Robin Hanson: Manage a Pandemic by Controlled Infection?

Coronavirus particleRobin Hanson is a an economics professor at George Mason University who is known for thinking way outside the box.  His 2016 book, The Age of Em, which was my nonfiction book of the year for 2016, envisioned a future where the overwhelming majority of sentient beings were emulations (“ems”) of human brains running on digital hardware up to a million times faster than our current biochemical substrate.  He coined the term “Great Filter” to explain why we do not observe the universe to be teeming with advanced civilisations, and explored the implications of whether the filter is an event in the past or future of our own species’ history.

Yesterday, on his Overcoming Bias blog, he posted a provocative article titled “Consider Controlled Infection”.  Noting that if the present coronavirus outbreak (or some other in the future) spreads and becomes a global pandemic, so many people may be infected at once that social support and medical infrastructure (for example, intensive care facilities and support equipment such as ventilators) could be overwhelmed.

I see two big potential problems. One is that our medical systems have limited capacities, especially for intensive care. So if everyone gets sick in the same week, not only won’t the vast majority get much of help from hospitals, they may not even be able to get much help from each other, such as via feeding. Perhaps greatly increasing death rates. This problem might be cut if we spread out the infection out over time, so that different people were sick at different times.

The other related problem is where many non-sick people stay away from work to avoid getting sick. If enough people do this, especially at critical infrastructure jobs, then the whole economy may collapse. And not only is a collapsed economy bad for most everyone, sick people do much worse there. Not only can’t they get to a doctor or hospital, they might not even be able to get food or heating/cooling. Infected surfaces don’t get cleaned, and maybe even dead bodies don’t get removed. Thieves don’t get stopped. And so on. We can already see social support partially collapsing in Wuhan now, and it’s not pretty.

He then suggests that if we cannot halt the spread of the disease or provide anything other than supportive care for those who are infected, and on the assumption that recovery from the infection confers immunity against re-infection (which is the case for most such diseases), perhaps a strategy to minimise overall deaths and social disruption would be to deliberately expose groups of people who would be quarantined and provided supportive care until those who survived recovered.

There’s an obvious, if disturbing, solution here: controlled infection. We could not only insist that critical workers go to work, but we might also choose on purpose who gets exposed when. We can’t slow down infection very much, but we can speed it up a lot, via deliberately exposing particular people at particular times, according to a plan.

Such a plan shouldn’t just expose random people early, as they’d be likely to infect others around them. Instead, groups might be taken together to isolated places to be exposed, or maybe whole city blocks could be isolated and then exposed at once. Such a plan should only expose a small fraction of each critical workforce at any one time, so that most of them remain available to keep the lights on.

He goes on to explore how such a program of controlled infection might be managed, and what incentives could be offered to those willing to go early.  Read the whole thing.  What do you think of the idea?  Is it something which, like controlled burns to reduce the impact of wildfires, which we should consider in responding to an outbreak of a potential global pandemic?

8+
avataravataravataravataravataravataravataravatar

Author: John Walker

Founder of Ratburger.org, Autodesk, Inc., and Marinchip Systems. Author of The Hacker's Diet. Creator of www.fourmilab.ch.

42 thoughts on “Robin Hanson: Manage a Pandemic by Controlled Infection?”

  1. Well, first, we have to know whether infection confers immunity.  It didn’t with smallpox, right?  Some people survived one bout only to die with a second.

    (How would we know that, except by re-exposing one or more survivors? )

    Another thing not clear to me is, what kinda shape are infected active cases in?  Are they incapacitated?  I get that it’s highly contagious, but so were measles, mumps—and most of us survived those as kids.
    And another thing I havent been able to find out: what percentage of people  infected die?

    I was just reading about the Spanish flu post WWI.  Now, THAT was virulent: people were feeling fine, then suddenly struck down by agonizing pains in their legs, things like haemhorragic nosebleeds.  Is Coronavirus like that?  Or is it just a URI which  has a high risk of developing into pneumonia?

    When you get it, how long is the symptomatic phase?

    The author doesn’t know the answers to any of these questions.

    Let’s  put it this way: allowing the new virus to saturate the human population WILL result in stabilization of the outbreak .  It always does.  Eventually.  Additionally, viruses go through many generations in a day.  It will mutate.  And as it does it may become less virulent, like syphilis, which used to absolutely ravage the human body in a matter of months. And as I think has happened with AIDS.

    idk, so far this makes me think of the Flagellants during the Black Plague.  They knew the scourge was divine punishment, so they decided to just get it over with by punishing themselves.   I think there’s a whiff of the same desperation, Masquerading as practicality, in this piece.  Don’t be a baby!

    2+
    avataravatar
  2. John Walker:
    What do you think of the idea? Is it something which, like controlled burns to reduce the impact of wildfires…

    Well, that *was* what I though.  Like a burn-back.

    Hmmm, if saturation is inevitable, then CI is a better idea than UCI, for the individual calculus as well as group.  If it’s not inevitable, then the deal looks better for groups than for individuals — if you’re in (or think you are in) a high-risk group, then your local facilities are likely already struggling.  If you’re in (or think you are in) a low-risk group, then you’ll try to remain uninfected.

