If COVID-19 really is only a bit more lethal than the seasonal flu, the benefits that can be expected from continued lockdowns, in terms of deaths prevented, are much lower than most projections assumed. If these results are confirmed, they should play an important role in discussions about when and how to reopen the economy.Suppose it turns out to be only as lethal as the seasonal flu.
Monday, April 20, 2020
Learning more …
… L.A. County Antibody Tests Suggest the Fatality Rate for COVID-19 Is Much Lower Than People Feared – Reason.com. (Hat tip, Dave Lull.)
Subscribe to:
Post Comments (Atom)
At over 41000 dead, it has already eclipsed most combined-seasonal flus. Here in Massachusetts, yesterday was out worst day yet. We've got more lives to sacrifice. Nationwide, 25,000 have died in the last 10 days. This killer is not done by any stretch.
ReplyDeleteNo expert is expecting the old IHME model number of 60,000 to be realistic at this point. No one who is behind that model is saying that it ought to be used as an accurate predictor. It's a handy-dandy tool, the same one that predicted upwards of 240,000. The reliability is not near good enough to make it a life-and-death decision-making model.
We can suppose that if we all keep our distance, wash hands, wear masks, and just stay put, it could theoretically just stop spreading in time -- and we do not know yet what that timing. If we all did that, the virus would kill only the people who have already caught it, because it would be prevented from spreading to more of us.
The 41000 already dead are extra dead, a nuclear bomb killing everyone in a small-sized city, more than Pearl Harbor, more than 9/11. When compared to the 4 million people born each year, and over 3 million who die in this country, it may seem like a drop in the bucket. We lost just under 60,000 to the entire Viet Nam Conflict. Covid-19 will eclipse that in just half a year easy. By now, April 20, 40% of us know someone who has died from Covid-19, who would have been around decades longer if not for the virus, if only we were all more careful and planned better. This number of those gone-too-early is rising daily.
We don;t know how to stop the virus exactly. We do know a lot about how to spread it, though.
According to the CDC, the mortality rate for the flu season just ended is estimated to have been 62,000. Of the 42,374 U.S. fatalities from COVID-19 (according to Johns Hopkins), more than 20,000 have been in New York City and its environs (including northern New Jersey). During the 2017-2018 flu season in the U.S., 80,000 people died. If the mortality rate from COVID-19 ends up being in the range of those two flu seasons, should we then plan to shut down the country again come Oct. 1?
DeleteFrank, again I urge you to read the article more carefully. The USC study was designed to get a better estimate of case fatality rate: the percentage of deaths among people WHO HAVE BEEN INFECTED with the virus. Total mortality -- how many people this virus will kill -- is the case fatality rate multiplied by the number of infected people. Each is equally important in determining the final death toll. Remember the explanation I gave in an earlier comment about the huge difference in infectivity between the influenza virus and SARS-COV-2? That difference will be critical.
ReplyDeleteWe already know that the final mortality count for COVID-19 will be much higher than the average seasonal flu toll. Only seven weeks into this outbreak in the U.S., we've already lost over 42,000 people. The highest death total for any U.S. seasonal flu outbreak during this century is 62,000 -- over an entire year. Asian flu in 1957 and Hong Kong flu in 1968 each killed 100,000 over the course of a full year. At the current growth rate, I fear that COVID-19 mortality will surpass all of those.
People seem to believe that, someday soon, this epidemic will suddenly, magically end. How would that happen? How many of these people have even an inkling about how person-to-person disease outbreaks subside?
The answer is simple: population-level immunity. And there are only two ways to get there. The first is for enough people to have been infected, and then recovered, to decrease the transmission rate (R-naught) to below 1.0. Typically, that requires 60% of the population to have been infected. The USC study quoted in the article to which you linked suggests that we're at a 4% infection level. That's a hell of a long way from 60%. Moreover, again using figures from this article, we'd have to sacrifice 196,000-392,000 people to reach this "herd immunity" level. (Simple math using a population of 327,000,000;
the empirical 60% herd immunity value; and the estimated case fatality rate of 0.1-0.2% mentioned in the article.
The other route to population-level immunity is vaccination. There will be no widely available vaccine for at least 18 months. That's just the way it works. Full stop. So, I have a suggestion for all the whiners and snowflakes on the right: man up and deal with the inconveniences of social distancing, self-quarantine if you've got even minor symptoms, and mask-wearing (for the benefit of others, not yourself), for the next year-and-a-half. Only then can we get the economy back in action. All freedoms come with corresponding obligations. If these chumps acted responsibly on their own, we wouldn't need heavy-handed government actions.
Hi Jeff,
ReplyDeleteI checked the CDC and they have 80,000 dying of the flu in the 2017-2018 season. And outside New York City and its environs the mortality rate in the U.S. is just over 20,000. Whoever wrote the headline and deck of the article seems to have understood it in much the same way I did. Anders Tegnell, Sweden's chief epidemiologist, seems to think that Stockholm may have built up a good measure of population immunity by the middle of May. (Vaccination, as you know, is not the only way to build up herd immunity.) The British journal Nature just published a piece suggesting that "coronavirus infections vastly exceed official counts." And the British Medical Journal reported on an admittedly small study that indicated that as many as 78 percent of those infected with the virus display no symptoms, which I suppose is the same as saying that they don't get sick. So there is a lot of information available that runs counter to the official view of this matter and a good many qualified individuals (e.g., board certified physicians, professors of epidemiology) who have voiced skepticism. Are their views simply not to be taken into account? It isn't just chumps who are skeptical, and this chump does take commonsense precautions and has been out and about from the start. Of course, I've also never had the flu. So I could just be one of those legendary exceptions that proves the rule. I do think that if, when this is past, the numbers used to support the lockdown do not hold up, many will have a lot of — as Ricky Ricardo used to say — splaining to do.
