In December 2019, several patients from Wuhan, People’s Republic of China, developed pneumonia and respiratory failure reminiscent of the SARS epidemic in 2003 (WMHC 2019, www.SARSReference.com). In early January 2020, a new virus, later denominated SARS-CoV-2, was isolated from bronchoalveolar lavage fluid samples and found to be a betacoronavirus (Zhou 2020). The virus spread first within China (Yu X 2020) and to several countries in Asia before reaching Iran and Italy where it caused major outbreaks. During the first 11 weeks of the pandemic, almost two-thirds of first cases in affected countries were in people reported to have recently travelled from only three affected countries (China, Iran, or Italy), showing how international travel from a few countries with substantial SARS-CoV-2 transmission might have seeded additional outbreaks around the world (Dawood 2020).

Open the references of this chapter in a separate window.

See also the new 5th Edition references with a short comment.

Despite some early successes in containment, SARS-CoV-2 eventually took hold in both Europe and North America during the first two months of 2020: in Italy around the end of January, in Washington State around the beginning of February, followed by New York City later that month (Worobey 2020 – see also Figure 6, Deng X 2020, McNeil Jr DG). In Brazil, it was found that there had been more than 100 international virus introductions, with 76% of Brazilian strains falling into three clades that were introduced from Europe between 22 February and 11 March 2020 (Candido 2020).

Between then and the time of this writing (31 October), SARS-CoV-2 has spread to every corner of the world. More than 40 million people have been diagnosed with SARS-CoV-2 infection and more than a million people have died of COVID-19, the disease caused by SARS-CoV-2. Not all cases, in particular if asymptomatic, have been diagnosed and the true number of infections and deaths is probably much higher. Relatively few large scale seroprevalence studies have been completed but the available seroprevalence data show that only a few places, like Mumbai and Manaus, have reached a high prevalence in the population, close to the level required for some kind of herd immunity (see Table 1). [Herd immunity is defined as the proportion of a population that must be immune to an infectious disease, either by natural infection or vaccination, to provide indirect protection (herd protection) to those who are not immune to the disease (D’Souza 2020, Adam 2020).

COVID Reference resources:

Home | Daily Top 10 | 7-Day Overview | Download | Epidemiology | Transmission | Prevention | Virology | Vaccines | Diagnosis | Clinical Manifestations | Treatment | Severe COVID-19 | Comorbidities / Special Populations | Pediatrics| Timeline | Preface

Table 1 shows that countries hit hardest by the COVID-19 pandemic have higher seroprevalence rates but, without an effective vaccine, no country can count on any kind of herd immunity in the near future.


Table 1. Seroprevalence data 2020
Sample collection
Italy* Nationwide May 25-July 15 2.5% Sabbadini 2020
Italy Lodi (red zone) 23% Percivalle 2020
Spain Nationwide
Pollán 2020
Spain Madrid 11% Soriano 2020
Switzerland Geneva 5.0-11% Stringhini 2020
Denmark Faroe Islands 0.6% Petersen 2020
South West
Ward 2020
China Wuhan March 9-April 10 3.2-3.8% Xu X 2020
US New York City
San Francisco Bay area
March 23-April 1
April 23-27
Havers 2020
US New York State 14% Rosenberg 2020



NYC, Health care personnel

Nationwide in patients receiving dialysis



July 2020




Moscola 2020


Anand 2020

India Mumbai July 57% Kolthur-Seetharam 2020
Brazil Manaus March-August 66% Buss 2020
Not peer-reviewed. Results have recently been questioned.

* Note that Italy’s national survey results are preliminary and probably an underestimation. The country only managed to collect 40% of the planned samples, with many people refusing to be tested. Insiders never believed these figures and favored a seropositivity rate of 5-10% like in Spain or in France. Now we have a new estimate of COVID-19 prevalence in Italy by Francesca Bassi and colleagues: 9%, corresponding to almost 6 million Italians (Bassi 2020).

The articles cited in Table 1 report some interesting findings:

  • Wuhan – Seropositivity for IgM and IgG antibodies was low (3.2%-3.8%) even in a highly affected city like Wuhan (Xu X 2020).
  • New York City – In New York, the prevalence of SARS-CoV-2 among health care personnel was 13.7% (5523 of 40,329 individuals tested) (Moscola 2020) which was similar to that among adults randomly tested in New York State (14.0%) (Rosenberg 2020).
  • UK – Black, Asian and minority ethnic (BAME) individuals were between two and three times as likely to have had SARS-CoV-2 infection compared to white people. An interesting trend: young people aged 18-24 had the highest rates (8%), while older adults aged 65 to 74 were least likely to have been infected (3%).
  • Mumbai – In a cross-sectional survey the prevalence of past SARS-CoV-2 infection in three areas in Mumbai was around 57% in the slum areas of Chembur, Matunga and Dahisar, and 16% in neighboring non-slums (Kolthur-Seetharam 2020). In some places of the world herd immunity may be within reach.
  • Geneva – Young children (5–9 years) and older people (≥ 65 years) had significantly lower seroprevalence rates than other age groups (Stringhini 2020).
  • Faroe Islands – At the beginning of the pandemic, small islands tended to have low seropositivity rates.

Free Download: 5th Edition

It is worth noting that we still have few nationwide population-based seroprevalance studies, that the sensitivity and specificity of serological tests being used can vary from place to place, and that some people might have been infected without showing detectable levels of antibodies. Based on all available serological studies, WHO has estimated that around 10% of the world population, or 760 million people, might have been infected as of October 2020. https://www.euronews.com/2020/10/05/around-10-of-the-world-s-population-may-have-had-covid-19-according-to-who

The mean incubation period of SARS-CoV-2 infection is around 5 days (Li 2020, Lauer 2020, Nie X 2020). The serial interval – defined as the duration of time between a primary case-patient having symptom onset and a secondary case-patient having symptom onset – has been estimated to be between 5 and 7.5 days (Cereda 2020). SARS-CoV-2 is highly contagious, with an estimated basic reproduction number R0 of around 2.5-3.0 (Chan 2020, Tang B 2020, Zhao 2020). [R0 indicates the average number of infections one case can generate over the course of the infectious period in a naïve, uninfected population. Read the guide by David Adam (Adam 2020) for more precious information on R0.]



SARS-CoV-2 is easily transmissible both by symptomatic and asymptomatic individuals, thrives in closed and densely inhabited environments, and is amplified by so-called ‘superspreader’ events.

The five golden rules to minimize the risk of SARS-CoV-2 infection

  1. Wear face masks in public spaces.
  2. Keep a distance of 2 (two!) meters to other people.
  3. Avoid crowded places (more than 5-10 people).
  4. Avoid in particular crowded and closed spaces (even worse: air-conditioned closed places where air is being moved around).
  5. Avoid in all circumstances – crowded, closed and noisy spaces where people must shout to communicate. These are SARS-CoV-2’s preferred playgrounds.

Find below a detailed discussion of SARS-CoV-2 transmission (pages 93) and its prevention (page 149).


As with the earlier SARS and MERS outbreaks (Shen Z 2004, Cho SY 2016), the spread of SARS-CoV-2 is characterized by the occurrence of so-called “superspreaders events” where one source of infections seems responsible for a large number of secondary infections. (Wang L 2020) This phenomenon is well described by a recent study of SARS-CoV-2 transmission in Hong-Kong (Adam DC 2020). The authors analyzed all clusters of infection in 1038 cases that occurred between January and April 2020 and concluded that 19% of cases were responsible for causing 80% of the additional community cases, with large clusters originating from bars, weddings, and religious ceremonies. Interestingly, decreased delays in confirmation of symptomatic cases did not influence the rate of transmission (suggesting higher rate of transmission at or before symptom onset), whereas rapid contact tracing and quarantine of contacts was very effective in terminating the transmission chain. Other authors (Endo 2020) have estimated a k of 0.1 outside China, meaning that only 10% of infected individuals transmit the virus (k or dispersion factor describes, in mathematical models, how much the disease tends to cluster).

A relatively low dispersion factor with few infected people causing most transmissions could explain some puzzling aspects of the beginning of the COVID-19 pandemic. For example, why the early introductions in Europe of SARS-Cov-2 in December 2019  (France) and in January 2020 (France, Germany) did not result in earlier outbreaks. Or why the large outbreak in Northern Italy in February 2020 did not lead to a similar rapid spread of the virus in the rest of the country.

