I recently read in the news from Reuters that Covid-19 was already present in the Europe in March 2019 as reported by the study by Chavarria-Miró et al (2020). According to Chavarria-Miró et al they found traces of SARS-CoV-2 in March 12, 2019 sewage samples from Barcelona.

I am interested in this news because I do research on effects of covid-19 pandemic and government responses to covid-19 on macreconomic indicators. In macroeconomcis we are unable to run randomized control trials but sometimes nature can create natural experiments or situations that allow for use of quasi-experimental statistical techniques, and I think I can use the situation above in such way.

Consequently, I have the following questions:

  1. How conclusive is that Chavarria-Miró et al study? Are there any red flags that would indicate the conclusion drawn from the study should not be taken at face value? Essentially, I am interested in knowing if the study is reliable or if there are possibly other corroborating studies.

  2. Is the SARS-CoV-2 found in march 12 the same SARS-CoV-2 that was circulating in 2020? Or we can't tell? Or to what degree it is similar if it is possible to know (in terms of infectiveness and mortality/morbidity)? If it is not possible to answer this empirically, is it at least possible to answer it theoretically based on how similar viruses mutate? (Here I would also appreciate some references to literature/data if possible).

The reason why I ask this second question is that given how dangerous and infectious is Covid-19 considered, I am surprised that there weren't spikes in deaths in Spain. In fact, according to statistics provided by Statista the number of deaths in 2019 was highest in January and even declined after March. When the number of death is totaled across whole year 2019, and compared to years prior years in data from Statista here, again the number of deaths was comparable to 2018. I realize that just looking at a crude naïve statistics like this is not rigorous but I would expect at least some visible effect if Covid-19 was already spreading in March 2019.

  • $\begingroup$ Do you maybe have your years mixed up? 2019 vs 2020? It's 2021 now, last year was 2020. $\endgroup$ – Bryan Krause Jan 4 at 2:10
  • $\begingroup$ @BryanKrause no actually no, as the Reuters reports "Spanish virologists have found traces of the novel coronavirus in a sample of Barcelona waste water collected in March 2019". Although I was surprised as well and thought it was mistake initially but apparently it is so $\endgroup$ – 1muflon1 Jan 4 at 2:12
  • 2
    $\begingroup$ It's a preprint, not a publication. Regardless of the publication quality, it makes zero sense. We have seen what happens clinically with COVID-19: respiratory distress in a sizeable fraction of infections, a pandemic lasting several months. It isn't feasible that the virus was somehow present in one place in Spain and yet resulted in no broader spread and no clinical notice. $\endgroup$ – Bryan Krause Jan 4 at 3:19
  • 1
    $\begingroup$ @1muflon1 Consider that it is much more likely that a PCR artefact (like Chris described in his answer) can give a one-off false positive results, such as the Italian or Spanish study, than 1000s of SARS-CoV-2 genomes all happen to coalesce to the wrong date, which also happens to be just before the first recorded case occured. The two studies aren't an equal weight of evidence. $\endgroup$ – user438383 Jan 4 at 16:07
  • 1
    $\begingroup$ Related q on Skeptics: skeptics.stackexchange.com/questions/47583/… Note that someone tried to give that Bacelona study as answer but it ended up negatively scored... On the other hand (and this is why I'm linking) you can see there what water analysis looks like in other less controversial instances (Italy, later in 2019); there's a continuous series of positive results... By the way, China could have published such studies too, but they seem very skittish to do so... $\endgroup$ – Fizz Jan 4 at 20:23

To judge about this assay, it is important to see how they worked. They used different sets of primers to analyze for SARS-CoV-2. Namely targets IP2 und IP4 from Institute Pasteur in Paris (see reference 1) and to validate results also primers targeting the "E" sequence (reference 2) and two against the nucleoproteins N1 and N2 (reference 3).

Then they tested frozen samples taken between January 2018 and December 19. Of those, only the sample from March 2019 come back positive for 2 out of 5 primer pairs, none of the samples before and after tested positive, despite mentioning that samples where already positive as early as January 2020 again.

If you look at the published amplification curve you see a very late Ct value for the sample (I would eyeball it somewhere between 39 and 40), which is basically in the area of non-specific amplification. Running the PCR another 10 cycles until 50 is highly unusual in my experience.

enter image description here

Conclusion: I see this result as a PCR artefact and I doubt, this will be published (at least I hope it will not). The reasons for it are:

  • only two of five sets of primers is positive. Since these are highly specific, I would expect all five of them positive.
  • the Ct value is very high, around the usual endpoint of qPCR. If you let PCR run long enough, it will eventually amplify something. The also show no melting curve to prove the have a good PCR product.
  • it is highly unlikely that the virus appears in March 2018 in the wastewater of Barcelona and then disappears again without leaving any trace. This is clearly not the experience we have been making throughout 2020.


  1. Protocol:Real-time RT-PCRassays forthe detection ofSARS-CoV-2InstitutPasteur, Paris
  2. Detection of 2019 novel coronavirus (2019-nCoV) by real-time RT-PCR
  3. CDC 2019-Novel Coronavirus (2019-nCoV) Real-Time RT-PCR Diagnostic Panel

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.