Current estimates about the Covid-19 fatality rate may be too high by orders of magnitude.

By Eran Bendavid and Jay Bhattacharya

March 24, 2020 6:21 pm ET

SAVE

SHARE

TEXT

338

A line at an emergency room in Brooklyn, N.Y., March 19.

PHOTO: ANDREW KELLY/REUTERS

If it’s true that the novel coronavirus would kill millions without shelter-in-place orders and quarantines, then the extraordinary measures being carried out in cities and states around the country are surely justified. But there’s little evidence to confirm that premise—and projections of the death toll could plausibly be orders of magnitude too high.

Fear of Covid-19 is based on its high estimated case fatality rate—2% to 4% of people with confirmed Covid-19 have died, according to the World Health Organization and others. So if 100 million Americans ultimately get the disease, two million to four million could die. We believe that estimate is deeply flawed. The true fatality rate is the portion of those infected who die, not the deaths from identified positive cases.

Why the Coronavirus Relief Bill Stalled

00:00 / 23:44

SUBSCRIBE

The latter rate is misleading because of selection bias in testing. The degree of bias is uncertain because available data are limited. But it could make the difference between an epidemic that kills 20,000 and one that kills two million. If the number of actual infections is much larger than the number of cases—orders of magnitude larger—then the true fatality rate is much lower as well. That’s not only plausible but likely based on what we know so far.

Population samples from China, Italy, Iceland and the U.S. provide relevant evidence. On or around Jan. 31, countries sent planes to evacuate citizens from Wuhan, China. When those planes landed, the passengers were tested for Covid-19 and quarantined. After 14 days, the percentage who tested positive was 0.9%. If this was the prevalence in the greater Wuhan area on Jan. 31, then, with a population of about 20 million, greater Wuhan had 178,000 infections, about 30-fold more than the number of reported cases. The fatality rate, then, would be at least 10-fold lower than estimates based on reported cases.

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%.

In Iceland, deCode Genetics is working with the government to perform widespread testing. In a sample of nearly 2,000 entirely asymptomatic people, researchers estimated disease prevalence of just over 1%. Iceland’s first case was reported on Feb. 28, weeks behind the U.S. It’s plausible that the proportion of the U.S. population that has been infected is double, triple or even 10 times as high as the estimates from Iceland. That also implies a dramatically lower fatality rate.

The best (albeit very weak) evidence in the U.S. comes from the National Basketball Association. Between March 11 and 19, a substantial number of NBA players and teams received testing. By March 19, 10 out of 450 rostered players were positive. Since not everyone was tested, that represents a lower bound on the prevalence of 2.2%. The NBA isn’t a representative population, and contact among players might have facilitated transmission. But if we extend that lower-bound assumption to cities with NBA teams (population 45 million), we get at least 990,000 infections in the U.S. The number of cases reported on March 19 in the U.S. was 13,677, more than 72-fold lower. These numbers imply a fatality rate from Covid-19 orders of magnitude smaller than it appears.

How can we reconcile these estimates with the epidemiological models? First, the test used to identify cases doesn’t catch people who were infected and recovered. Second, testing rates were woefully low for a long time and typically reserved for the severely ill. Together, these facts imply that the confirmed cases are likely orders of magnitude less than the true number of infections. Epidemiological modelers haven’t adequately adapted their estimates to account for these factors.

The epidemic started in China sometime in November or December. The first confirmed U.S. cases included a person who traveled from Wuhan on Jan. 15, and it is likely that the virus entered before that: Tens of thousands of people traveled from Wuhan to the U.S. in December. Existing evidence suggests that the virus is highly transmissible and that the number of infections doubles roughly every three days. An epidemic seed on Jan. 1 implies that by March 9 about six million people in the U.S. would have been infected. As of March 23, according to the Centers for Disease Control and Prevention, there were 499 Covid-19 deaths in the U.S. If our surmise of six million cases is accurate, that’s a mortality rate of 0.01%, assuming a two week lag between infection and death. This is one-tenth of the flu mortality rate of 0.1%. Such a low death rate would be cause for optimism.

This does not make Covid-19 a nonissue. The daily reports from Italy and across the U.S. show real struggles and overwhelmed health systems. But a 20,000- or 40,000-death epidemic is a far less severe problem than one that kills two million. Given the enormous consequences of decisions around Covid-19 response, getting clear data to guide decisions now is critical. We don’t know the true infection rate in the U.S. Antibody testing of representative samples to measure disease prevalence (including the recovered) is crucial. Nearly every day a new lab gets approval for antibody testing, so population testing using this technology is now feasible.

If we’re right about the limited scale of the epidemic, then measures focused on older populations and hospitals are sensible. Elective procedures will need to be rescheduled. Hospital resources will need to be reallocated to care for critically ill patients. Triage will need to improve. And policy makers will need to focus on reducing risks for older adults and people with underlying medical conditions.

A universal quarantine may not be worth the costs it imposes on the economy, community and individual mental and physical health. We should undertake immediate steps to evaluate the empirical basis of the current lockdowns.

Dr. Bendavid and Dr. Bhattacharya are professors of medicine at Stanford. Neeraj Sood contributed to this article

(http://www.autoadmit.com/thread.php?thread_id=4486572&forum_id=2#39856493)

heck, trump already knows all this. not to mention there is a cure already in sight, it just has to be approved and shipped.

(http://www.autoadmit.com/thread.php?thread_id=4486572&forum_id=2#39856523)

NO SHIT

(http://www.autoadmit.com/thread.php?thread_id=4486572&forum_id=2#39856546)

(http://www.autoadmit.com/thread.php?thread_id=4486572&forum_id=2#39856583)

JFC, in less than a month 1% of the entire country was infected.

(http://www.autoadmit.com/thread.php?thread_id=4486572&forum_id=2#39856560)

plus, the Dems are going to push the story because it helps them...and the media will go along to some extent...

but I LOVE the stock market crash, and I LOVE the fulltime telework, so I hope the scary, but likely false, propaganda keeps getting pushed....

(http://www.autoadmit.com/thread.php?thread_id=4486572&forum_id=2#39856565)

It's almost like that's a better number than using the sample of 70 year old people showing up at the er coughing up blood.

(http://www.autoadmit.com/thread.php?thread_id=4486572&forum_id=2#39856569)

When people are old and immunocompromised, it's easier for them to catch anything.

This was posted earlier, quoting an advisor to italys minister of health.

www.telegraph.co.uk/global-health/science-and-disease/have-many-coronavirus-patients-died-italy/amp/

(http://www.autoadmit.com/thread.php?thread_id=4486572&forum_id=2#39856674)

about 10% it looks like. given how readily the illness spreads absent controls, and limitations on medical resources at any given time, i think you can see how that could easily become a big problem right.

(http://www.autoadmit.com/thread.php?thread_id=4486572&forum_id=2#39856690)

As i recall the average age of the passengers was really high

(http://www.autoadmit.com/thread.php?thread_id=4486572&forum_id=2#39856590)

thats the crux of the whole article

if you assume 20% of infected require some kind of hospitalization then we need to shut down for a couple of months

but the article is showing that real number of infections is like 60x of confirmed cases in which hospitalization rate is only 1% and can be easily handled except for hotspots like NYC.

basically it all comes down to this. we have 55K confirmed cases. how many were totally infected? MSM hysteria assumes only around 400K~600K were infected. this guys is saying its like 4M~6M already

(http://www.autoadmit.com/thread.php?thread_id=4486572&forum_id=2#39856701)

being very conservative and assuming we have captured only 25% of covid deaths by now (maybe lot covid deaths of old ppl in jan/early feb were never tested for covid and thought of as regular flu???)

so roughly 7M cases and 3500 deaths by very conservative estimate gives us 0.05% mortality rate

iceland study is very interesting. wonder if that holds true for US

(http://www.autoadmit.com/thread.php?thread_id=4486572&forum_id=2#39856651)

eg https://www.nature.com/articles/s41591-020-0822-7

Because both of your examples are extremely consistent with the lower estimates of eg 0.5-0.8% - in SK, yes they did widespread testing but there's still selection bias, ie the selection mechanism into getting tested is based on more severe cases / symptoms, - because if you're completely asymptomatic or very mild symptoms, you don't have an incentive to go get medical help like you do if you're very severe, so the testing , although much more widespread and so lower than the 3% cfr elsewhere, is still missing asymptomatic / very mild cases.

The diamond princess cruise ship - ~1% sCFR , is actually cause for optimism, because this group of people skews so overwhelmingly older on the ship, 1% is less than the estimates for CFR of eg 65+ of something like 1.4% in that paper I linked above.

All of those examples we see observationally are still biasing the CFR upwards whether due to age group of selection into testing, see eg the paper linked above or

(http://www.autoadmit.com/thread.php?thread_id=4486572&forum_id=2#39863259)

have it and the mass deaths are solely due to hospital overcrowded ness and not disease potency?

(http://www.autoadmit.com/thread.php?thread_id=4486572&forum_id=2#39856882)

i think the culprit is mostly that it spread much much faster in northern italy than anywhere except maybe wuhan or NYC. basically they have an outlier R0 due to large chinese population, social customs like hugging/kissing, multiple generations living together

so R0 of italy is like 4 or 5 vs 2 everywhere. this overwhelmed their hospital capacity plus their older smoking demographic has lead to 8% death rate among confirmed cases

i wont be shocked if like 20% of population of northern italy has already been infected. NYC is heading that way

(http://www.autoadmit.com/thread.php?thread_id=4486572&forum_id=2#39856966)

I'll try to dig it up, I saw someone on twitter post an analysis of the time series of deaths in Italy every March, and as we know an overwhelming proportion of those deaths are people with comorbidities and 80+ (literally only 3 people, 0.8%, in the study linked on bloomberg had no previous disease ,

"Almost half of the victims suffered from at least three prior illnesses and about a fourth had either one or two previous conditions.

More than 75% had high blood pressure, about 35% had diabetes and a third suffered from heart disease.")

And this analysis (i'll try to find it again) showed that Italy's deaths for March really aren't anomalous , which is a suggestion that maybe it's just moving cause of death from column A to column B, consistent with recent prevalence estimates and other analyses about how hospitals are huge vectors for these infections...

(http://www.autoadmit.com/thread.php?thread_id=4486572&forum_id=2#39863244)

http://www.xoxohth.com/thread.php?thread_id=4479648&mc=53&forum_id=2#top

(http://www.autoadmit.com/thread.php?thread_id=4486572&forum_id=2#39857144)

The China numbers really make no sense to me. How did it create so much localized death and destruction, and then just stop spreading almost completely?

(http://www.autoadmit.com/thread.php?thread_id=4486572&forum_id=2#39857465)