This page aims to summarise our understanding of the current science on key questions about COVID-19 (as of 3 April, 2020), as best we can given the state of the evidence and the fast moving situation. We provide more explanation as well as sources in the footnotes.
Table of Contents
Symptoms and severity
- The most common reported symptoms are cough (appearing in about 80% of confirmed cases – meaning those who have been tested and found to be infected with the virus) and fever (80%-90%).1 Many also experience shortness of breath, usually later in the disease progression. Diarrhea and other GI symptoms have also been seen in some patients.2 Nasal congestion and runny nose seem uncommon (<5%).3 Anecdotally, loss of the sense of taste or smell have also been reported.4
- Once someone is infected, it seems to typically take ~7 days for symptoms to develop. One study with a large sample size found that for 11.5% of confirmed cases it took more than 14 days. 5
- According to initial data from China, around 81% of confirmed cases are ‘mild’ (though can still involve pneumonia), 14% are severe (requiring hospitalisation), and 5% are critical.6 A large proportion of people infected with the virus have mild symptoms, and around 20% may have no symptoms,7 though there is not reliable data here.
- Most current estimates of the fraction of infected people (rather than people with confirmed cases) who die from the disease (the ‘IFR’) seem to be between 0.1% and 2%.8 The fraction will vary based on whether healthcare capacity is overwhelmed, as well as the age and health of the population. One reason estimates are uncertain is that we don’t know how many mild and symptom-free cases there are, which usually go unconfirmed; if there are more, the IFR is lower.
- Risks of long-term health effects are unknown because not enough time has passed, though there are concerns that some portion of COVID-19 patients (perhaps those with more severe cases) might face ongoing issues like reduced lung capacity.9
Who is most at risk?
- Children can catch and spread the virus, though they may be less likely to, and they rarely have serious symptoms.10
- Men seem to be about 50% more likely to die of the disease than women on average.11
- Age is a large risk factor for mortality, as suggested in the set of estimates in the table below, used by researchers at Imperial College London. The estimates are for infections (rather than for confirmed cases) and based on data from the early phase of the disease in China:
How is the virus spread?
- How COVID-19 spreads is not fully understood, but most experts seem to think droplets from coughs and sneezes are most important. ‘Aerosols’ — smaller droplets that can be expelled even while talking and hang in the air longer — may also be significant.12
- This means the main ways you can become infected are by being close to someone talking, coughing, or sneezing, and by touching an infected surface and then touching your face. Sources commonly recommend staying six feet away from others, but this just is a best guess for how far viral particles can travel through the air, and further is of course safer. On plastic and stainless steel, virus particles can remain viable for three days or more; they remain viable for less time on copper and cardboard.13
- It’s thought that people can be infectious perhaps one or two days before they develop symptoms, though they’re less infectious than when they have symptoms.14 Totally asymptomatic people (including children) may also be infectious. It’s currently debated how much transmission is driven by asymptomatic and very mild cases.15
- It’s unknown how long people stay infectious after they develop symptoms. The CDC recommends home isolation for seven days after symptoms start and three days after they resolve. By contrast, the WHO recommends isolation for 14 days after symptoms resolve.16
How many will be infected?
- You can track the total number of confirmed cases and deaths attributed to COVID-19 by country on the Johns Hopkins dashboard (or see here if you want a lot more detail).
- The number of infections is higher than the number of confirmed cases everywhere. By how much depends on how extensive testing is, which varies extremely widely – it seems plausible some countries could be capturing up to half of infections while others are capturing as little as 1% (or even less).17
- How many will eventually become infected (and how quickly) depends on how we respond. If spread is uncontrolled, models suggest that at the start of the epidemic each infected person will infect on average another ~2-6 people (and perhaps more).18 Interventions like social distancing reduce this number substantially. If each person infects between 2 and 6 others, it’s been estimated that ~50-80% of the population would need to develop immunity (most likely through being infected and surviving) in order for the population as a whole to have “herd immunity.”19. If the virus spreads uncontrolled worldwide, this would likely result in tens of millions of deaths.
- However, many governments are now acting to suppress the spread. The best all-things-considered forecasts we know for total eventual reported cases and mortality are from the Good Judgement Project. For example, as of 3 April, forecasters think there’s a 24% chance that more than 8 million people will die worldwide from COVID-19 before 31 March, 2021 and a 17% chance that less than 800,000 will.
Reducing the risk of infection
- The most highly recommended ways of limiting the spread are staying home, limiting contact with others, washing hands thoroughly, and covering coughs and sneezes with a tissue.20 If you want to go all out see the guidelines for individuals from endcoronavirus.org.
- Getting plenty of sleep and exercise can help strengthen your immune system.21 We’ve also heard some evidence that Vitamin D reduces the frequency or severity of respiratory infections, though this is a less common recommendation.
- Expert consensus seems to be that wearing a surgical mask does help prevent you from infecting others. Due to the shortage among healthcare workers, healthy members of the public shouldn’t buy surgical masks (or N-95s) at the moment. For this reason, we opted not to review the evidence for apparently healthy people wearing surgical masks in ordinary life, though our impression is that they do offer some measure of protection.21
- Homemade masks have not been rigorously tested, so we have to rely on weaker forms of evidence. The jury is still out on whether homemade masks are overall helpful, but there’s a common-sense case in their favour, and their use in some circumstances has recently been endorsed by the US CDC.22
- There is currently no vaccine or other proven pharmaceuticals. The main treatments provided by hospitals are things like life support in severe cases (e.g. assistance breathing) so patients live long enough for their immune systems to eliminate the virus.
- Blood plasma from recovered COVID-19 patients has been used with some success in treating critically ill cases. The treatment has not been thoroughly tested or approved, but results from early cases of “compassionate use” seem promising.23 Unfortunately, scaling up this treatment is a challenge because it requires blood from donors.
- There are many trials of antivirals being undertaken, which could reduce severity. It’s very hard to predict, but it seems plausible we could see some antivirals developed within a year.24
- Developing a vaccine will take at least a year, and potentially several years.25
Learn more: we made a list of other primers on COVID-19.