Understanding the reasons underlining superspreader events can be key to the success of preventive measures, so the big question is, “Why do some COVID-19 patients infect many others, whereas most don’t spread the virus at all?” (Kupferschmidt 2020). It is possible that some individuals simply shed more virus that others, or that there is much more shedding at a specific moment of higher contagiousness in the natural history of the infection, possibly when viral load is at its peak. Environmental conditions also clearly play a role, with crowded, closed places where people talk loudly, shout, sing or exercise being at higher risk, possibly because of the higher production and diffusion of small particles like aerosols. A “superspreader individual“ in a “superspreading setting“ may result in a very large number of infections, as seen in the Shincheonji church cluster in South Korea where, in March 2020, one single person was estimated to have generated more than 6000 cases.

A better understanding of superspreader events may help in defining the most effective measures to reduce SARS-CoV-2 transmission by reducing the likelihood of superspreading events. We will explore below the most common “hotspots” of SARS-CoV-2 infection, where the likelihood of single or multiple infections is higher.


In this chapter, we will discuss:

  1. Hotspots of SARS-CoV-2 infection
  2. Special aspects of the pandemic in selected places
  3. The Pandemic: Past and Future

Hotspots of SARS-CoV-2 Transmission

The following settings were, are or can be catalyzers of outbreaks:

  • Hospitals
  • Nursing facilities
  • Homes (also including intense social life with friends and colleagues)
  • Leisure facilities such as bars, clubs, choirs, discos, sports facilities and restaurants
  • Workplaces
  • Schools
  • Mass gatherings
  • Marriages
  • Funerals
  • Religious gatherings
  • Closed and densely populated spaces
  • Prisons
  • Homeless shelters
  • Ships (closed spaces)
    • Cruise ships
    • Aircraft carriers and other military vessels


During the first months of the SARS-CoV-2 pandemic, when suspicion of the disease was low, transmission in hospitals and other health care centers (including doctors offices) played a prominent role in the origin and spread of local epidemics. This was reminiscent of SARS and of the largest MERS outbreak outside of the Arabian peninsula which occurred in the Republic of Korea in 2015, where 184 of 186 cases were infected nosocomially (Korea Centers for Disease Control and Prevention 2015). Hospitals, as many other places where strangers meet, can be a favorable environment for the propagation of SARS-CoV-2 (Wison 2020). Within the first 6 weeks of the epidemic in China, 1716 cases and at least 5 deaths (0.3%) were confirmed among health care workers by nucleic acid testing (Wu 2020). In some instances, hospitals could have been even the main COVID-19 hub, facilitating transmission between health workers and uninfected patients (Nacoti 2020).

One hospital environment study reports that the virus was widely present in the air and on object surfaces in both the intensive care units and general wards, implying a potentially high infection risk for medical staff. Contamination was greater in ICUs (Self 2020). Virus RNA has been found on floors, computer mice, trash cans, sickbed handrails, and was detected in the air up to approximately 4 m from patients (Guo 2020). The virus was also isolated from toilet bowl and sink samples, suggesting that viral shedding in stool could be a potential route of transmission (Young 2020, Tang 2020). However, most of these studies have evaluated only the presence of viral RNA, not its infectivity.

Although nosocomial spread of the virus is well documented, appropriate hospital infection control measures can prevent nosocomial transmission of SARS-CoV-2 (Chen 2020, Nagano 2020, Callaghan 2020). This was nicely demonstrated by the case of a person in her 60s who travelled to Wuhan on Dec 25, 2019, returned to the US on Jan 13, 2020, and transmitted SARS-CoV-2 to her husband. Although both were hospitalized in the same facility and shared hundreds (n = 348) of contacts with HCWs, nobody else became infected (Ghinai 2020).

However, working in a high-risk department, longer duty hours, and sub-optimal hand hygiene after coming into contact with patients are all associated with an increased risk of infection in health care workers (Ran 2020). At one time, during the early epidemic in March 2020, around half of 200 cases in Sardinia were among hospital staff and other health care workers. On 14 April, the US CDC reported that 9282 Health Care Personnel had been infected with SARS-COV-2 in the US. Health care workers from COVID-19 have a higher risk of being SARS-CoV-2 infected (5.4%) than those from non–COVID units (0.6%) (Vahidy 2020). In a prospective cohort study in London, 25% of HCWs were already seropositive at enrolment (26 March to 8 April) and a further 20% became seropositive within the first month of follow-up (Houlihan 2020). However, a Chinese study of 9684 healthcare workers (HCW) in Tongji Hospital showed a higher rate of infection in non-first-line HCW (93/6.574, 1.4%) when compared to those who worked in fever clinics or wards (17/3110, 0.5%) (Lai X 2020). Interpretation: those who worked in clinical departments other than fever clinics and wards may have had less access to, or have neglected to adopt, adequate protective measures.

The risk factors for SARS-CoV-2 infection in health care workers have been summarized in a recent review (Chou 2020). There is evidence that more consistent and full use of recommended PPE measures was associated with decreased risk for infection. Association was most consistent for masks but was also observed for gloves, gowns, and eye protection, as well as hand hygiene. Some evidence was found that N95 respirators might be associated with higher reduction of risk for infection than surgical masks. Evidence also indicated an association with certain exposures (such as involvement in intubations, direct contact with infected patients, or contact with bodily fluids).

SARS-CoV-2 outbreaks have also been documented in dialysis units (Schwierzeck 2020, Rincón 2020). The prevalence of SARS-CoV-2 antibodies was lower among personnel who reported always wearing a face covering while caring for patients (6%), compared with those who did not (9%) (Self 2020).

Long-term care facilities

Long-term care facilities (LTC) are high-risk settings for infectious respiratory diseases. The first important study published in May 2020 reported an outbreak in a skilled nursing facility in King County, Washington, US, where 167 cases of COVID-19 (101 residents, 50 health care personnel and 16 visitors) were diagnosed within less than three weeks from the identification of the first case: (McMichael 2020) (Table 2).

Table 2. COVID outbreak in a long-term care facility
(N = 101)
Healthcare personnel
(N = 50)
(N = 16)
Median age (range) 83 (51-100) 43.5 (21-79) 62.5 (52-88)
Female (%) 68.3 76 31.2
Hospitalized (%) 54.5 6.0 50.0
Died (%) 33.7 0 6.2
Chronic underlying conditions (%)
Hypertension 67.3 8.0 12.5
Cardiac disease 60.4 8.0 18.8
Renal disease 40.6 0 12.5
Diabetes mellitus 31.7 10.0 6.2
Obesity 30.7 6.0 18.8
Pulmonary disease 31.7 4.0 12.5


Among residents (median age: 83 years), the case fatality rate was 33.7%. Chronic underling conditions included hypertension, cardiac disease, renal disease, diabetes mellitus, obesity, and pulmonary disease. The study demonstrated that once introduced in a long-term care facility, often by a care worker or a visitor, SARS-CoV-2 has the potential to spread rapidly and widely, with devastating consequences.

By mid-April 2020, more than 1300 LTC facilities in the US had identified infected patients (Cenziper 2020, CDC 200311). As most residents had one or more chronic underling conditions such as hypertension, cardiac disease, renal disease, diabetes mellitus, obesity and pulmonary disease, COVID-19 put them at very high risk for premature death.

Later studies found a high percentage of asymptomatic residents (43%) during the two weeks prior to testing (Graham 2020b); extraordinarily high seropositivity rates (72%; Graham 2020a); and a higher infection rate in residents (9.0%) than in staff (4.7%) (Marossy 2020).

A national survey covering 96% of all long-term care facilities in Italy found that in Lombardy, the epicenter of the epidemic, 53.4% of the 3045 residents who died between 1 February and 14 April were either diagnosed with COVID-19 or presented flu-like symptoms. Among the 661 residents who were hospitalized during the same period, 199 (30%) were found positive by a PCR test.

As soon as a single case is detected among residents of a nursing facility, it is recommended to test all residents, as many of them may still be asymptomatic. After an outbreak at a long-term care nursing facility, all residents, regardless of symptoms, underwent serial (approximately weekly) nasopharyngeal SARS-CoV-2 RT-PCR testing. Nineteen of 99 (19%) residents had positive test results for SARS-CoV-2 (Dora 2020). Fourteen of the 19 residents with COVID-19 were asymptomatic at the time of testing. Among these, eight developed symptoms 1-5 days after specimen collection and were later classified as pre-symptomatic.

