In March 2020, only 1-3% of actual COVID cases were detected in the U.S., researchers found in a study published in the highly prestigious science journal, Nature, this week, titled "Cryptic transmission of SARS-CoV-2 and the first COVID-19 wave." Only <3% of detected cases mean that SARS-CoV-2, the causative virus of COVID-19, was actively spreading by early January 2020, right after the discovery of COVID-19 in China in December 2019.
Researchers at the Northeastern University, Boston, performed a massive mathematical modelling analysis that integrates real-time human mobility and population data in the continental U.S. and 30 other European countries, as well as SARS-CoV-2's incubation time and transmission rates (the R-naught). The researchers have also validated their model's prediction accuracy with the COVID-19 scenario in recent months.
"The model estimates of the median daily incidence of new infections up to February 21, 2020, for both the U.S. and Europe," the study authors wrote. "These values are much larger than the number of officially reported cases, highlighting the substantial number of potential transmission events that may have already occurred before many states and countries had implemented testing strategies independent of travel history."
Their findings make sense. As SARS-CoV-2 testing technology improves, we can detect more COVID-19 cases that are happening. Before March 2020, many countries had not developed and implemented a robust SARS-CoV-2 surveillance system. Therefore, cases before then were vastly underestimated. And we vastly underestimated the threat of SARS-CoV-2, allowing it to spread without us knowing. No wonder SARS-CoV-2 or Covid-19 took us by surprise, spreading globally at an unprecedented rate.
"Our results suggest that many regions in the U.S. and Europe experienced an onset of local transmission in January and February 2020, during the time when testing capacity was limited," the authors explained. " As testing capacity increased and more cases were detected, many governments began to issue social distancing guidelines to mitigate the spread of SARS-CoV-2."
The model further identifies that the first week of SARS-CoV-2 transmission likely occurred by the week of 26 January 2020 in California, 2 February in 2020, and end of January 2020 in Europe, Italy, the U.K., Germany and France. But "these dates also suggest that it is not possible to rule out introductions and transmission events as early as December 2019," the study authors cautioned.
More concerningly, we also did not know much about COVID-19 in early 2020, especially how to treat it. It was basically trial and error back then. We put patients on invasive mechanical ventilation for lung support too soon, for example, which may have done more harm than good. We give drugs like hydroxychloroquine that could cause toxicity in high doses. Therefore, the earlier spread of SARS-CoV-2 also means that we may have underestimated the true global health burden of COVID-19.
This study finding is also consistent with what other studies have found. For instance, the CDC published a study, titled “Serologic testing of U.S. blood donations to identify SARS-CoV-2-reactive antibodies: December 2019-January 2020,” back in November 2020. In this study, researchers collected leftover sera from 7,389 donated blood samples from donors without suspected viral or bacterial respiratory infection. Antibody testing was then performed.
Results detected antibodies specific for the spike protein of SARS-CoV-2 in 1.43% (106 out of 7,389) of samples. Of these 106 cases, 39 belonged to blood samples collected between 13–16 December 2019 from California, Oregon, and Washinton. The other 67 cases were sampled from 30 December 2019 to 17 January 2020. And note that antibodies typically form 1-3 weeks after infection, as the adaptive immune system needs time to activate. This means that the true infection encounter in the CDC study might have even been three weeks before 13 December 2019.
Researchers in Italy also did a similar study as the CDC, titled “Unexpected detection of SARS-CoV-2 antibodies in the prepandemic period in Italy.” Herein, the study caught antibodies specific for the SARS-CoV-2 receptor-binding domain (RBD) in 11.6% (111 out of 959 persons) of blood samples, of which 14% were sampled during September 2019.
Data from China also indicates that the earliest detected COVID-19 case was on 17 November 2019 in a 55-year-old person in Hubei. By the end of November 2019, there were nine cases of Covid-19. This data corroborates a study published in The Lancet that describes a Covid-19 patient with symptom onset dated 1 December 2019 in China.
In the end, SARS-CoV-2 has most probably entered the human population much earlier than we thought, further exposing shortcomings in the current infectious disease surveillance system.