged meaningfully. The US as mentioned has tested far more people than any other country, but the extensive testing came much later in the outbreak when the virus had already widely spread (vis-a-vie South Korea). Testing in absolute terms may be high in the US, but in relative (to the outbreak) it remains low (as of mid- may data).
Overall, to observe the success of the countries in terms of their response to the pandemic; one cannot look at the number of confirmed cases or testing data in isolation, they need to look at both of them together.