Mortality in LTCs is almost always high. In a study from Ontario, Canada, the incidence of mortality was more than 13 times higher than the one seen in community-living adults older than 69 years during a similar period (Fisman 2020). In one UK investigation involving 394 residents and 70 staff in 4 nursing homes in central London, 26% of residents died over a two-month period (Graham 2020). It is estimated that residents in long-term care facilities contributed 30–60% of all COVID-19 deaths in many European countries (ECDC 2020; see also the statement to the press by Hans Henri P. Kluge, WHO Regional Director for Europe). Excess mortality data suggests that in several countries many deaths in long-term care facilities might have occurred in patients not tested for COVID-19, which are often not included in the official national COVID-19 mortality statistics.


Infection rates at home varied widely (between 11% and 19%) in three studies. One group noted that household contacts and those travelling with a COVID-19 case had a 6 to 7 times higher risk of infection than other close contacts, and that children were as likely to be infected as adults (Bi Q 2020). Another group found that the odds of infection among children and young people (< 20 years old) was only 0.26 times that among the elderly (≥ 60 years old) (Jing QL 2020). A third group calculated that the secondary attack rate in children was 4% compared to 17.1% in adults, and that the secondary attack rate in contacts who were spouses of index cases was 27.8% compared to 17.3% in other adult members in the households (Li W 2020). It has been objected that these transmission rates may be an underestimate if index cases were isolated outside of the home (Sun 2020). In yet another study, 32.4% (48 of 148) of household contacts of 35 index cases were infected (Wu J 2020).

In Spain, during the summer of 2020, social settings such as family gatherings or private parties accounted for 14% of cases (854/6208) in an analysis of 551 outbreaks. SARS-CoV-2 positive cases linked to leisure venues such as bars, restaurants, or clubs were even more frequent (NCOMG 2020) (see next paragraph).

Leisure venues (bars, clubs, choirs, karaoke, discos, etc.)

In Spain, an analysis of 551 outbreaks from mid-June to 2 August linked 1230 of 6208 cases (20%) to leisure venues such as bars, restaurants, or clubs (NCOMG 2020). Data from Japan showed that of a total of 61 COVID-19 clusters, 10 (16%) were in restaurants or bars; 7 (11%) in music-related events, such as live music concerts, chorus group rehearsals, and karaoke parties; 5 (8%) in gymnasiums; 2 (3%) in ceremonial functions (Furuse 2020). In South Korea, superspreading events in nightclubs in downtown Seoul were shown to have the potential to spark a local resurgence of cases (Kang 2020). In Hong Kong, an explosive summer outbreak was best explained by the sudden increase in social gatherings after the easing of public health control measures, especially gatherings at eateries (To 2020). College trips and summer camps represent another environment for efficient SARS-CoV-2 transmission. In one case, a spring break trip from Austin to Mexico resulted in 14 asymptomatic and 50 symptomatic cases (Lewis 2020). CDC reported an outbreak with 260 (44%) out of 597 attendees of an overnight summer camp in Georgia becoming infected in June 2020 (Szablewski 2020). The camp adopted most of CDC’s suggested preventive measures for Youth and Summer Camps but wearing cloth masks and opening windows and doors for increased ventilation in buildings were not implemented L.

Choirs, too, are places of efficient SARS-CoV-2 transmission. On 8 March 2020, the Amsterdam Mixed Choir gave a performance of Bach’s St John Passion in the city’s Concertgebouw Auditorium. Days later, the first singers developed symptoms and in the end 102 of 130 choristers were confirmed to have COVID-19. One 78-year-old choir member died, as did three partners of choir members; some singers required intensive care (The Guardian, 17 May). On 9 March, members of the Berlin Cathedral Choir met for their weekly rehearsal. Three weeks later, 32 out of 74 choir members were positive for SARS-CoV-2 (NDR 2020). All recovered. On 10 March 2020, 61 members of a Skagit County choir in Washington met for a 2,5-hour practice. A few weeks later, researchers reported that 32 confirmed and 20 probable secondary COVID-19 cases had occurred (attack rate = 53.3% to 86.7%); three patients were hospitalized, and two died. The authors conclude that transmission was likely facilitated by close proximity (within 6 feet) during practice and increased viral diffusion by the act of singing (Hamner 2020).

In an unintentional experiment, the German national team of amateur boxers proved that even 100% transmission rates can be achieved within days. In a training camp, some of the 18 athletes and 7 coaches and supervisors started having cold symptoms. Four days later, all 25 persons tested positive for SARS-CoV-2 (Anonymous 2020).

These data suggest that any noisy, closed and stagnant air environments (e.g., discos, pubs, birthday parties, restaurants, meat processing facilities, etc.) where people stand, sit or lie close together are ideal conditions for generating large SARS-CoV-2 outbreaks. If they need to shout for communication, the situation may become explosive.


As early as January 2020, SARS-CoV-2 was found to spread during workshops and company meetings (Böhmer 2020). A few weeks later, an outbreak of SARS-CoV-2 infection was reported from a call center where 94 out of 216 employees working on the same floor were infected, translating to an attack rate of 43.5% (Park SY 2020). Particularly instructive is the case of a scientific advisory board meeting held in Munich, Germany, at the end of February. Eight dermatologists and 6 scientists (among them the index patient) met in a conference room of about 70 m2 with a U-shaped set-up of tables separated by a central aisle > 1 meter wide. During the meeting that lasted 9,5 hours, refreshments were served in the room 4 times. In the evening, the participants had dinner in a nearby restaurant and shook hands for farewell, with a few short hugs (no kisses!). Finally, the index patient shared a taxi with three colleagues for about 45 min. Outcome: the index patient infected at least 11 of the 13 other participants. When isolated either in a hospital or at home these individuals infected an additional 14 persons (Hijnen 2020). In the presence of an infected individual, workplaces can be important amplifiers of local transmission.

In May 2020, outbreaks with hundreds of infected individuals were reported from meat-packing plants in Germany (DER SPIEGEL), the US (The Guardian) and France (Le Monde). Outbreaks in meat processing facilities were also reported from other countries. In March and April, 25.6% (929) of employees and 8.7% (210) of their contacts were diagnosed with COVID-19 in South Dakota, USA; two employees died (Steinberg 2020). The highest attack rates occurred among employees who worked < 6 feet (2 meters) from one another at the production line. Another study reported 16,233 COVID-19 cases and 86 COVID-19–related deaths among workers in 239 facilities (Waltenburg 2020). The percentage of workers with COVID-19 ranged from 3.1% to over 20% per facility (Waltenburg 2020). Promiscuity, noise, cold and humid conditions are currently favored as explanations for these unusual outbreaks. In Spain, the above-mentioned analysis of 551 outbreaks from mid-June to 2 August linked around 500 of 6208 cases (8%) to occupational settings, in particular, workers in the fruit and vegetable sector and workers at slaughterhouses or meat processing plants (360/6208 cases) (NCOMG 2020).


Schoolchildren usually play a major role in the spread of respiratory viruses, including influenza. However, while the SARS-CoV-2 virus has been detected in many children, they generally experience milder symptoms than adults, need intensive care less frequently and have a low death rate. An analysis of data from Canada, China, Italy, Japan, Singapore and South Korea found that susceptibility to infection in individuals under 20 years of age was approximately half that of adults aged over 20 years, and that clinical symptoms are manifest in 21% of infections in 10-to-19-year-olds, rising to 69% of infections in people aged over 70 years (Davis 2020).

The role of children in SARS-COV-2 transmission is still unclear. Several studies have suggested that children rarely transmit the infection. In a small COVID-19 cluster detected in the French Alps at the end of January, one person returning from China infected eleven other people, including a nine-year-old schoolboy. The researchers closely tracked and tested all contacts (Danis 2020). The boy had gone to school after showing COVID-19 symptoms and was estimated to have had more than sixty high-risk close contacts. No one was found positive to the coronavirus, though many had other respiratory infections. Also, no virus was found in the boy’s two siblings who were on the same Alpine vacation.

A study by the Institut Pasteur in April 2020 (before school closure) that included 510 primary school children concluded that “it appears that the children did not spread the infection to other students or to teachers or other staff at the schools”. Another study in 40 patients less that 16 years old in Geneva, Switzerland (Posfay-Barbe 2020) also concluded that unlike with other viral respiratory infections, children do not seem to be a major vector of SARS-CoV-2 transmission, with most pediatric cases described inside familial clusters and no documentation of child-to-child or child-to-adult transmission.“

However, a review of 14 published studies (Rajmil 2020) was less categorical, simply concluding that children are not transmitters to a greater extent than adults. A more recent metanalysis of published evidence (Viner 2020) states that there is insufficient evidence to conclude whether transmission of SARS-CoV-2 by children is lower than by adults.

