If you are confused about whether to wear a face mask, you're not alone. Even scientists cannot come to a consensus on whether they are a useful health intervention or could make the situation worse.
At the beginning of the coronavirus pandemic, the World Health Organisation (WHO) advised against mask wearing, only changing its advice last month to encourage their use where physical distancing is difficult, such as on public transport.
So what do we know about the effectiveness of masks?
One major problem with the current science on masks is that there are few randomised control trials (RCTs) that have been conducted in community settings with conclusive results.
Most studies have looked at health workers in hospitals, making it tricky to separate the use of masks from other personal protective equipment (PPE) such as eye shields, gowns and gloves, coupled with scrupulous hygiene practices.
Arguably it is unethical to restrict masks in a pandemic just to provide a control group which could give a definitive answer. But even among the gold-standard trials that are available, the results are mixed.
In May, the University of Hong Kong published an analysis of 10 RCTs looking at whether face masks prevented influenza transmission in the community. They found no significant reduction.
A similar meta-analysis by the University of East Anglia (UEA), found wearing a face mask slightly reduces the odds of infection by the wearer, but only by roughly six per cent.
Paul Hunter, professor of medicine at UEA and one of the authors of the study said: "The value of face masks in the community is still an issue that the scientific community has not yet reached consensus on.
“Randomised controlled trials – the gold standard of scientific evidence – of face mask use in the community have not proven that mandating their widespread use is protective. Other types of study have given mixed results, some finding no benefit and others, like case control studies, generally finding them to be protective."
However real-world data appears to show that masks may be effective. The University of Iowa found that the growth rate of coronavirus slowed down after masks were made compulsory in 15 US states and the District of Columbia.
Likewise, Virginia Commonwealth University studied 198 countries and found lower death rates in those with compulsory mask-wearing, or in which face masks were considered culturally normal.
Earlier this year, the Beijing Research Centre for Preventive Medicine found that if people wear face masks at home before they develop symptoms, it prevents transmission 79 per cent of the time. Yet after the onset of symptoms, masks make no difference.
A paper published this month by The Royal Society, a learned society comprising many of the world's most eminent scientists and the oldest scientific academy in continuous existence, found that, overall, the use of cotton masks lowered the risk of infection by 54 per cent and the use of paper masks by 39 per cent.
They found that the material and thickness was important, with effectiveness ranging from five to 95 per cent depending on the fabric, number of layers and fit. A double layer of cotton and chiffon performed best.
How do masks offer protection?
Put simply, at least some transmission of coronavirus occurs in droplets from coughs, sneezes and even talking, so covering the nose and mouth stops an infected person projecting the virus into the air around them.
In March the Massachusetts Institute of Technology (MIT) found that viral droplets expelled in coughs and sneezes can travel in a moist, warm atmosphere at speeds of 33ft to 100ft per second, creating a moving cloud that can span approximately 23ft to 27ft.
The researchers warned that droplets can stay suspended in the air for hours, moving along airflow patterns created by ventilation or climate-control systems.
Laboratory testing by Northeastern University, in Boston, found that surgical masks block out 75 per cent of respiratory-droplet-size particles, while the University of Hong Kong found that loose-fitting surgical masks block almost all contagious droplets breathed out by infectious people.
Prof Rowland Kao, the chair of veterinary epidemiology and data science at the University of Edinburgh, said: "Basic mechanical principles tell us that in indoor spaces, where air is relatively still, this effect is even more important.
"As we move towards larger concentrations of people coming together in indoor spaces, and as the rules on social distancing become relaxed, use of face masks has a potentially even greater role to play in reducing transmission."
However it now appears that masks are better at preventing infected people from passing on the disease than helping to protect the wearer.
One of the major scientific discoveries of the past few months is that there are huge numbers of people with the disease who are asymptomatic – maybe even as much as 80 per cent.
Studies have found that viral load – the amount of virus present once a person has been infected and the virus has had time to replicate in their cells – peaks in the days before symptoms begin and that simply talking is enough to expel infectious droplets.
A study by the National Institutes of Health in the US, using high-speed video, found that hundreds of droplets ranging from 20 to 500 micrometers were generated when saying a simple phrase. But nearly all of the droplets were stopped from spreading if the mouth was covered with a damp cloth.
Adam Finn, professor of paediatrics at the University of Bristol, said: "Most masks will do a much less efficient job of protecting the wearer from breathing in droplets and no mask will stop you infecting yourself with your hands.
"In fact, if you spend a lot of time touching and adjusting your mask and your hands are contaminated, they could even increase your risk of self-inoculation."
The situation in the UK?
To make the situation more confusing, recommendations across Britain have varied considerably.
On April 28, Scotland advised the public to wear face masks in enclosed spaces where social distancing is difficult to achieve, and imposed mandatory wearing on public transport as of June 22. On June 9, Wales also recommended that face coverings could be used where it might be difficult to stay two metres away from others and advised using three-layer, non-medical face coverings.
England made wearing masks mandatory on public transport on June 15 but has so far stopped short of advising that they should be used elsewhere. Some scientists believe there is still not sufficient evidence to advise widespread use.
Dr Antonio Lazzarino, of the Department of Epidemiology and Public Health at University College London, said: "The evidence has not changed in the last few weeks – it's still extremely weak. The question is why politicians may be changing their minds now.
Scientists are also concerned that if wearing a mask becomes mandatory in shops, they would also need to be brought into pubs and restaurants, where they could do more harm than good.
"The constant need to be always moving your face covering to take a sip of beer could be a risk,” added UEA's Prof Hunter.
Yet other experts think there is enough data now to suggest masks could be useful.
Linda Bauld, professor of public health at the University of Edinburgh, said: "A number of new studies and systematic reviews have persuaded most researchers and public health officials that they should be worn, including those who were sceptical a few months ago.
"Growing evidence on potential airborne transmission of the virus adds to the case for face coverings."