Uncertainty surrounds aspects of COVID‐19.

1. Nomenclature : the World Health Organization refers to the new disease as COVID‐19, but virologists await an acceptable name for its virus. The international taxonomy committee's proposal of “severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2)” is cumbersome and seems unlikely to catch on.
2. Origin : the source of the current SARS‐like coronavirus is unknown. The live food markets of South East Asia evidently provide opportunities for genomic re‐assortment of the coronaviruses of various species, and late in 2019 one re‐assortant may have acquired the ability to infect humans. The presence of an advanced virology facility in the Chinese city of Wuhan may be coincidental, but conspiracy advocates are suggesting that such a virus, grown there to high titre in cell culture, might have infected one or more laboratory workers and they might then have become the index cases for local outbreaks; but that is just supposition.
3. Virus‐host interaction : the course of human infection with SARS‐CoV‐2 is as ill‐defined as its origin. After an incubation period of several days, there is in most clinical cases fever, headache and persistent cough lasting about a week. Thereafter a slow recovery is usual, but older and in other ways more susceptible patients can go on to develop respiratory failure, which is often not reversible. The proportion of infections that are clinically manifest in this way, or are milder or sub‐clinical, is as yet unknown. No reliable death rate can therefore be calculated. These proportions may furthermore change with time. It is possible that the natural history of the better characterised coronaviruses of livestock, other animals and birds may indicate how the virus of COVID‐19 can be expected to behave in the future.
4. Virus excretion : at this stage of the pandemic, individual local outbreaks are still discernible; but even where these have been formally studied, the length of the period of transmissibility from an index case is still ill‐defined. Both the onset of infectiousness and the interval of 14 days adopted by many countries as the period of continuing infectivity need to be examined critically.
5. Virus dose : the amount of infectious virus needed for transmission to occur is unknown. Another unknown is whether not just transmission but subsequent clinical severity might be determined by the infecting dose. The relative importance of the various routes of infection, for example, coughing into the face, aerosol, surface contamination and transfer from hand to mouth, nose or eye, is also uncertain. All these routes are assumed to be significant and are the basis for the precautionary measures widely promoted by governmental “lockdown.”
6. Diagnostic reliability : it would help to resolve some of the foregoing uncertainties if polymerase chain reaction (PCR)‐based diagnostic reports contained cycle threshold information. They could then be regarded as semi‐quantitative, with some weak PCR signals perhaps being due to the presence of non‐viable virus only. Other weak signals may be non‐specific. Although PCR tests will continue to be used on a large scale, they are technically quite demanding and, were it possible, a SARS‐CoV‐2 antigen assay comparable to that used for the diagnosis of hepatitis B would be valuable.
7. Safety constraints : laboratory investigators may be reluctant to inoculate specimens into cell culture for safety reasons, but this is the best means of quantifying infectivity. Laboratory containment needs to be available and regularised so that more such studies can be done.
8. Antibody to SARS‐CoV‐2 : as of early May 2020, no serological test indicating past COVID‐19 or sub‐clinical infection has yet been entirely validated, and an international standard serum for antibody to the virus is urgently needed with related standards then being distributed nationally. It is not yet established which assay formats can best offer sensitivity and specificity, either as in‐house assays or in the form of commercial kits. The current political focus is on bedside and other point of care tests, but these may lack sensitivity so that the accuracy of each one of them needs to be defined. A valid antibody test would resolve the present uncertainty about the proportion of national populations that have already experienced infection. So far the virus may be far from prevalent worldwide and still so ill‐adapted to humans that only a modest proportion of populations is yet or perhaps ever will be infected. For those people shown to have antibody, there is no guarantee of long‐lasting immunity.
9. Protection from disease : therapy and prophylaxis are uppermost in the public mind. The analogy with influenza, weak though it is, suggests that genomic “drift” of SARS‐CoV‐2 might occur over time and complicate both anti‐viral treatment and immunisation. The use of any drug, as well as of convalescent serum, will depend on their ability to prevent or mitigate illness; and the evaluation of candidate vaccines will depend on their capacity to generate both an immune response and, as a marker, the development of anti‐SARS‐CoV‐2. With so much public pressure to roll out a vaccine, there may have to be some derogation from the orthodox conduct of vaccine trials. Vaccine evaluation will also require access to an accurate antibody test. To give lasting immunity, it may in the future be necessary to develop an attenuated whole virus vaccine.
10. Continuing virus activity : It has been possible to study the pandemics consequent upon shifts in the human influenza virus genome for over a century, and virtual pandemics have not been rare, vide, recently, Nipah, Sars1 and Zika. The COVID‐19 pandemic may follow an influenza‐like pattern, or it may establish an equilibrium following widespread human exposure to the virus, or it may be a short‐lived phenomenon.