CDC reported in September on twelve children who acquired COVID-19 in three different child-care facilities in Utah. It documented transmission from these children to at least 12 (26%) of 46 non-facility contacts and that transmission was observed from two of three children with confirmed, asymptomatic COVID-19. In addition, several studies have found that both symptomatic and asymptomatic children can shed the SARS-CoV-2 virus for several days or weeks after infection (Liu 2020, Han 2020). However, qualitative positive or negative findings for molecular detection of virus may not necessarily correlate with infectivity (DeBiasi 2020).

In the early autumn of 2020, how to re-open schools was a hot debate. In Taiwan authorities established general guidelines, including a combination of strategies such as active campus-based screening and access control; school-based screening and quarantine protocols; student and faculty quarantine when warranted; mobilization of administrative and health center staff; regulation of dormitories and cafeterias; and reinforcement of personal hygiene, environmental sanitation, and indoor air ventilation practices (Cheng SY 2020). Most European countries have decided to reopen schools, considering the possible increase in infections as being less damaging than the loss of education in schoolchildren.  At the time of this writing (31 October), the reopening of schools in European countries does not seem to have contributed substantially to the national epidemics. It can indeed be difficult to determine if children were infected at home, at school (by their peers or by their teachers) or outside during social or sport gatherings. In some school clusters, the index cases identified were teachers and/or parents (Torres 2020), so school prevention should focus on enforcing preventive measures and avoiding new cases among teachers. In any case, close monitoring of school clusters will provide much needed additional data that might help clarify the role of children of different ages in the spread of the virus, and whether schools can be considered hotspots or not of SARS-CoV-2 transmission.

Mass gatherings

Sports events

A football match played in Milan, Italy on 19 February 2020 has been described as “Game zero” or “a biological bomb”. The match was attended by 40,000 fans from Bergamo and 2500 from Valencia and was played just two days before the first positive case of COVID-19 was confirmed in Lombardy. Thirty-five percent of Valencia’s team members tested positive for the coronavirus a few weeks later, as did several Valencia fans. By mid-March, there were nearly 7,000 people in Bergamo who had tested positive for the coronavirus with more that 1,000 deaths, making Bergamo the most heavily hit province during the initial COVID-19 outbreak in Italy.

Other sport events have been implicated in the COVID-19 spread, including the match between Liverpool and Atletico Madrid, held at Anfield stadium on 11th March and  attended by 3,000 supporters from Madrid, the center of the pandemic in Spain, and the Cheltenham horseracing festival, with races attracting crowds of over 60,000 people (Sassano 2020). Most national and international large sport events, cancelled or postponed in the first half of 2020, have resumed during the summer months, though with closed doors or major limitations in the number of spectators. Large sports events including tens of thousands of spectators might not take place for several years.

Religious gatherings

Several mass gathering religious events have been associated with explosive outbreaks of COVID-19. As mentioned above, in April 2020, a total of 5212 coronavirus cases were related to an outbreak at the Shincheonji Church in South Korea, accounting for about 48.7% of all infections in the country at that time.

The annual gathering of the Christian Open Door Church held between 17 and 24 February in Mulhouse, France, was attended by about 2500 people and became the first significant cluster in France. After a parishioner and 18 family members tested positive on 1 March, a flurry of reported cases brought the existence of a cluster to light. According to an investigative report by France Info, more than 1,000 infected members from the rally in Mulhouse contributed to the start of the COVID-19 epidemic in France. Many diagnosed cases and deaths in France as well as Switzerland, Belgium and Germany were linked to this gathering.

Another report described 35 confirmed COVID-19 cases among 92 attendees at church events in Arkansas during March 6–11. The estimated attack rates ranged from 38% to 78% (James 2020). In Frankfurt, Germany, one of the first post-lockdown clusters started during a religious ceremony held on 10 May. As of 26 May, 112 individuals were confirmed to be infected with SARS-CoV-2 (Frankfurter Rundschau). May we suggest that going to church does not protect you from SARS-CoV-2?

Huge religious mass gatherings should probably be postponed. Gatherings that attract millions of pilgrims from many countries (with pilgrims typically > 50 years old and often suffering from chronic disease such as diabetes or cardiovascular disease [Mubarak 2020]) have clearly the potential to create giga-spreading events, saturating designated wards and ICU capacity within days. Reducing the number of pilgrims and excluding foreign pilgrims is therefore a wise decision (Khan 2020). Events attended by even more people, such as the Sabarimala annual 41-day long Hindu pilgrimage (average attendance: 25 million people) would need even more careful planning (Nayar 2020).

Closed and densely populated spaces


According to WHO, people deprived of their liberty, such as people in prisons and other places of detention, are more vulnerable to COVID-19 outbreaks (WHO 200315). People in prison are forced to live in close proximity and thus may act as a source of infection, amplification and spread of infectious diseases within and beyond prisons. The global prison population is estimated at 11 million and prisons are in no way “equipped” to deal with COVID-19 (Burki 2020).

In US prisons, COVID-19 attack rates are high. By June 6, 2020, there had been 42,107 cases and 510 deaths among 1.3 million prisoners (Saloner 2020, Wallace 2020). Among 98 incarcerated and detained persons in Louisiana who were quarantined because of virus exposure, 71 (72%) had SARS-CoV-2 infection identified through serial testing, among them 45% without any symptoms at the time of testing (Njuguna 2020). In July 2020, more than one-third of the inmates and staff (1600 people) in San Quentin Prison tested positive (Maxmen 2020). Six had died. Still in July 2020, the rate of COVID-19 among incarcerated individuals in Massachusetts was nearly 3 times that of the general population and 5 times the US rate (Jiménez 2020).

Homeless shelters

Testing in 1192 residents and 313 staff members in 19 homeless shelters from 4 US cities (see table online), initially triggered by the identification of a COVID-19 cluster, found infection rates of up to 66% (Mosites 2020).

In another report from Boston, Massachusetts, 147/408 (36%) homeless shelter residents were positive. Of note, 88% had no fever or other symptoms at the time of diagnosis (Baggett 2020).

In yet another study of 14 homeless shelters in King County, Washington, researchers divided the number of positive cases by the total number of participant encounters, regardless of symptoms. Among 1434 encounters, 29 (2%) cases of SARS-CoV-2 infection were detected across 5 shelters. Eighty-six percent of persons with positive test results slept in a communal space rather than in a private or shared room (Rogers 2020).

Cruise ships, aircraft carriers etc.

Cruise ships carry many people in confined spaces. On 3 February 2020, 10 cases of COVID-19 were reported on the Diamond Princess cruise ship. Within 24 hours, all sick passengers were isolated and removed from the ship and the rest of the passengers quarantined on board. Over time, more than 700 of the 3,700 passengers and crew tested positive (around 20%). One study suggested that without any intervention 2920 individuals out of the 3700 (79%) would have been infected (Rocklov 2020). The study also showed that an early evacuation of all passengers on 3 February would have been associated with only 76 infected. For cruise ships, SARS-CoV-2 may spell disaster – carrying village-loads of people from one place to another may not be a viable business model for years to come.

Big navy vessels such as aircraft carriers can become floating petri dishes for emerging viral respiratory diseases. Already in 1996, an outbreak of influenza A (H3N2) occurred aboard a navy ship. At least 42% of the crew became ill within few days, although 95% had been appropriately vaccinated (Earhart 2001). Since the beginning of the year, several outbreaks of COVID-19 on military ships have been reported, facilitated by the small enclosed areas of work and the lack of private quarters for the crew. The largest outbreaks have been reported on the USS Theodore Roosevelt and the French aircraft carrier Charles de Gaulle. During the Theodore Roosevelt outbreak in late March, around 600 sailors out of a crew of 4800 were infected with SARS-CoV-2 (see also the March 30 entry of the Timeline); around 20% reported no symptoms and one sailor died. (USNI News). Preventive measures, such as using face-coverings and observing social distancing, reduced risk of infection: among 382 service members, those who reported taking preventive measures had a lower infection rate than did those who did not report taking these measures (e.g., wearing a face-covering, 56% versus 81%; avoiding common areas, 54% versus 68%; and observing social distancing, 55% versus 70%, respectively) (Payne 2020).

