DW: What do you think about the way that some countries in Latin America are managing their numbers of infected and deceased by COVID-19?

Prof. Dr. Dietrich Rothenbacher: The figures currently circulating are not scientifically comparable because they are not collected in a systematic and representative way. The testing strategy is primarily case-based, which means that people with specific symptoms or a certain severity of illness are especially tested. However, what we really need to have valid figures is a defined and systematic method of selecting a representative sampling frame that is comparable across countries.

Subscribe to our special newsletter on coronavirus and click here.

How can a correct accounting of coronavirus cases be carried out in the countries of the region?

Comprehensive testing always costs money, and financial and technological resources are often not available in all countries. The most problematic part of the mortality rate is, for example, the denominator. The denominator depends, to a large extent, on the total number of people evaluated. In countries where few tests are performed, the most serious cases will be diagnosed, because these are cared for in the public health system. Extensive testing in society to include mild or symptom-free cases results in a higher denominator for the calculations, thereby lowering the estimated mortality rate.

So the key would be to run massive tests?

Mass testing of potentially ill people is an important measure of obtaining robust data to estimate mortality. Early testing is also important to quickly find infected and isolate them. So other people don’t get infected. Early interruption of infection chains prevents the virus and, subsequently, diseases from spreading explosively.

Which tests do you trust the most, molecular or rapid tests?

Only in the polymerase chain reaction (PCR) based test is it suitable for virus detection. The precision, that is, the validity, of the tests can vary greatly, which can lead to many false positive and negative results.

When will we know the correct contagion figures? Who has the real numbers?

A detailed observation will always be better done in retrospect, but we have to make decisions today and often based on data that has many uncertainties.

Professor Dr. Dietrich Rothenbacher is director of the Institute of Epidemiology and Medical Biometry at the University of Ulm in Germany.