# Is the coronavirus less fatal than early predictions suggested?

The coronavirus is an icky disease that takes a cruel toll on the elderly, the sick, and the unlucky.  In this modern era, we can outwit many things that once routinely killed people, but the Grim Reaper is still out there, and he'll eventually get all of us.  Scary headlines have hinted that the coronavirus is now the Grim Reaper's preferred method.

Media reports have told us that the coronavirus is significantly more deadly than the flu, which annually kills 30,000 to 60,000 Americans.  Based on the speed with which it killed in China, Italy, Iran, and Spain, it looked as if the American death toll could easily top two million people annually.  While that's small potatoes compared to past pandemics (e.g., the Plague of Justinian, the Black Death, Spanish Influenza), it's a staggering toll in modern America.  Any actions seemed worthwhile to keep America from turning into a viral slaughterhouse.

But that might not be what's happening.

At the Wall Street Journal (behind a paywall), Eran Bendavid and Jay Bhattacharya, two medical professors at Stanford, propose that we're using the wrong math and that we are still missing the numbers we need to do the math correctly.  However, by extrapolating from available data, one can argue that the coronavirus's mortality rate is significantly lower than the early estimates.

According to the doctors, "The true fatality rate is the portion of those infected who die, not the deaths from identified positive cases."  When calculating the mortality rate, while we know the numerator (the number who have died); we're using the wrong denominator.  If the denominator is only those sick enough to get the test in the first place, that small number will return a much higher mortality rate.

The real denominator should be the total number of people who catch this contagious virus.  Ideally, we'd already be testing everyone with an antibody test, which would allow us to know how many have already caught this new virus and recovered without incident.  However, while antibody tests are in the pipeline, they're not there yet.

Thankfully, there is information that helps us estimate the contagion factor.  In Italy, the scariest country of them all, the math says that while Italy is being slammed by the speed of the coronavirus's spread, the ultimate mortality rate won't be as devastating as the early numbers predict:

Next, the northeastern Italian town of Vò, near the provincial capital of Padua. On March 6, all 3,300 people of Vò were tested, and 90 were positive, a prevalence of 2.7%. Applying that prevalence to the whole province (population 955,000), which had 198 reported cases, suggests there were actually 26,000 infections at that time. That's more than 130-fold the number of actual reported cases. Since Italy's case fatality rate of 8% is estimated using the confirmed cases, the real fatality rate could in fact be closer to 0.06%.

The doctors do the same calculation for other defined populations, such as planes that evacuated U.S. citizens from Wuhan (all the passengers were tested), Iceland's population, and NBA players.  In the case of the Wuhan evacuees, only 0.9% of the Wuhan travelers tested positive.  Extrapolating that data out to Wuhan as a whole would mean that 178,000 Wuhan residents were probably infected, a number 30 times greater than the number China identified as having the virus.  Accepting as true China's claim that only 3,287 people died, the real mortality rate would be 1.85% — which is still a high mortality rate, but less than the original 4% claimed.

Considering that in December and January, there was substantial and steady traffic between Wuhan and the U.S., the doctors believe that the coronavirus's high infection rate means that some six million Americans may have been infected by March 9.  As of March 23, though, only 499 Americans had died from the coronavirus.

The professors point out that this is a mortality rate of only 0.01%, which is one tenth of the annual flu mortality rate of 0.1%.  If more people are infected than with the flu, but only 1/10 the number of people die...well, it's not Armageddon.  (For more "not Armageddon" info, check out John Hinderaker's post about virus fatalities.)

Drs. Bendavid and Bhattacharya suggest that the most sensible plan is to focus on vulnerable populations: the elderly and the sick, whether in hospitals, nursing homes, or their own homes.  This means isolating them and, because of the virus's rapid spread, temporarily reallocating hospital resources, such as elective procedures.

The doctors close by warning that a universal quarantine, with its massive impact on the economy and, as a byproduct, on Americans' physical and mental well-being, may well be overkill and must, therefore, be reevaluated.

It's a great article with more information than I've given here.  If you have access to The Wall Street Journal, try to read what the good doctors have to say.