On the French aircraft carrier Charles-de-Gaulle, a massive epidemic was confirmed on 17 April. Among the 1760 sailors, 1046 (59%) were positive for SARS-CoV-2, 500 (28%) presented symptoms, 24 (1.3%) sailors were hospitalized, 8 required oxygen therapy and one was admitted in intensive care. Smaller clusters have also been reported on 5 other US military vessels, and in one each from France, Taiwan, and Holland. However, given usual security policies and communication restrictions of national armies and navies, it is possible that other unreported cluster of cases and even deaths might have occurred.

Special Aspects of the Pandemic

The COVID-19 pandemic has highlighted a number of specific aspects and lessons learned from different countries that should be kept in mind during the management of future pandemics (by coronaviruses, influenza viruses or by as yet unknown viruses):

  • First outbreak (China)
  • Surprise or unpreparedness (Italy)
  • Unwillingness to prepare (UK, USA, Brazil)
  • Partial preparedness (France)
  • Preparedness (Germany)
  • Herd immunity? (Sweden)
  • Deferred beginning (South America)
  • Splendid isolation (New Zealand, Australia)
  • Unknown (?) outcome (Africa)

First outbreak (China)

China was caught by surprise by the COVID-19 outbreak – as any other nation would have been – but “thanks” to the SARS outbreak in 2003 (Kamps-Hoffmann 2003), was prepared for it. At first, the epidemic spread within Wuhan and Hubei Province (December 2019, Li Q 2020) and then nationwide to all provinces in January 2020, favored by travelers departing from Wuhan before the Chinese Spring Festival (Zhong 2020, Jia JS 2020). However, within 3 weeks from the identification of the new virus, the government ordered the lockdown of more than 50 million people in Wuhan and the surrounding province of Hubei, as well as travel restrictions for hundreds of millions of Chinese citizens. This astonishing first in human history achieved what even specialists didn’t dare dream: curbing an epidemic caused by a highly contagious virus (Lau 2020).

As early as four weeks after the Wuhan lockdown, there was evidence that strict containment measures were capable of curbing a SARS-CoV-2 epidemic as shown in Figure 1 (page 35). The lesson from China: it is possible to lockdown entire provinces or countries and lockdown works. Some authorities in the Western Hemisphere followed the example of China (Italy, for example, ordered a lockdown as early as 18 days after the diagnosis of the first autochthonous case), other governments did not. It cannot be overemphasized that China has basically managed to control the spread of SARS-CoV-2 since March. How was that possible (Burki 2020)?

Preparedness (Taiwan, Vietnam, Japan)

On 7 June, Taiwan (24 million people with a population density of 650/km2) had reported only 443 cases and 7 deaths. Most SARS-CoV-2 infections were not autochthonous. As of 6 April 2020, 321 cases were imported by Taiwanese citizens who had travelled once or more to 37 countries for tourism, business, work, or study (Liu JY 2020). From the beginning, Taiwan drew on its SARS experience to focus on protecting health care workers’ safety and strengthening pandemic response (Schwartz 2020 + The Guardian, 13 March 2020). An early study suggested that identifying and isolating symptomatic patients alone might not suffice to contain the epidemic and recommended more generalized measures such as social distancing (Cheng HY 2020). Big data analytics were used in containing the epidemic. On one occasion, authorities offered self-monitoring and self-quarantine to 627,386 persons who potentially had contact with the more than 3,000 passengers of a cruise ship. These passengers had disembarked at Keelung Harbor in Taiwan for a 1-day tour five days before the COVID-19 outbreak on the Diamond Princess cruise ship on February 5, 2020 (Chen CM 2020).

Vietnam, too, did remarkably well. One hundred days after the first SARS-CoV-2 case was reported in Vietnam on January 23rd, 270 cases were confirmed, with no deaths. Although there was a high proportion of asymptomatic and imported cases as well as evidence for substantial pre-symptomatic transmission, Vietnam controlled SARS-CoV-2 spread through the early introduction of mass communication, meticulous contact-tracing with strict quarantine, and international travel restrictions (Pham QT 2020).

Finally, in Japan, public adherence to the rules, along with cluster tracing and a ban on mass gatherings, seem to have helped in bringing the outbreak under control. Where widespread mask use and hygiene is a normal part of etiquette, combatting SARS-CoV-2 is easier (Looi 2020).

Experiences from these countries show that effective testing and contact tracing, combined with physical distancing measures, can keep the pandemic at bay and an economy open. Health is the key to wealth.

Surprise or unpreparedness (Italy)

In Italy and France, SARS-CoV-2 was circulating as early as January among asymptomatic or pauci-symptomatic people (Cereda 2020, Gámbaro 2020). Italy was the first European country struck by the pandemic. Complete genome analysis of SARS-CoV-2 isolates suggests that the virus was introduced on multiple occasions (Giovanetti 2020). Although the first local case was diagnosed only on 20 February, the force of the outbreak suggests that the virus had been circulating for weeks, possibly as early as 1 January (Cereda 2020).

However, it was not straightforward to decipher the subtle signs of coming events, in Italy as elsewhere. During the yearly flu season, COVID-19 deaths in elderly people could easily be interpreted as flu deaths. And the rapid spread of SARS-CoV-2 among the most active social age group – young people crowded in bars, restaurants and discos –would not have caused visible life-threatening symptoms. Before being detected, the epidemic had plenty of time (at least a month) to grow.

One additional possible reason for the delay in recognizing the encroaching epidemic in Italy might have been the Italian ‘suspected case definition for COVID-19’. It included (like the suspected case definitions recommended at that time by WHO) the mandatory epidemiological criteria of ‘history of travel to  China or in contact with a person from China’ before requesting a PCR test. A strict application of this case definition discouraged testing suspected pneumonia cases where the link with China was not clear (which would eventually happen everywhere after the first asymptomatic infections). The anesthesiologist who eventually requested the PCR test for Mattia, the Italian patient #1, did it “under her own responsibility since not in line with MOH guidelines”.

It is as yet unclear why the epidemic took such a dramatic turn in the northern part of Italy, especially in Lombardy (Gedi Visual 2020), while other areas, especially the southern provinces, were relative spared. Overdispersion might be an explanation (see above). Of note, healthcare in Italy is run regionally and for a long time, the Lombardy Region has favored the development of a mostly private and hospital-centered system, with great facilities but poor community-based services. This meant that COVID-19 patients were quickly run to the hospital, even with minor symptoms, resulting in overcrowded emergency services and major nosocomial spread. A more decentralized and community-based system like in the Veneto Region (plus maybe a bit of luck) could have greatly reduced the mortality from COVID-19 in Lombardy. In addition, Italy had not updated nor implemented the 2006 national pandemic preparedness plan (https://www.saluteinternazionale.info/2020/04/cera-una-volta-il-piano-pandemico). The lack of preparedness and the overlap of responsibilities hampered considerably the initial coordination of the national response between the regions and the central government.

Unwillingness to prepare, or denial (UK, Iran, USA, Brazil)

In the UK, clumsy political maneuvering delayed the start of effective lockdown measures by a week or more. As the epidemic doubles in size about every 7 days (Li 2020), around 50% and 75% of all deaths might have been prevented had lockdown or social distancing measures been ordered one or two weeks earlier, respectively. Early data from Ireland and the United Kingdom seem to confirm this assumption. Each day of delay increased mortality risk by 5 to 6% (Yehya 2020). The consequences were dramatic (Stoke 2020, Maxmen 2020).

Like in Iran, where the regime covered up news of the coronavirus for three days to avoid impacting on the turnout at parliamentary elections on 21 February, domestic politics (or paranoia; BMJ, 6 March 2020) influenced the epidemic response in the US. Scientific advice from CDC and other national public health institutions was ignored (The Lancet 2020). The US is the country with the highest number of cases and deaths. Without this unprecedented vacuum in leadership (NEJM Editors 2020), most of these deaths would have been prevented.

Brazil, which is also not an example of good governance performance, has become the country with the second highest number of deaths in the world.

Partial preparedness (France)

France was partially prepared. During the first national outbreak near Mulhouse, hospitals were overwhelmed. Despite the updated and well-structured pandemic plan (https://www.gouvernement.fr/risques/plan-pandemie-grippale), all over the country protective equipment was in short supply; in particular, face masks were sorely lacking after a decision by the Hollande government to greatly reduce the stocks of 1.7 billion protective masks (surgical and FFP2) available in 2009 and considered too expensive to only 145 million surgical masks in 2020 (“We are not going to manage mask stocks, it is expensive, because we have to destroy them every five years. Nous n’allons pas gérer des stocks de masques, c’est coûteux, parce qu’il faut les détruire tous les cinq ans.”)  (Le Monde 200506).

