This is part 2 to Diving Deep into the Numbers. My point is not to minimize the harm caused by COVID-19, or efforts to minimize its growth and effects. Rather, my hope is to put things into perspective, using data and numbers, since putting large numbers into perspective can be challenging. Moreover, while the idea of a deadly virus is certainly scary, with a unique combination of the ability to spread exponentially (contagion rate) and to kill (CFR), I believe there is also a measure of the availibility heuristic at play. Movies and our imagination lead us to believe that we are more likely to die in a plane crash, terror attack, or by a mysterious disease spreading across the globe, rather than from heart disease, smoking, car crash, more “ordinary” accident, or a combination of health factors that reflect genetics, lifestyle, and circumstance. Statistics says otherwise.
In this post I will take another look at the data available, and try to parse it to present some different perspectives, as well as a few policy suggestions.
- The mortality rate is misleading. There is not a 3.4% chance that you will die of Coronavirus. It’s not even a 2% rate. Those estimates are the mortality rate of infected patients. More accurately, they reflect death rates from known cases. The first question is whether you as an individual will contract the disease. That is an unknown variable dependent on a number of factors, but probably much lower than 100%. Watching the rise in cases in Italy and several other countries is certainly scary, but China, South Korea, and other countries offer hope that the growth can be slowed if not controlled altogether, which reduces the chances of ever getting it. It is only within that subset of people who catch Corona, that we can begin to discuss the mortality rate.
- It is clear that children, and to a lesser degree women and those under 40, aren’t nearly as affected by the virus. That could mean that they are less likely to catch (and spread it), or less likely to suffer symptoms severe enough to be detected (in which case they are more likely to spread it unintentionally), or a combination of both (my guess). Either way this would affect the likelihood of contracting it, or the CFR.
- Assuming one has the virus, the odds are still probably much lower than reported. Although exact numbers are tricky to calculate (those who are still sick may be included in the calculation, whereas some may eventually die of related complications), there are most likely many who have not been diagnosed because 1. the symptoms resemble the flu and other less virulent diseases; 2. the symptoms haven’t manifested or haven’t manifested as strongly (see youth), 3. limitations on testing capabilities. Most of the data comes from hospitalized victims. What does this mean? Many more have or have had Corona, and were fine, meaning the mortality rate is far lower, perhaps under 1%.
- So if I do test positive for Corona then I have a 1% chance of dying? No–that’s still not true. The data gives us more information than that. In reality you’re risk factor could range from close to 0 up to 15 or 20%, even higher in extreme cases. As noted, the virus either doesn’t infect, or causes only extremely mild symptoms, in most children. In general, those who are under the age of 40 who do get sick are still at extremely low risk, probably less than 0.1% (1 out of a thousand), and those under 50 are still well under half a percent, assuming no other health conditions. As people get older, the risk grows significantly, with those in the 70-80 bracket estimated at 8% likely to die, and those over the age of 80 at nearly 15% (WHO data). In Italy for example, a preliminary study showed a median age of 81 for those who died of Coronavirus, with most victims having 3 or more prexisting medical conditions. To put that in perspective, global life expectancy is about 72. Most of those who died in Italy were 9+ years older than the average person expects to live around the world (Italy has one of the highest life expectancies in the world; more on that later). More women are also dying from Corona than men, although like children, the data is not conclusive about whether they are less susceptible to contracting it, or just having milder responses (or healthier overall and thus better equipped to fight it off; women have higher life expectancies than men by several years due to a number of factors).
- Comorbidity means having multiple conditions simultaneously. This is most likely to be what kills someone with Coronavirus- the virus impacting an already weakened immune system and body, and/or reducing its ability to cope with prior conditions (in at least some cases conditions which may have already been terminal). This is probably the case with all victims, but in particular with those under 60, when the likelihood of getting seriously ill from Corona is generally quite low, and most people are generally healthy.
- What do 4 and 5 mean? While all life is sacred, we can still to put these numbers in perspective using actuarial life tables. Actuarial tables provide probabilities of dying at different ages, and expected life expectancy. As a person gets older, their odds of dying increase (with the exception of the few years), and the number of years they can expect to live decrease. For a simple example, I will include an example using 2015 US data:
Male Female Age Death
10 0.000088 66.65 0.000093 71.54 20 0.001173 56.91 0.000423 61.68 30 0.001794 47.72 0.000803 52.01 40 0.00242 38.59 0.001403 42.5 50 0.005007 29.69 0.003193 33.26 60 0.011533 21.61 0.006848 24.6 70 0.023122 14.4 0.015413 16.57 75 0.035963 11.18 0.025035 12.97 80 0.057712 8.34 0.042539 9.74 85 0.096603 5.94 0.073763 7.01 90 0.163689 4.08 0.129706 4.85
These numbers range from country to country, and of course would change drastically with more precise information (smoking; weight; various medical conditions, etc.), but it’s a reminder that the general mortality rates are not contant for all demographics, and that if this disease did affect children as many others do, the effects would be much worse. (and one last point to put in perspective–life expectancy a century ago was under 60).