Clearly, I understand that vaccination is not the only route to population level immunity -- I explained the other route in my comment. Was the explanation inadequate? My epidemiology professors taught me that herd immunity requires 60% of the population to have been infected. Is Sweden anywhere near 60%? Please provide a reference to the actual level of infection in that country. Your phrase "a good measure" has no scientific meaning. And I hope to God that you're not making the elementary logical error of thinking that 78% of infected people showing no symptoms means that 78% of the population has been infected.
ReplyDeletePlease explain the relevance of the distribution of cases inside and outside of the New York metropolitan area, and why you think that makes a difference. My explanation is that viruses spread faster in dense populations, meaning that illnesses and deaths accumulate more quickly. But I don't conflate speed with final extent. My epidemiology professors taught me that, without mitigation measures, the final rate of infection will be the same everywhere; it just happens faster in densely populated areas. What's your explanation?
I've asked three questions in the preceding two paragraphs. They were not rhetorical, and I'd appreciate answers.
I form my own hypotheses after considering solid data from as many sources as I can find, using my training and experience as a professional biologist, including two years working in a human virology lab and a year of graduate-level courses in epidemiology. I really don't give a shit about what you call "the official view". And I'm completely willing go back and try to understand where I might have made a conceptual mistake if events prove me wrong. That's what scientists do. More importantly, that's what honorable people do. Are you equally willing? Just three weeks ago, you predicted it would take five more months to reach 30,000 total COVID-19 deaths in the U.S. We're at 42,000 now.
For me it's not about being right or wrong. It's about thinking clearly. About seeing with eyes unclouded by ideological biases or wishful thinking, and reasoning with a mind uncontaminated by conspiracy theories or knee-jerk contrarianism.
(Incidentally, my comments about right-wing whiners and chumps were not directed at you. If you want to identify with those folks, I guess it's your choice. Personally, I consider ideology to be the intellectual equivalent of training wheels for a bicycle. As soon as I was coordinated enough to ride without training wheels, I ditched them and left paved paths behind. As soon as I possessed the mental agility to think independently, I ditched ideology as well, and sought the unpaved intellectual paths.
Oh, I was just joking about the chumps — though some emails I have received indicate your reference thereunto was not taken kindly by some. I always like to identify with chumps. So much better than thinking too well of yourself. And worse, of putting on airs. I was recently asked by a doctor treating wife if I had ever had any episodes of depression. He broke out laughing when I told him, "Me? No. I'm shallow."
ReplyDeleteThe folks who drive around honking their horns, waving flags and screaming about liberty imagine themselves to be patriots. The real patriots are volunteering at food banks, or running errands and shopping for the vulnerable. I was fortunate to be raised by the real deal, a man whose valor during WWII earned him burial at Arlington National Cemetery with full honors -- horse-drawn caisson, riderless horse, seven-man rifle squad, seven-man burial team, bugler. He's the man who taught me how to distinguish between true patriots and loudmouth chumps.
ReplyDeleteA layperson's question: if so many people are asymptomatic, what has caused the sudden 'explosion' of symptomatic cases, and deaths? Jeff?
ReplyDeleteHey Lee,
ReplyDeleteThe short answer is that an enormous portion of the earth's population remains as yet uninfected, and it's this group that the epidemic is expanding into.
Let's just focus on the USC study of LA County summarized in the piece to which Frank linked. Through antibody testing of a random population sample, the researchers concluded that 4% of the county's population already had been infected, knowingly or not. Given LA County's population of 10 million, that means about 400,000 have been infected and 9.6 million remain as yet uninfected. Imagine a pie from which a slice representing 1/25 of the total has been removed. That leaves a whole lotta pie still to be eaten.
Now let's focus on the 400,000 who probably have been infected. Most studies I've seen estimate that 50% of infected people are asymptomatic. But it's important to remember that even asymptomatic people can spread the virus. How can that happen? Well, if I had the virus right now, I'd be a super-spreader, even if I was asymptomatic with no fever or cough. I'd still be secreting the virus into my mucus membranes, and spreading it widely with the frequent sneezing caused by my allergy to tree pollen. That's why I wear a mask when in a public space. There are dozens of other ways that infected, but asymptomatic, people can spread the virus.
The USC paper also estimated a case fatality rate of 0.1-0.2% (case fatality rate is the percentage of infected people who die). If the number of infected people in LA County is really 400,000, then the estimated case fatality rate suggests that 400-800 people would have died to this point. The actual number, as of today, is 666.
Which brings me to the theory that Frank brings up: herd immunity. Development of herd, or population, immunity requires letting the virus move freely through the population, with the infection rate slowing down as the virus encounters ever greater
difficulty finding as yet uninfected hosts. Generally, the epidemic burns itself out once the population infection level exceeds 60%. In LA County, that would require 6 million people to become infected, with 6,000-12,000 of them dying (based on the USC case fatality rate of 0.1-0.2%).
I hope this helps.
In short, then: the fact that so many people seem to be asymptomatic means that Covid-19 is more dangerous, not less so.
ReplyDelete