Why Epidemiological Testing Is Important

I encourage you to read this post by Luigi Zingales at City Journal:

The Italian experience suggests that locking down towns is a necessary but insufficient condition to stop the spread of Covid-19. If 50 percent of the infected are asymptomatic, there is no hope of containing the disease unless we subject ourselves to massive testing. On February 22, 89 inhabitants of Vò Euganeo, a small town close to Padua, were found to be infected. Weeks later, the town is free of infection. The lesson of Vò Euganeo is of value to the world as it confronts this unimaginable crisis.

Vò Euganeo is a small town of 3,341 people located on the hills just outside of Padua, where I grew up. It came to international prominence on February 21, when Adriano Trevisan, a 77-year-old inhabitant, died of coronavirus. What will ensure Vò Euganeo a place in the history of medicine is the decision made by the Governor of the Veneto region (which includes Padua, Venice, and Verona) to test all 3,341 inhabitants of the town twice: the first time before closing it off from the rest of Italy, and a second time two weeks later.

In this respect, Vò Euganeo resembles the Diamond Princess, the ship that was quarantined in the port of Yokohama with about 3,700 passengers, making it comparable in size to the town. In both cases, we can observe an entire population exposed to the virus over time, with comprehensive testing. This is different from all other cases, where only a fraction of the population—usually symptomatic—is tested.

To understand what works in containing the spread of the disease, we need to have an accurate count of the infected. Reported infection numbers differ wildly from actual rates of infection and this gap might change over time, as regions expand their testing protocols differently. The experience of Vò and the Diamond Princess demonstrate how various testing approaches can have radically different results. On February 22, 3 percent of the inhabitants of Vò were infected. After two weeks, during which the town was locked down, only 0.25 percent were infected. Once these few infected people were isolated, the town reopened and has experienced no new cases. Compare this strategy with the one adopted in the Diamond Princess. Initially, passengers and crew members were tested when they showed symptoms, and only after that were they taken off the boat. Only toward the end did the authorities test all the passengers. As a result, what started with the infection of one passenger eventually contaminated 20 percent of the people on board. Thus, in one case we go from 3 percent to 0 percent infections, while in the other case from 1 percent to 20 percent.

Read the whole thing.

There are actually two messages here. The first is that the type of testing we’re doing now cannot achieve the objective of reducing the spread of the disease. The second is that epidemiological testing can. That can either be accomplished by sampling, universal testing, or a combination of the two.

Just as is the case with perfect compliance, completely universal testing will remain beyond our reach. We can’t even get everybody to immunize their kids against chickenpox or measles, for goodness sake. We can do better than we are but that will require many, many more tests with quicker results than are presently available and more reliance on epidemiological testing even if it means a reduction in diagnostic testing.