- Median age is the midpoint of the population in a country. Global mean is 29.6 years old. Of the 30 countries with the most reported cases of Coronavirus, all have median ages above the global median, and 25 of them are 7 years or more above the global median. Most are 10 years or more above this average. What does this mean? That the virus has spread fastest, or at least affected the most, the oldest countries in the world. What else does it mean? That the effects could be far less when it hits much younger countries (these are averages, so for every China, US, or Italy, ten years above the average median, there are countries like Afghanistan, Nigeria, Ethiopia, Yemen and Kenya, where the median is under 20, and we can expect the virus to either die, or see a much lower effect, depending on the true effects on younger populations. Either way, that mortality rate would be lower. Here’s another nice little graph to look at:
Country Median age Global 29.6 1 China 38.4 2 Italy 47.3 3 Iran 31.3 4 S. Korea 41.8 5 Spain 44.9 6 France 41.2 7 Germany 45.9 8 USA 38.2 9 Switzerland 42.2 10 Norway 39.2 11 Denmark 41.6 12 Diamond Princess 46.7 13 Japan 47.3 14 Sweden 40.9 15 Netherlands 42.1 16 UK 40.5 17 Austria 43.2 18 Belgium 41.2 19 Qatar 32.3 20 Bahrain 32.5 21 Singapore 42.2 22 Australia 37.9 23 Canada 40.8 24 Malaysia 30.3 25 Greece 43.4 26 Hong Kong 43.2 27 Israel 30.2 28 Czechia 43.2 29 Iceland 36.5 30 Finland 42.5
- What do we do with all of these numbers? Firstly, not panic. Avoiding unnecessary travel and reducing large gatherings will certainly help slow the spread of the disease, and washing hands is a good idea regardless. But it’s also important to not overreact, because of the economic effects, and the stress it’s creating. Both of those can also impact life and health. It’s easy to say “Health is a priority; economy comes second.” But there’s something of a false dichotomy there; if we had unlimited resources, then much of the concerns about having enough medical supplies would be irrelevant. Medical care is expensive, and it can save, lengthen, and improve the quality of, life, as intense debates in the US, and many other countries, often remind us. Decisions about work, travel and school must thus be weighed against their impact, and policy decisions must thread a difficult needle. Moreover stress has a negative effect on the body, and prolonged stress, as individuals and society, can have some very negative effects, including on health. These are just some of the factors that decision makers must take into account.
- With that in mind, here are some thoughts policy-wise. Firstly, sports and entertainment should go on, without crowds. Major American sports events have all been suspended or canceled. I think that’s a terrible idea. Sure, athletes are extremely highly paid, and run some risk of exposure, even without fans. But the risk of exposure and serious risk to 20-30 year olds in excellent physical shape is minimal, and the diversion that watching games and highlights online offers to the rest of society is extremely important. On the other hand, visits to nursing homes should be suspended, and even greater care should be taken to promote hygiene there and in hospitals, where there are many immunocompromised people. Setting up separate medical facilities for Corona patients in every area affected is essential, for the same reason- those in hospitals are the ones most at risk from the virus; every effort to reduce exposure should be made (medical staff should ideally not be treating Corona patients and other patients, and taking frequent tests). The elderly and those with pre-existing conditions should avoid traveling more than necessary, or going to public places, but could also benefit from programs to provide basic supplies via drone, be given temporary housing options (if living with family and it proves that children can be carriers), and given special lines in stores, etc. This is the medical equivalent of hardening soft targets from attacks–finding ways to protect those at special risk.
- Schools pose a more complicated challenge–given what we know and don’t know, schools do not pose a serious risk to students, but may be a risk to teachers, and potentially contribute to spreading the virus. But it’s also a serious disruption, with a significant long-term impact, and an immediate strain on the entire society. A better approach might be to have older teachers stay home, while classes continue as usual for all but those with known medical conditions or compromised immune systems.
- The situation is not static: there will be improvements in testing, treatment, and eventually vaccines. Likewise, the virus could mutate, making it more difficult to combat, or deadlier. It could go the way of measles, or the flu–it depends on a number of factors. Either way, the status quo does not guarantee the future.
- Long term effects: there will be some interesting long-term effects. The economic effects are drastic, and overall negative for many people. Dispruption in travel patterns could be long-lasting, as airlines struggle to recover, and subconscious fears linger. Corona has taken attention away from many other issues of global importance, like climate change, poverty, gender equality, and other more endemic issues. But the potential for global cooperation has been raised as well, with the need to share information, scientific knowhow, and collaborate highlighted. The solutions to this challenge will be the result of truly international collaboration, meaning that closing borders is a temporary blip in globalization. Among the more interesting effects will be how countries recover from Corona-if it affects them on micro and macro levels. The countries thus far most affected, and likely to last that way, are countries with some of the lowest birthrates in the world (obviously correlated with and related to higher median average and higher life expectancy). Does that mean that some of these countries may never recover economically? Could this trigger a re-emphasis on the nuclear family, and a spike in birth rate? Will this ultimately speed up the movement of the global economy south and east, as many have already predicted? Or will it instead recalibrate the global economy? What effects will it have on how we work, interact, and otherwise live, in an increasingly digitial age? These are just some of the questions raised.
Country Fertility rate Global 2.4 1 China 1.6 2 Italy 1.4 3 Iran 1.7 4 S. Korea 1.2 5 Spain 1.3 6 France 2 7 Germany 1.5 8 USA 1.8 9 Switzerland 1.5 10 Norway 1.7 11 Denmark 1.7 12 Japan 1.4 13 Sweden 1.9 14 Netherlands 1.7 15 UK 1.8 16 Austria 1.5 17 Belgium 1.7 18 Qatar 1.9 19 Bahrain 2 20 Singapore 1.2 21 Australia 1.8 22 Canada 1.6 23 Malaysia 2 24 Greece 1.3 25 Hong Kong 1.2
- Interesting sources: