The pandemic was only a few weeks old when the American Paul Romer sat down with his colleagues at the computer and calculated. Romer is an economist who received the Nobel Prize for his analyzes of technical innovations. In March 2020, the classic containment strategies had failed everywhere. Sick people were tested and infected people and their contacts isolated, but the spread of the virus was still not under control. Almost all countries imposed lockdowns.
end of April Romer published a study about a possible way out of the pandemic. If all people were tested regularly, the effective number of reproductions could be reduced to below 1. This number is crucial in the pandemic. Due to its infectivity, the virus specifies the so-called basic reproduction number, which indicates how many people are infected by an infected person on average. Containment measures can “effectively”, i.e. in reality, reduce this number. If it is below 1, the case numbers decrease. Also through mass tests?
The Romers model was just a bold thought in April 2020. The PCR test capacities were not even remotely sufficient for this. Around two years later, however, one country comes closer to this utopia than any other: Austria. So far, more than 138 million corona tests have been carried out, 56 million of which were PCR tests. In Vienna alone, two million samples are evaluated every week, statistically one from every Viennese. In reality, the distribution is different. But more on that later.
The fact that Austria became the self-proclaimed test world champion was not strategically planned at all. As so often in this pandemic, the government stumbled into this measure. It began with the ultimately unsuccessful mass test in December 2020, which ex-Chancellor Sebastian Kurz hastily announced. The planned repetition of this experiment after the holidays was supposed to become a “test out” from the Christmas lockdown, but the SPÖ still negotiated it into a “pure test” in gastronomy and events. Austria was thus one of the first countries to decide on such access authorizations.
This did not lead to a comprehensive screening of all people, but it increased the initially tentative demand for tests. Especially from the point when the lockdown actually ended. The primary aim of the access tests was to reduce the risk of an unnoticed infected person visiting a bar and possibly infecting many people there. But of course it was also hoped that more people would test themselves as a result.
As early as January 2021, the simulation researcher Niki Popper also has the epidemiological effect of a broad population screening. A “testing away” of the virus, as Romer had calculated a year earlier, cannot be read from Popper’s study, but there is a “significant reduction” if around 100,000 households per day were tested using PCR. Back then, in January, this number was still unrealistic, but in May 100,000 were actually gurgling in Vienna every day. At the time, however, the virus was on its summer break.
Test strategy has already cost 2.6 billion euros
The federal government decided to roll out the system of pooled PCR tests throughout Austria. The federal states were responsible for implementation, but failed to do so. Instead, some ÖVP state governors called for the tests to be charged. Some politicians wanted to get out of the strategy that Austria had stumbled into in order to achieve something completely different, namely a higher vaccination rate.
But there was also the cost argument. Every day, millions of euros flow into the corona tests in Austria. According to the Ministry of Finance, the federal government spent around 2.6 billion euros on this in 2020 and 2021. Vienna City Councilor Peter Hacker countered the cost argument in August: “Lockdowns cost X times as much.” A few months later, Austria was not only the “test world champion”, but also one of only three European countries in another lockdown (at a cost of 400 million euros per week, as determined by Wifo). Despite the measure of multiple testing, it was not possible to get through this delta wave better than countries and regions with a similar vaccination rate such as Switzerland and Bavaria. Why?
An evaluation of this measure is difficult. On the one hand, important data are missing. A seroprevalence study, in which infections that have gone unnoticed can be detected via specific antibody detection, would be particularly important, as complexity researcher Peter Klimek explains. “This would allow you to see how much of the underreporting was skimmed off by testing healthy people.” But you would also have to determine how many infected people are found via which test tracks.
On the other hand, people’s behavior is an unknown that is difficult to determine. Already in the fall of 2020, when the antigen tests came up and epidemiologists from Harvard with them calculated new “path testing” models, researchers from Cambridge thought a study against, which also took into account changed behavior patterns. If the tested had more contacts than the untested, the positive effect could even be reversed, since a negative result is only a snapshot. But do the (tested) Austrians behave differently than the untested Bavarians? That could not be determined seriously, says Klimek.
Vienna came through the delta wave a little better
What was striking about the delta wave: Vienna, which mainly relied on PCR tests and not on antigen tests, had a lower incidence and also a lower burden on the hospitals than other federal states. Whether this was due to the test strategy or whether other reasons (also) play a role here cannot be said from observation alone.
What is certain is that the virus has become faster from variant to variant. This is also confirmed by the statistician Erich Neuwirth. “The time between the waves of infection and the waves in the hospital has also become shorter and shorter,” he says. If it only takes two to three days from infection to infectivity, testing can hardly keep up. And not even with PCR tests. Popper also says, especially with regard to Omikron, that “the effectiveness is even worse”.
This also applies to the area of application of tests as a safety net before private meetings, where the virologist Elisabeth Puchhammer-Stöckl has already warned of “false security” that could convey a negative result from the previous day. Since Delta was already faster than previous variants, this could also have been a reason why the overall epidemiological benefit was less than hoped for in the fall.
In addition, the tests are not evenly distributed across the population. This is regularly asked in the Austria Corona Panel, most recently 25 percent stated that they had tested themselves more than four times in the past month, but a similar number did not do this at all and another 30 percent only once or twice in four weeks. It is particularly striking that people with a tertiary education (university) disproportionately use this offer as well as young people (under 26 years of age).
However, both Popper and Klimek do attribute an epidemiological effect to the test strategy. Klimek speaks of 10 percent, Popper of 15 percent (with 400,000 tests per day) by which the effective number of reproductions is reduced by this measure alone. That’s not nothing, but it alone won’t prevent a lockdown. In the case of a reproduction number of 1.3, it would drop to 1.1 to 1.2 through the tests alone. “But you save yourself one or the other measure,” says Klimek.
Early diagnosis important for treatment
His team, led by Jana Lasser, also looked at the effectiveness of school tests. Although the screening has a relevant effect, if you look at all measures separately, Ventilation reduces the clusters the most. (see page 20.)
How the test strategy will develop is still uncertain. The national crisis coordination is planning a further PCR expansion from April. However, this is not without controversy within the committee. The question is: What goals are you pursuing with the strategy? In any case, there is a lack of evidence on the epidemiological effectiveness, and it is certain that Omikron reduces the previous effectiveness again.
One aspect that could also speak in favor of broad and low-threshold availability is the possible therapeutic benefit in connection with new drugs such as molnupiravir or paxlovid. They are effective in preventing severe courses, but must be taken early in the course of the disease. However, if only sick people are tested, perhaps not even on the first day of symptoms because the surgery is already closed, the medical diagnosis and recommendation for this medication can come too late. On the other hand, if you gargle quickly at home when you first notice a symptom and take the sample to the supermarket, you can gain two to three days. The time gained in this way can be decisive for the success of the treatment.