Once a new patient is identified and isolated, he or she would no longer be able to transmit the virus. Only the as-yet-unidentified patients could continue to transmit the virus. If transmission can be limited, the number of unidentified patients would decline over time, falling to zero eventually, and the epidemic would be under control.

Preventing transmission by the unidentified patients is extremely important; otherwise, the number of new (and as-yet-unidentified) patients will continue to rise.

The Sars virus is transmitted primarily through close person-to-person contact, via respiratory droplets produced when an infected person coughs or sneezes.

A critical parameter was the number of new patients that a Sars patient would be likely to infect before he or she was identified. Obviously, this depended on the length of time that a Sars patient could remain unidentified, as well as the degree of stringency of the avoidance of contact with others.

Working backwards from the new-patient data and assuming an incubation period, I estimated these numbers, which turned out to be quite low on average, on the order of low single digits, and declining over time. I could therefore predict when this number would fall to zero, which would signal the end of the Sars epidemic.

Controlling the current coronavirus epidemic requires a similar strategy. We need to identify all the infected patients as soon as possible, and isolate and treat them, so they can no longer transmit the virus. Moreover, once a new patient is identified, we can also identify other people with whom he or she has had close contact, and proactively test these potential patients for possible infection, and isolate and treat if necessary.

For us, the public, we also need to take proper precautions by minimising physical contact with other people, both friends and strangers. If we succeed in doing so, we will be able to reduce the transmission of the virus to virtually zero. This is possible if we are willing to adopt and maintain good hygienic behaviour and practices.

Coronavirus forces world’s largest work-from-home experiment

For example, instead of shaking hands to greet others, we could use the traditional Chinese way of greeting, gong shou – that is, holding up your hands and cupping one hand in the other. It does not require physical contact, and can be done in a group as well as between two people.

We should wear masks, use disposable gloves when touching things in public areas, wash our hands with disinfectant or wipe them with alcohol regularly, and refrain from touching our eyes, nose and mouth with our hands.

The transmission of the virus will stop if everyone takes the proper precautions for his or her own protection. By taking measures to protect ourselves, we will also protect others by helping to reduce the probability of secondary transmission to zero.

There is no need to panic. In any case, the mortality rate of the coronavirus is, as of now, not high, about 2 per cent, and so far the deaths caused by the virus were concentrated among high-risk groups such as older people and people with pre-existing health conditions.

Now that the virus has spread beyond China, potential unidentified patients can in principle come from anywhere, not just from the Chinese mainland. Even closing the border completely to the rest of the world will not stop the increase of new coronavirus patients in Hong Kong because of the existence of as-yet-unidentified patients.

Now may be the time to close the border

All travellers to Hong Kong, wherever they come from and whether they are permanent residents, should be monitored and required to undergo tests and, if necessary, quarantined. If all of us continue to be cautious, there should be no new infected patients, and we would have stopped the epidemic.

Lawrence J. Lau is Ralph and Claire Landau Professor of Economics at the Chinese University of Hong Kong, and Kwoh-Ting Li Professor in Economic Development, Emeritus, at Stanford University