However, France, thanks to Italy, had an important advantage: time. It had several weeks to learn from the events in Lombardy. When, on the weekend of 21 March, virtually from one day to the next, patients started pouring into the hospitals of the Greater Paris Region, the number of available intensive care unit beds had already increased from 1400 to 2,000 during the preceding week. Furthermore, two years before, in a simulation of a major terrorist attack, France had tested the use of a high-speed TGV train for transporting casualties. At the height of the COVID epidemic, more than 500 patients were evacuated from epidemic hotspots like Alsace and the Greater Paris area to regions with fewer COVID-19 cases. Specially adapted high-speed trains as well as aircraft were employed, transporting patients as far away as Brittany and the Bordeaux area in the South-West, 600 km from Paris and 1000 km from Mulhouse. The French management of ICU beds was a huge logistical success.

Good virologists, huge lab network, family doctors (Germany)

Germany’s fatality rate is lower than in other countries. It is assumed that the main reason for this difference is simply testing. While other countries were conducting a limited number of tests in older patients with severe disease, Germany was doing many more tests that included milder cases in younger people (Stafford 2020). The more people with no or mild symptoms you test and isolate, the lower the fatality rate and the spread of infection.

Furthermore, in Germany’s public health system, SARS-CoV-2 testing is not restricted to a central laboratory as in many other nations but can be conducted at quality-controlled laboratories throughout the country. Thanks to reliable PCR methods that had been developed by the end of January from the Drosten group at Berlin’s Charité (Corman 2020), within a few weeks the overall capacity reached half a million PCR tests a week. The same low fatality rate is seen in South Korea, another country with high testing rates.

Finally, another important reason for the low mortality in Germany might be the age distribution. During the first weeks of the epidemic, most people became infected during carnival sessions or ski holidays. The majority were younger than 50 years of age. Mortality in this age group is markedly lower than in older people.

As a result of these first-wave distinctive features, the case-fatality rate (CFR) of COVID was 0,7% in Germany, compared with CFRs as high as 9,3% and 7,4% in Italy and the Netherlands, respectively (Sudharsanan 2020, Fisman 2020). Age distribution of cases may explain as much as 66% of the variation of SARS-CoV-2 cases across countries (Sudharsanan 2020).

Herd immunity? Not yet!  (Sweden)

Sweden has never really imposed a lockdown, counting on the population to adopt individual social distancing and other protective measures to curb the transmission of SARS-CoV-2. The price was high (Habib 2020). In October 2020, Sweden had a death rate 10 times higher than Norway and five times higher than Denmark, with most deaths occurring in care homes and immigrant communities. Still worse, Sweden didn’t benefit economically of its no-lockdown approach as its economic performance contracted at a similar rate as countries in the rest of Europe (Financial Times, 10 May 2020).

Will the autumn and winter reduce the mortality gap between Sweden and Norway and Denmark? Will Sweden, after accepting many deaths in spring, see fewer of them in the future? Will those who died early reduce the number of deaths seen later? Will a (still low!) level of community immunity help slow down the epidemic in winter? In any case, evaluations of cell phone data show that Swedes traveled much less during the summer than, for example, Norwegians or Danes, so they may have imported less infections from summer vacation hotspots. For a detailed discussion of herd immunity, see Randolph 2020.

Deferred beginning, then major impact (South America)

The first case of COVID-19 in Latin America was reported on 26 February in Brazil and by early April all countries had reported at least one imported case. However, in the initial months of 2020, the number of cases was comparatively low in South America compared to Europe or Asia (Haider 2020). As a matter of fact, the local epidemics took off roughly 4 weeks later than in Europe (see www.worldometers.info/coronavirus).

However, the epidemic accelerated during the month of May when South America become the epicenter of the coronavirus pandemic according to WHO. In September, Latin America, home to around 8% of the world’s population, accounted for over a quarter of all confirmed COVID-19 cases and nearly a third of all related deaths. There is, however, wide variation between countries, with Brazil and Mexico having some of the worst epidemics in the world, while Uruguay infection rates are comparable to the best performing countries in Asia or Europe (Taylor 2020).

According to Marcos Espinal and colleagues from WHO, there are several factors in Latin America that make this pandemic more difficult to manage: inequality, belts of poverty surrounding big cities, informal economies, and difficult areas of access. Here, as elsewhere, leadership and sound public health policies made a difference. Both Brazil’s and Mexico’s presidents have been widely criticized for playing down the threat of COVID-19, not taking action to slow its spread, and suggesting alternative ineffective ways of protection (for example, the use of traditional scarves (?) instead of face masks).

However, other countries have performed much better, managing to keep infections low. For example, Cuba and Costa Rica have enforced strict testing, isolation and quarantine measures. The most successful country so far has been Uruguay that managed, though a mix of effective testing, contact tracing, isolation and quarantine, to keep infection rates very low without generalized lockdowns. The President simply asked, rather than ordered, people to stay home for their own well-being and that of fellow citizens (Taylor 2020).

Splendid isolation (New Zealand, Australia)

Australia, New Zealand, French Polynesia, Fiji, New Caledonia and Papua New Guinea and Oceania are among the least hit areas in the world. Geographically isolated islands or island states should be the ideal candidates for elimination trials. However, even New Zealand, which viewed itself in the post-elimination stage and where public life had returned to near normal (Baker 2020), was suddenly called back into COVID-19 reality when new cases were discovered in August 2020.

In Australia, transmission was initially driven by multiple SARS-CoV-2 importations by returned international travelers which accounted for over half of locally acquired cases (Seemann 2020). However, on 20 June, the State of Victoria reported a spike in community transmitted cases, apparently following lax implementation of quarantine measures, that resulted in a large outbreak with more than 20,000 cases and 800 deaths and the imposition of strict lockdown measures in the State and a night curfew in Melbourne. An easing of restrictions only started mid-September, following a major decrease in the number of new cases.

Both Australia and New Zealand have considered a strategy of COVID-19 elimination, i.e. the absence of sustained endemic community transmission in the country. The recent outbreaks have raised the question of whether elimination is a reasonable goal (Hewyood 2020). The elimination of any infectious disease is an ambitious objective, requiring strong public health measures and substantial resources. In principle, a zero-case scenario of not less than three months would be the condition for declaring a state or country SARS-CoV-2-free. Then, strict travel and border restrictions and quarantine measures must be implemented over a prolonged period, since the virus continues to spread around the world. It looks like international travel to New Zealand and Australia may continue to be banned for quite some time.

Africa: The unknown (?) outcome

The transmissibility of SARS-CoV-2, combined with the scarcity of crucial health equipment and facilities and the challenges of implementing widespread case isolation (Wells 2020), was supposed to result in a devastating impact of COVID-19 on African countries. These predictions have not materialized. (To put the area into focus, remember that Europe without Russia has a surface of roughly 6 million km2, Africa has 30 million km2. That should explain by itself that the burden and outcomes associated with COVID-19 in Africa shows substantial variations across African countries [Twahirwa 2020]. There is no ‘one’ Africa.)

Some official figures are certainly underestimates, voluntary or not, due to regional difficulties in reporting. In some cities, such as Kano, Nigeria, major outbreaks may already be underway. The New York Times reported on 17 May, “so many doctors and nurses have been infected with SARS-CoV-2 that few hospitals are now accepting patients”. Gravediggers are working overtime. In Mogadishu, Somalia, officials say burials had tripled, according to the same report. In Tanzania, the US embassy has warned of the risk of “exponential growth” of COVID-19 cases in the country, adding that hospitals were “overwhelmed” (The Guardian, 19 May).

However, there has been no COVID-19 explosion in Africa. Has time come to hypothesize an “African exception”? It is probably too early to say but demographics might explain in part the difference. In the Democratic Republic of the Congo and Malawi, for instance, only 2-3% of the population is older than 65 years (Kalk 2020), in sharp contrast to Europe at 20,5% or Lombardy at 26%. If 65-year-old SARS-CoV-2 infected individuals are 100 times more likely to die from COVID-19 than a 25-year-old, we should expect two different epidemics. Simply, the age pyramid might make the difference.

