In 1665 a tailor in the Derbyshire village of Eyam bought some fine cloth from London. Along with the cloth, he brought the plague to his village, perhaps from fleas infected with the bacteria Yersinia pestis that were living in the fabric. Once villagers started to die from plague, many tried to flee Eyam. That is, until the local priest persuaded people to stay put, to prevent the infection spreading, and for the greater good. Dozens died in Eyam, but nearby villages and towns were not infected. At least, that is the version told in the play, The Roses of Eyam, published in 1970, based on true events.

Quarantine works. Originally for forty days, sailors would observe quarantine when arriving into the Italian city states, such as Venice, during the 17th Century, to prevent any diseases they had being spread (quaranta is Italian for forty).

Lockdown is a form of quarantine, and it works too. The evidence is clear that what epidemiologists call non-pharmaceutical interventions (an intervention that doesn't involve a drug or a vaccine) have reduced, or in some locations, halted the spread of the coronavirus. A pair of studies published in Nature estimated that 530 million infections worldwide were prevented, with 3.2 million lives saved in Europe as a result of these interventions.

Last weekend in the UK, lockdown began to relax. Pubs were open and the high street was busy. But when lockdown ends, the virus can return. Despite millions of infections worldwide, most of the world has not yet caught the virus, and so remain susceptible. Even in Wuhan, less than 4% have evidence of infection by serology testing, which looks for antibodies against the virus. A past infection protects monkeys from re-infection, and the majority of those with a past infection develop antibodies, so those who have been infected likely have some protection from future disease, although there are some unknowns here, such as how long immunity lasts and how effective it is.

How do we relax lockdown whilst preventing the virus from returning? The goal, as many now know, is to keep the reproductive number below 1. This epidemiological term is now common parlance, and it means the number of people that each infected person passes the virus on to. The reproductive number seems to be between 2 and 3 in normal circumstances, meaning that each infected person passes on the virus to 2 or 3 others.

There are two goals for ending lockdown safely. Firstly, to reduce transmission of the virus by reducing social interactions. This can be measured by keeping the reproductive number below 1. The second goal is to prevent severe infections - both by shielding people that are particularly vulnerable to a severe infection, and by the use of drugs that stop people dying in hospital.

It is absolutely clear that the virus does not affect us all equally. Some people are more likely to catch the virus, due to their occupation (those working as drivers, in healthcare, or in food preparation seem particularly vulnerable), living in overcrowded conditions, living in care homes, or being older. Part of the reason why black people and South Asian people in the UK, and African Americans and Latinx people in the US, are more likely to catch the virus is likely due to their occupation. Inter-generational living and overcrowding also play their part, as well as existing health inequalities. COVID-19 is an infectious disease of poverty, and those living in poorer areas are more likely to catch the virus.

As well as some people being more likely to catch the virus, others are more likely to have severe disease. Here age is the biggest predictor, as well as those with pre-existing conditions such as hypertension, obesity or diabetes. The virus itself may cause diabetes in some people, and it also affects blood clotting, and causes thrombosis in some severe cases.

To slow the spread of the virus once lockdown ends, there is a continued role for face masks, which reduce the risk that you pass on the virus to others (especially if you are one of the up to 40% with no symptoms), and also a role for social distancing. In addition, the concept of bubbles or social networks seems to be effective. Again, mathematical modelling can be used to predict this, but the concept is logical. If you interact with people you already know, in the area that you live, then if you are infected, your bubble or social network is infected, but it doesn't spread beyond that. This essentially means you can interact with anyone you already know. But a mass interaction with many people you don't know, who then all go back home to their own networks, is dangerous. This is what happened at the end of the first world war, when the troops that were gathered in France all travelled home, to the UK, South Africa, India, and Australia, taking influenza with them.

It is therefore no surprise that the UK Chief Medical Officer has said (with apologies for the anthropomorphism) that the role of the pub in bringing people together is a great thing - from the virus' point of view.

To protect those at greatest risk, the answer is continued shielding, and testing. Test doctors, test nurses, test delivery drivers, test factory workers, test teachers, test bar staff, test baristas, test anyone who is forced to interact with people they don't know every day; those who have to mingle with people who are not in their existing social network. This allows outbreaks to be stopped as they start - as happened successfully in the Italian town of Vo', which prevented an outbreak from spreading through contact tracing, mass testing, and immediate shielding of vulnerable groups once the outbreak started.

Some people will inevitably fall ill with the virus, but hopefully not those at risk of severe disease. Here the (somewhat obvious) goal is that they survive. The overall case fatality rate of the virus is around 1 in 100, considerably higher than influenza, but considerably less than Ebola. The fatality rate of those that end up in hospital is much higher, around 1 in 5. This can be lowered.

Remdesivir looks promising. It didn't reduce the fatality rate in a clinical trial, but it did lead to a quicker recovery. The drug was given around a week after infection (it has to be administered intravenously, and so cannot be used in the community), so earlier treatment may be more effective. Hydroxychloroquine does not work, and may even cause harm, especially when combined with azithromycin. Drugs that tackle thrombosis, or immune dysregulation, both of which are seen in some severe cases, may also work. The RECOVERY trial found that dexamethasone, a common and cheap corticosteroid, reduces mortality from COVID-19, although the results are not yet been peer-reviewed.

If cases can be reduced through social bubbles and regular testing of those in at-risk occupations, and deaths can be reduced through shielding those at higher risk of severe illness, and a combination of new drugs such as remdesivir, and old drugs such as dexamethasone, then the second wave of the outbreak can be long, slow, and mild, rather than short, quick, and deadly. To use the current parlance - we can flatten the curve.

This matters because a vaccine is still many months away, and the economic and health consequences of lockdown are significant. Up to 28 million operations are predicted to be cancelled, ambulances take longer to reach patients, meaning someone having a stroke is less likely to survive it, deaths from heart disease, diabetes, and Alzheimer's are all up, and some patients aren't being treatment for cancer.

Lockdown cannot continue indefinitely, but when it ends, life cannot return to normal.

Image by Free-Photos from Pixabay.