The coronavirus is an icky disease that takes a cruel toll on the elderly, the sick, and the unlucky.  In this modern era, we can outwit many things that once routinely killed people, but the Grim Reaper is still out there, and he'll eventually get all of us.  Scary headlines have hinted that the coronavirus is now the Grim Reaper's preferred method.

Media reports have told us that the coronavirus is significantly more deadly than the flu, which annually kills 30,000 to 60,000 Americans.  Based on the speed with which it killed in China, Italy, Iran, and Spain, it looked as if the American death toll could easily top two million people annually.  While that's small potatoes compared to past pandemics (e.g., the Plague of Justinian, the Black Death, Spanish Influenza), it's a staggering toll in modern America.  Any actions seemed worthwhile to keep America from turning into a viral slaughterhouse.

But that might not be what's happening.

At the Wall Street Journal (behind a paywall), Eran Bendavid and Jay Bhattacharya, two medical professors at Stanford, propose that we're using the wrong math and that we are still missing the numbers we need to do the math correctly.  However, by extrapolating from available data, one can argue that the coronavirus's mortality rate is significantly lower than the early estimates.

According to the doctors, "The true fatality rate is the portion of those infected who die, not the deaths from identified positive cases."  When calculating the mortality rate, while we know the numerator (the number who have died); we're using the wrong denominator.  If the denominator is only those sick enough to get the test in the first place, that small number will return a much higher mortality rate.

The real denominator should be the total number of people who catch this contagious virus.  Ideally, we'd already be testing everyone with an antibody test, which would allow us to know how many have already caught this new virus and recovered without incident.  However, while antibody tests are in the pipeline, they're not there yet.

Thankfully, there is information that helps us estimate the contagion factor.  In Italy, the scariest country of them all, the math says that while Italy is being slammed by the speed of the coronavirus's spread, the ultimate mortality rate won't be as devastating as the early numbers predict:

Next, the northeastern Italian town of Vò, near the provincial capital of Padua. On March 6, all 3,300 people of Vò were tested, and 90 were positive, a prevalence of 2.7%. Applying that prevalence to the whole province (population 955,000), which had 198 reported cases, suggests there were actually 26,000 infections at that time. That's more than 130-fold the number of actual reported cases. Since Italy's case fatality rate of 8% is estimated using the confirmed cases, the real fatality rate could in fact be closer to 0.06%.

The doctors do the same calculation for other defined populations, such as planes that evacuated U.S. citizens from Wuhan (all the passengers were tested), Iceland's population, and NBA players.  In the case of the Wuhan evacuees, only 0.9% of the Wuhan travelers tested positive.  Extrapolating that data out to Wuhan as a whole would mean that 178,000 Wuhan residents were probably infected, a number 30 times greater than the number China identified as having the virus.  Accepting as true China's claim that only 3,287 people died, the real mortality rate would be 1.85% — which is still a high mortality rate, but less than the original 4% claimed.

Considering that in December and January, there was substantial and steady traffic between Wuhan and the U.S., the doctors believe that the coronavirus's high infection rate means that some six million Americans may have been infected by March 9.  As of March 23, though, only 499 Americans had died from the coronavirus.

The professors point out that this is a mortality rate of only 0.01%, which is one tenth of the annual flu mortality rate of 0.1%.  If more people are infected than with the flu, but only 1/10 the number of people die...well, it's not Armageddon.  (For more "not Armageddon" info, check out John Hinderaker's post about virus fatalities.)

Drs. Bendavid and Bhattacharya suggest that the most sensible plan is to focus on vulnerable populations: the elderly and the sick, whether in hospitals, nursing homes, or their own homes.  This means isolating them and, because of the virus's rapid spread, temporarily reallocating hospital resources, such as elective procedures.

The doctors close by warning that a universal quarantine, with its massive impact on the economy and, as a byproduct, on Americans' physical and mental well-being, may well be overkill and must, therefore, be reevaluated.

It's a great article with more information than I've given here.  If you have access to The Wall Street Journal, try to read what the good doctors have to say.