The SARS-CoV-2 pandemic: Past and Future

Natural course of a pandemic

The COVID-19 epidemic started in Wuhan, in Hubei province, China, and spread within 30 days from Hubei to the rest of mainland China, to neighboring countries (in particular, South Korea, Hong Kong and Singapore) and west to Iran, Europe and the Americas. The first huge outbreaks occurred in regions with cold winters (Wuhan, Iran, Northern Italy, the Alsace region in France).

Fifty years ago, the course of the COVID-19 pandemic would have been different, with slower global spread but high burden due to limited diagnostic and therapeutic capacities and no option of nation-wide lockdowns (see also a report of the influenza pandemics in 1957 and 1968: Honigsbaum 2020). According to one (controversial) simulation, in the absence of interventions and with a mortality rate of around 0.5%, without interventions COVID-19 would have resulted in 7.0 billion infections and 40 million deaths globally during the first year (Patrick 2020). The peak in mortality (daily deaths) would have been observed approximately 3 months after the beginning of local epidemics. Another model predicted that 80% of the US population (around 260 million people) would have contracted the disease. Of those, 2.2 million Americans would have died, including 4% to 8% of those over age 70 (Ferguson 2020). In Germany alone, the SARS-CoV-2 pandemic could have resulted in 730,000 deaths (Barbarossa 2020) and in 500,000 deaths each in France, Italy, Spain and the UK.

The 2020 Lockdowns

Fortunately, for now, the world has been spared from a freely circulating SARS-CoV-2. If humanity can change the climate, why shouldn’t we be able to change the course of a pandemic? Although economists warned that unemployment could surpass the levels reached during the Great Depression in the 1930s, at first, almost all governments considered saving hundreds of thousands lives more important than avoiding a massive economic recession. First in China, six weeks later in Italy and another a week later in most Western European countries, more recently in the US and in many other countries in the world, unprecedented experiments of gigantic dimensions were started: ordering entire regions or the whole nation to lockdown. By the first week of April, 4 billion people worldwide were under some form of lockdown — more than half of the world’s population. Lockdowns in Europe were generally less strict than in China, allowing the continuation of essential services and industries and the circulation of people when justified.

People were generally compliant to mandatory stay-at-home orders, even in the US. Based on location data from mobile devices, in 97.6% of US counties these orders were associated with decreased median population movement (Moreland 2020). Lockdowns were generally also well accepted. During the week of May 5–12, 2020, a survey among 2402 adults In New York City and Los Angeles found widespread support of stay-at-home orders and non-essential business closures and a high degree of adherence to COVID-19 mitigation guidelines (Czeisler 2020). In New York City, SARS-CoV-2 prevalence varied substantially between boroughs between 22 March and 3 May 2020 (for example, Manhattan: 11,3%; South Queens: 26,0%). These differences in prevalence correlate with antecedent reductions in commuting-style mobility between the boroughs. Prevalence was lowest in boroughs with the greatest reductions in morning movements out of and evening movements into the borough (Kissler 2020).

Lockdowns were also successful in slowing down the pandemic. According to one study, between 12 and 15 million individuals in Europe had been infected with SARS-CoV-2 by May 4th, representing between 3.2% and 4.0% of the population (Flaxman June 2020). Projected percentages of the total population infected were for Austria 0,76%, Belgium 8,0%, Denmark 1,0%, France 3,4%, Germany 0,85%, Italy 4,6%, Norway 0,46%, Spain 5,5%, Sweden 3,7%, Switzerland 1,9% and the UK 5,1%. In South America, lockdowns were successful, too, although they worked best among the wealthy and less well among the less wealthy who had to choose between the risk of dying from COVID or dying from hunger.

There is no real pandemic in Africa, a never-ending wave in the Americas, and now a second wave in Europe. The worst may be yet to come (The Lancet 2020) with more people dying and every death leaving 10 more people mourning a grandparent, parent, sibling, spouse, or child (Verdery 2020). Will the winter SARS-CoV-2 pandemic follow the scenario of the 1918 influenza pandemic (Horton 2020)?

In the French Bouches-du-Rhône department which includes Marseille, the first signs of the second wave were detected in wastewater on July 13[1]. Three weeks later, the first post-lockdown rise in new SARS-CoV-2 infections was seen in young adults 20 to 29 years old, and again a few weeks later, infection rates increased in older age groups. In Spain (NCOMG 2020), Switzerland (see Figure 1) and other European countries, the second wave looked equally triggered mostly by transmission among young adults in leisure venues such as bars, restaurants, discos or clubs during the summer 2020.


Figure 1. Weekly positive SARS-CoV-2 tests in Switzerland by age group (August 3 through October 5).Source: SRF, So entwickeln sich die Corona-Zahlen in der Schweiz (https://www.srf.ch/news/schweiz/coronavirus-so-entwickeln-sich-die-corona-zahlen-in-der-schweiz; accessed 12 October 2020).


Let’s briefly discuss

  • Measuring the epidemic
  • Herd immunity: Not yet
  • Vaccines: Be patient
  • ‘Variolation’ – Finding of the year?
  • Protection: People at risk
  • Prevention: Testing, tracing, isolating
  • Curfews

Measuring the epidemic

In the current second European wave, the number of newly diagnosed SARS-CoV-2 cases and the positive rate of PCR tests are certainly useful markers for the evolution of national epidemics; however, the number of hospitalizations and, most importantly, the number of new admissions to intensive care units (ICU) and deaths are the crucial figures in terms of disease burden (Figure 2 and 3).

Note that all these markers have limitations. For example, the number of positive cases identified are related to the number of tests performed and testing strategies. Hospital admissions also have limitations (hospital admission criteria may change from place to place and be modified over time) and can be influenced by, for example, the availability of quality home-based care or the collapse of an overburdened health system. In addition, many governments are not publicly providing numbers of daily hospital admissions and discharges (Garcia-Basteiro 2020).

In anticipating local epidemics, politicians should prepare for the worst, at least until spring 2021. An important feature of this second wave of infections is its widespread nature, as opposed to earlier, more localized outbreaks (e.g. northern Italy, Madrid, Spain or Mulhouse, France.) More populated and better-connected municipalities were generally affected earlier by the SARS-CoV-2 epidemic, and less populated municipalities at a later stage of the epidemic (de Souza 2020). However, relaxation of mitigation measures leading to a resumption of “normal” diffusion may initially appear to have few negative effects, only to lead to deadly outbreaks weeks or months later (Thomas 2020). Public health messaging may need to stress that apparent lulls in disease progress are not necessarily indicators that the threat has subsided, and that areas “passed over” by past outbreaks could be impacted at any time.

Herd immunity: Not yet

Herd immunity, the notion introduced to a wider public by a foolish politician, may not be on the agenda for a long time. Herd immunity, also known as indirect protection, community immunity, or community protection, refers to the protection of susceptible individuals against an infection when a sufficiently large proportion of immune individuals exist in a population (Omer 2020). As for now, not a single country is anywhere close to reaching herd immunity. Even in past hotspots like Wuhan, the prevalence of SARS-CoV-2 IgG positivity was 9.6% among 1021 people applying for a permission (the SARS‐CoV‐2 nucleic acid test needed to be negative) (Wu X 2020). A French study projected 2.8 million or 4.4% (range: 2.8–7.2) prevalence of infections in France. In Los Angeles, the prevalence of antibodies was 4.65% (Sood 2020). (And even this low number may be biased because symptomatic persons may have been more likely to participate.) A nationwide coronavirus antibody study in Spain showed that about 5% of the population had contracted the virus. These infection rates are clearly insufficient to avoid a second wave of a SARS-CoV-2 epidemic (Salje 2020). Achieving herd immunity without overwhelming hospital capacity would require an unlikely balancing of multiple poorly defined forces (Brett 2020).

Vaccines: Be patient

Some fools – politicians and experts alike – announced efficient and safe vaccines two months before Christmas 2020. Reality will see such thoroughly tested vaccines delivered to the first groups of vaccinees (i.e., health care workers) way into 2021, and nobody should expect vaccines to have a noticeable impact on the SARS-CoV-2 pandemic before the end of next year. In the meantime, people will need to be patient and look for alternative ways of protection.

‘Variolation’ – Finding of the year?

Reducing the viral SARS-CoV-2 inoculum might not only reduce the probability of infection but also favor an asymptomatic infection while still generating immunity. This suggestion (Bielecki 2020) was later further developed (Ghandi 2020; see also the comments to the paper by Rasmussen 2020, Brosseau 2020): if facial masking may help in reducing the size of the viral inoculum, universal facial masking might ensure that a greater proportion of new infections are asymptomatic. If universal masking could be proved to be a form of ‘variolation’ (inoculation), it would be an additional argument in favor of strict mask wearing.

