In the absence of an efficient vaccine, the control of the COVID-19 pandemic currently relies on non-pharmaceutical interventions (NPIs).

Since these interventions are socially disruptive and have substantial economic costs, they have caused controversy. Two distinct approaches have been considered. The mitigation approach tries to reduce the spread of the disease so that it does not overwhelm the healthcare system. By reducing, but not suppressing, transmission the goal is for the population to achieve herd immunity. The UK government initially opted for this approach, and Sweden still follows it. When herd immunity is reached the disease gradually disappears from the population. For herd immunity against SARS-CoV-2, an estimated 60% of the population need to develop protective immunity to bring the epidemic to a halt. Achieving herd immunity for SARS-CoV-2 is difficult for several reasons. First it comes at a cost, which can be calculated with the infection fatality rate (IFR); the probability of death for an infected subject, which causes an ethical problem. The best current IFR estimate for SARS-CoV-2 is 0.6% (on average, because IRF increases markedly with age). Herd immunity would be achieved at the prize of 150,000 deaths in the UK, or more than one million deaths in the USA if the epidemic were simply left to roll over the population. The threshold for herd immunity depends on the ‘force’ of the infectious agent, mathematically expressed as the basic reproduction number R, which is defined as the number of new infections transmitted by an infected person. For this strategy, the R value needs to be kept slightly above one, since with R<1 the epidemic retreats and the virus will be eliminated over time. However, it is difficult to maintain R at 1.2, which ensures that the epidemic growth is small enough to prevent the collapse of the health system and to keep the death rate low. Furthermore, even regions severely affected by COVID-19 have not yet reached herd immunity. Seroprevalence studies, which measure the presence of specific serum antibodies against SARS-CoV-2 as a footprint of a past infection, indicated that ‘only’ a third of the population in New York had anti-viral antibodies. In addition, whether the presence of serum antibodies indicates protective immunity in humans is still unknown. Indeed, we do not know whether infected people are protected against reinfection, as singular cases of reinfection have been reported. Finally, the antibody part of the immune response against SARS-CoV-2 seems to be relatively short-lived and herd immunity, once achieved, might rapidly wane. However, cellular immune response against SARS-CoV from 2003 was maintained over years.


Other governments have followed an alternative strategy. With the suppression concept they want to reduce R<1 to eliminate the virus, which is easier said than done. Once eliminated from a region, one must prevent reintroduction of the virus by travellers from abroad. Since animals can be infected with SARS-CoV-2, animal reservoirs also need to be considered in the future. China and Germany are examples of a stricter and a softer suppression approach, respectively, reflecting different political systems. For both countries, no excess mortality (increased death rate compared with the means of previous years) was observed over the first six months of 2020. In contrast, excess mortality in New York was comparable to that of the Spanish flu epidemic of 1918. The suppression strategy can be summarised as ‘test, trace and isolate’. It comprises identifying new cases by surveillance and testing and isolating them to prevent further transmission. These measures must be complemented by tracking all contacts of cases and quarantining them. Additional physical barriers to viral spread are necessary, such as social distancing to reduce exposure to viruses transmitted by speech, sneezing and coughing; ventilation of rooms to curtail aerosol transmission of virus; and wearing face masks over the nose and mouth to reduce the viral load excreted by an infected person and to a lesser extent the inhalation of virus by a susceptible person. Epidemiologists also recommend the banning of mass gatherings and travel restrictions. The ultimate tool is restricting personal mobility in lockdowns. This tool has shown efficacy in curbing epidemics in many countries but comes at such a high cost that it can only represent a last-resort emergency tool. The challenge is now to find a suppression strategy that is efficient without causing too much collateral damage. Since these control measures depend on the cooperation of the population, communicating science becomes a pivotal element of this strategy. Counterintuitively, when the suppression strategy works, the population starts doubting its usefulness. This trend is amplified by vocal groups on social media and on the street who question the viral threat or campaign against vaccines, organising disinformation campaigns that spread on the internet faster than information by health organisations. In this situation, scientific societies have an important role to play in providing trusted scientific information in an understandable language. Indeed, this pandemic (which had long been predicted by virologists) has scared populations, since even industrialised societies were found to be unprepared for such an event, as shown by the initial lack of personal protective equipment for healthcare workers. Being unfamiliar with the scientific method, changes in strategies (e.g. school closures, mask wearing) were perceived as indecisiveness and not as adaptation to new scientific insights.


To illustrate this point, as children are a motor of influenza virus transmission and suffer clinically from flu, it was assumed early in the pandemic that this might also be the case for COVID-19, which led to a widespread closure of schools. Later research showed that SARS-CoV-2 infected children mostly experience only mild disease, or no symptoms (except for rare children showing a severe multisystem inflammatory syndrome). Early data from China showed transmission in households with some also initiated by children. However, most household transmissions were between spouses, and it is worth noting that only 30% were infected from their bed-sharing partner, indicating that COVID-19 is not a flying viral infection such as measles. Social network analysis showed that children had numerous contacts when going to school but, when reconstructing infection chains, epidemiologists discovered the importance of super-spreader events. Indeed, bars and social events such as weddings and religious ceremonies served as hotspots of viral transmission, but not schools. This led to a cautious opening of schools, which was also politically motivated by educational deficits (one year loss of school translates into 10% lifetime income loss) and are needed to lower the burden on working parents. Testing and tracing identified further hotspots of infection such as crowded housing and transport conditions of migrant workers (Singapore and India, but also meat factories in the USA and Germany). These public health activities have now led to more targeted or ‘granular’ interventions, where lockdown measures are imposed only at local district level where and when daily new cases transgress a threshold (50 per 100,000 inhabitants in Germany).


The US Centers for Disease Control and Prevention (CDC) was initially hesitant to recommend masks because objective proof of their effectiveness was scientifically lacking. Meanwhile, numerous physical experiments, mostly by engineers, have demonstrated the substantial filtration effects, even of self-made masks on the output of viral loads. When analysing the stepwise regional introduction of masks in Germany, epidemiologists found strong evidence that mask wearing reduced the infection rate in regions where it became compulsory, compared with neighbouring regions that had not yet introduced these rules. Masks are now a cornerstone of epidemic control measures. US physicians have even suggested an interesting hypothesis according to which general mask use would lead to subjects getting infected with lower viral doses, which could cause milder disease symptoms, but still allow the development of an immune response. If the autumn rise of infections in Europe is not followed by an increase in hospitalisation and death rates, widespread mask wearing might decrease IFR. We must for the moment learn to live with the epidemic. This implies better protection of vulnerable populations, especially elderly citizens and particularly nursing home residents who in some countries contributed 80% of the COVID-19-associated deaths. Research in the USA and the UK showed that ethnic minorities are also disproportionally affected by the pandemic, partly due to underlying health problems (obesity, diabetes) and partly due to jobs that do not allow home working and expose the workers to higher infection risks, further increasing social disparity in affluent countries. Creating safe working places will become an important task to alleviate the economic downturn. COVID-19 has reversed positive trends in poverty reduction and vaccination coverage in developing countries. Increasingly popular selfish national instincts are certainly not an answer to global health and economic problems.