First detected in December 2019, almost 4000 people have died from the novel coronavirus, COVID-19 (as of March 12, 2020). While it started in China, it is currently hitting South Korea, Italy, and Iran hard. Spain, France, Germany, and the US all have over 1000 reported cases, cases doubling rough every 5-7 days. Sports leagues have been shuttered, schools have been moved online, flights have been canceled, bans on travel have been enacted, and hospitals are becoming overwhelmed.
Some people are panicking, stockpiling water, masks, and even toilet paper. Other people are claiming the world is overreacting, and that we need to calm the …. down. That there’s no need for the bans on travel, the closing of factories, or the stockpiling of toiletries.
Who is right? This is a question best answered by calculating risk.
Calculating Risk
When assessing risk, we generally look at two things, the probability of the risk and the magnitude of the risk. In other words, how likely is it to occur and if it does occur, what impact will it have. For infectious diseases, two metrics help capture the magnitude and probability of impact – death rate and the contagiousness of the disease, captured as R0.
Death rate is a simple calculation consisting of deaths due to the disease/total # of cases. At least, it seems simple. However, cases may be susceptible to under or over reporting. We assume that the tests being provided are accurate, but that’s not always the case. Some tests can be guilty of false positives, incorrectly stating that some has COVID-19 when in fact they don’t. In fact, doctor’s are worried about false positives of COVID-19 tests. If a country adopted a test that resulted in significant false positives, they may paint a picture that the virus is not as dangerous. That could lead to a lower expectations of risk.
Under reporting of cases can also impact death rate metrics. Missed cases could be due to false negatives in testing, no tests available, or tests only given to people with symptoms, missing asymptomatic cases. Under reporting cases could lead to higher fatality rate than actually exists, causing people to overreact and impose more draconian measures to stop the pandemic.
While the number of deaths could also be miscounted, there’s less chance of that due to most people using the health care system if their symptoms get bad enough. So the official numbers there are likely close to correct. However, death rates might be impacted by the quality of health care, the overall health of the population, and maybe even the weather.
Calculating R0 (pronounced R naught), entails using the infectious period, contact rate, and mode of transmission to estimate the average number of people infected by each sick person. My understanding is that this figure is much harder to calculate. Calculations based on contact rate would be directly related to the social reaction to a disease. If the public self-quarantines during an outbreak, the R0 may drop considerably because the virus finds it hard to find more people to infect. If the certain sub-population is extremely social, the R0 could be higher. So at best, doctors come up with a range for that figure.
How serious is Coronavirus?
According to the graph below, COVID-19 has a much higher fatality rate and transmission rate than the seasonal flu, even by the lower estimates. [Edit: In a recent interview with infectious disease expert Dr. Amesh Adalja, Amesh suspects the true number is at the lower bound.] But is less dangerous than either smallpox or polio, but we have vaccines for those two. It’s not as deadly as Ebola or MERS and less likely to spread than measles and chickenpox. But overall, it seems to spread quickly and have a non-trivial fatality rate, especially for the elderly.
Based on what we know so far, the risk seems to be low, especially among healthy younger population. That is why I have no fear for myself, my wife, or my kids. For older, sick, or immune compromised individuals, the risk is moderate to high. Some areas may experience increased risk due to a shortage of hospital beds should a large number of sick individuals need care suddenly. Florida retirement communities, I’m thinking of you here. This seems to be the major problem in Italy. The population is generally older and many of the hospitals are running beyond full capacity trying to care for those older citizens in critical condition due to the virus.
However, I don’t believe the government should be mandating restrictions on travel. While a case may be made that the government should have that power in severe pandemics, I’m not convinced that COVID-19 has reached that level. Besides the attack on individual liberties, these restrictions will inevitably lead to a short, severe recession. This could hurt many more people indirectly than the virus does directly. In fact, I would rate the risk of severe recession much higher than the risk from the disease. In the next few months, expect to see many small businesses fail and unemployment skyrocket, especially in service and hospitality industries.
At the current rate of spread, I suspect this virus will infect over 10% of the world population, perhaps as much as 50%. Even if we slow the spread through these restrictions, COVID-19 will likely stay with us throughout the rest of this year and into next – if not beyond. The best things we can do is be prepared, but not to panic. If you have older or sick relatives or friends, see that they have the resources to limit social gatherings for several months. If they get sick, be sure they follow best practices in care and avoid transmitting it to others.
In short, the risk of this disease suggests we should not freak out but cognizant of the dangers to those around us. Preparing for the inevitable virus is not a bad thing and may save us some overreacting, causing secondary problems if the economy falters. What’s hope that as we improve the accuracy of the risk metrics, they narrow to the low side of the estimates, fewer deaths and fewer infected people.