Protecting people at risk

Protecting those at higher risk of SARS-CoV-2 infection, for example the elderly and healthcare workers (Nguyen 2020), will continue to be the highest priority over the coming months. Specific population groups might be at higher risk too. In the UK and the US, Black, Asian, and minority ethnic health care workers are at especially high risk of SARS-CoV-2 infection, with at least a fivefold (!) increased risk of COVID-19 compared with the non-Hispanic white general community. The infection rate is also higher in the most poor and vulnerable areas, thus emphasizing existing inequalities (Grasso 2020: COVID of the rich, COVID of the poor).

In a cross-sectional study of 396 pregnant New York City residents, large household membership, household crowding, and low socioeconomic status were associated with a 2-3 fold higher risk of infection (Emeruwa 2020). American Indian and Alaska Native (AI/AN) persons, too, appear to be disproportionately affected by the COVID-19 pandemic. In one study, the overall COVID-19 incidence among AI/AN persons was 3.5 times that among white persons (594 per 100,000 AI/AN population compared with 169 per 100,000 white population) (Hatcher 2020).

Prevention: Testing, tracing, isolating

Screening, case investigation, contact tracing, and isolation of infected persons is paramount during periods of community transmission. In a random sample of 350 adults aged ≥ 18 years who had positive RT-PCR in outpatient and inpatient settings at 11 US academic medical centers, only 46% were aware of recent close contact with someone with COVID-19, most commonly a family member (45%) or a work colleague (34%) (Tenforde 2020).

Testing presents numerous challenges (Clapham 2020), but the more people you test for SARS-CoV-2, the better. In a worldwide cross-sectional study (Liang LL 2020), COVID-19 mortality was

  • Negatively associated with
    • Test number per 100 people
    • Government effectiveness score
    • Number of hospital beds
  • Positively associated with
    • Proportion of population aged 65 or older
    • Transport infrastructure low quality score

Testing was a major limiting factor in assessing the extent of SARS-CoV-2 transmission during its initial spread into the US (Perkins 2020). After a national emergency was declared, fewer than 10% of locally acquired, symptomatic infections in the US were detected over a period of a month. This gap in surveillance during a critical phase of the epidemic resulted in a large, undetected reservoir of infections by early March. Other countries did better. Citywide mass nucleic acid testing of SARS-CoV-2 for all citizens is possible as shown in Wuhan city (14 May to 1 June 2020). The results are sometimes meager, revealing just 6 persons who test positive for SARS-CoV-2 (0,006% of 107,662 residents around the Huanan Seafood Market), but are able to suffocate a nascent epidemic (Jingwen L 2020).

It is important to recognize that despite aggressive efforts by health departments, many COVID-19 patients do not report contacts, and many contacts cannot be reached (Lash 2020). Staff members in North Carolina, US have investigated 5514 (77%) persons with COVID-19 in Mecklenburg County and 584 (99%) in Randolph Counties: during periods of high COVID-19 incidence, 48% and 35% of patients reported no contacts, and 25% and 48% of contacts were not reached. Median interval from index patient specimen collection to contact notification was 6 days. Some countries are obviously better prepared for mass testing than others and capable of performing 9 million tests in 5 days after the detection of 12 cases in a previously COVID-free area (Vidal Liy 2020 + BBC).


Lockdowns are effective but frighteningly costly. The spring lockdown cost most countries around 10% of their PIB with unforeseeable economic, political and also health consequences; in exchange, they can “flatten the curve” and did succeed in keeping seroprevalence rates low, somewhere between 1% and 10%. General lockdowns are clearly not a viable model for the future.

Might curfews be a less costly alternative, both economically and socially? In French Guiana, an overseas départment, a combination of curfews and targeted lockdowns in June and July 2020 was sufficient to avoid saturation of hospitals. On weekdays, residents were first ordered to stay at home at 11 p.m., then at 9 p.m., later at 7 p.m., and finally at 5 p.m. On weekends, everyone had to stay at home from 1 p.m. on Saturday (Andronico 2020). Whether curfews can be successfully adapted to other areas than French Guiana, is not known. French Guiana is a young territory with a median age of 25 years and the risk of hospitalization following infection was only 30% that of France. About 20% of the population had been infected with SARS-CoV-2 by July 2020 (Andronico 2020). Following Belgium and Germany, France has just implemented now its night curfew in Paris and a few other major cities. Be prepared to see more curfews orders over the coming six months.


How long will SARS-CoV-2 stay with us? How long will it be before we return to pre-COVID-19 ‘normalcy’? For how long will a combination of physical distancing, enhanced testing, quarantine, and contact tracing be needed? Historical evidence from prior influenza pandemics indicates that pandemics tend to come in waves over the first 2–5 years as population immunity builds-up (naturally or through vaccination) and that this is the most likely trajectory for SARS-CoV-2 (Petersen 2020). Even vaccines are not expected to have a substantial impact on the pandemic before 2022, if ever. In the meantime, classical infection control measures are the only way to reduce the number of infections and avoid healthcare systems from breaking down, leaving patients with other morbidities – common emergencies and surgery, cancer treatment, management of patients with chronic diseases – stranded and abandoned in a medical no-man’s land.

Summer of 2020 has shown that the post-lockdown epidemic dynamic was driven by younger adults with gradual ‘spill-over’ into older age groups. However, the formula ‘young adults -> parents -> grandparents -> death’ shall not be used as a simplistic model for the European second wave. SARS-CoV-2 is introduced and spread in communities via all conceivable routes. It is therefore important to define the behaviors than can minimize the risk of local lockdowns and economic hardship.

In situations of intense SARS-CoV-2 community transmission, the prevention triad is simple:

  1. Stop people from meeting each other in large gatherings.
  2. If they MUST meet, have them wear face masks.
  3. In any case reduce the time infected or suspected infectious people meet any other people at all: test as much as possible, isolate cases quickly and track the close contacts.

In transmission hotspots, restrictive social-distancing measures will need to be combined with widespread testing and contact tracing to slow down the ongoing pandemic (Giordano 2020 + less realistic, Peto 2020). People should concentrate on the essential activities of providing food and shelter as well as continuing their job, school and university activities. All ‘après-work’ and ‘après-school’ activities should be reduced to a minimum (no evening bars, no night life). In such social slowdowns, people will need to avoid prolonged meetings with people from outside their inner-core “friends-and-family-bubble”, in particular social events which bring people from many different families together (marriages, funerals, religious events). Even inside the inner-core “friends-and-family-bubble”, meetings should be restricted to a handful of people. Economically, a social slowdown implies the temporary closure of places where foreigners, strangers or simply unacquainted people meet: discos, amusement parks, bars, restaurants, brothels and many more. In a situation of intense SARS-CoV-2 community transmission, strangers must not come into contact.

Coronaviruses have come a long way (Weiss 2020) and will stay with us for a long time. Questions abound: When will we move freely around the world as we did before? Will air traffic return to pre-COVID-19 levels by 2024 or only later? Will we be inclined to plan vacations nearer to home rather than on the other side of the globe? Will we wear face masks for years? Will there be any nightlife event with densely packed people dancing and shouting and drinking in any city in the world anytime soon? Nobody knows. We only know that old and obese countries are hard hit and young and slim countries are relatively spared.

The French have an exquisitely precise formula to express unwillingness for living in a world you do not recognize: “Un monde de con!” Fortunately, we will be able to walk out of this monde de con thanks to a scientific community which is larger, stronger, and faster than at any time in history. (BTW, should politicians who are skeptical of science be ousted out of office? Yes, please! It is about time!) As of today, we do not know how long-lasting, how intense, and how deadly this pandemic will be. We are walking on moving ground and, in the coming months and years, we will need to be flexible, resilient, and inventive, looking for and finding solutions nobody would have imagined just months ago. Sure enough though, science will lead the way out. If we could leap five years into the future and read the story of COVID-19, we would not believe our eyes.

[1] SARS-CoV-2 can be detected in wastewater using RT-qPCR. In one study, the total load of gene equivalents in wastewater correlated with the cumulative and the acute number of COVID-19 cases reported in the respective catchment areas [Westhaus 2020]. Note that wastewater is no route for SARS-CoV-2 transmission to humans! All replication tests were negative tests.



By Bernd Sebastian Kamps
Stefano Lazzari

Please find the figures
in the free PDF.