    3+
    avataravataravatar
  3. It’s unclear how Mr Hanson’s plan is helpful. He acknowledges that

    We can’t slow down infection very much, but we can speed it up a lot, via deliberately exposing particular people at particular times, according to a plan.

    If the concern is that medical facilities would be overwhelmed and critical functions would be impaired, how does additional burdening of medical facilities and impairment of critical functions help? All Mr Hanson’s concerns would only be exacerbated. Sure, the plan might result in having a small cadre of immune individuals (after a few unfortunate and unnecessary deaths) but all the other critical individuals would still be infected at the same time. As a bonus, additional medical facilities would be taken out of commission as the result of the controlled infection. The term “controlled infection” is misleading. There’s nothing controlled about it. A better one would be “additional, intentional infection.”

    Mr Hanson reasons by analogy:

    …as soon as one kid came down with an illness, parents would put the other kids in close contact, so they could all catch it at the same time. Because it was less trouble to care for all the kids in a family at once than to care for them one at a time.

    This reasoning does not apply to Hanson’s concerns. Having even more critical workers sick all at once does not alleviate the problem. Likewise, having even more ICU beds occupied at the same time does not alleviate the shortage of ICU beds. It makes the shortages more acute.

    Furthermore, Hanson extrapolates the current six-day doubling time for deaths (and new infections) indefinitely into the future. The following graph (from the link Hanson uses for his data) illustrates the trouble with his approach. Simple extrapolations are poor predictors. The doubling time has been decreasing, presumably because treatment has improved and spreading is better controlled as we learn more. Time is on our side; accelerating the infection’s spread is going in the wrong direction. Even if there is a reversal, Hanson’s policy prescription wouldn’t help.

    Given the poor predictive value of economic models, one could be forgiven for hoping that economists will be kept as far away as possible from decision-making on COVID19. Maybe extend that to all public policy.

    Hanson is getting savaged in the comments to his blog post, for good reason. As an aside, I noticed that one of the commenters mentioned Paul Ewald‘s ideas about infectious diseases, which should give pause to anyone with ideas of gratuitously infecting people with diseases.

    3+
    avataravataravatar
  4. drlorentz:
    If the concern is that medical facilities would be overwhelmed and critical functions would be impaired, how does additional burdening of medical facilities and impairment of critical functions help?

    I think he is making the assumption that, once you have survived coronavirus infection, that then subsequent infections will be mild, or at least less debilitating.   Which was Hypatia’s question above.   I don’t think we know yet whether this is true.

    2+
    avataravatar
  5. There’s some internet churn about deadly reinfection.  I’m dismissing this based on the few and poor common sources.

    3+
    avataravataravatar
  6. MJBubba:

    drlorentz:
    If the concern is that medical facilities would be overwhelmed and critical functions would be impaired, how does additional burdening of medical facilities and impairment of critical functions help?

    I think he is making the assumption that, once you have survived coronavirus infection, that then subsequent infections will be mild, or at least less debilitating.   Which was Hypatia’s question above.   I don’t think we know yet whether this is true.

    Even if that were true, I don’t see how it’s helpful to add more cases than would occur naturally. Recall that he also claims the infection’s spread can’t be slowed down very much, viz. the first passage I quoted above. Hence, the number of cases would be even greater under his plan, thereby burdening the healthcare system more. Deadly reinfection would just be a bonus.

    2+
    avataravatar
  7. drlorentz:
    Even if that were true, I don’t see how it’s helpful to add more cases than would occur naturally.

    As I read the article, I think he’s assuming that the infection will grow exponentially and eventually infect everybody on the planet who does not have some kind of natural immunity to the disease (and they, of course, would be unaffected by the controlled infection).  If you assume this kind of saturation infection, then the question is whether the disruption would be greater if most of the infections and acute illness happens effectively at the same time (which is a consequence of exponential growth), which would completely overload the health care resources even not considering how many workers in the sector would themselves be incapacitated by the infection or, accepting that everybody is eventually going to be infected anyway (and note that he does say that may not happen if we find a vaccine or effective treatment), to space out the infections and illness over a period of time sufficiently long that the health care system can cope with the number of acute cases at any given time during the period.

    Given how little is known about the details of this disease, and estimates for infection multiplier and fatality rate all over the map, any argument about a strategy has to assume many facts not in evidence.

    5+
    avataravataravataravataravatar
  8. G.D.:
    Wasn’t that discussed in one of the books about “The Iron Dragon”?

    Yes, in The Dawn of the Iron Dragon, volume 2 of the trilogy, the spacemen deliberately spread a less virulent version of the plague which the Cho-ta’an have developed to exterminate the population of Europe.  The version they spread is still deadly, but it “only” kills a third or so of those infected, rather than the 100% deadly weaponised version bio-engineered by the Cho-ta’an.  The weaker plague, however, makes survivors immune to the weaponised plague, so by spreading it they are actually saving many lives who would be killed if the Cho-ta’an plague reached them first.

    This isn’t exactly what Hanson is proposing, but both count upon those who survive the deliberate infection developing immunity to re-infection when the virus reaches the exponential blow-off phase of infection around the globe.

    4+
    avataravataravataravatar
  9. I just read Midnight in Chernobyl, which, among much other information, revealed the astounding dishonesty of officials at every level of the Soviet Union. By analogy, I therefore believe nothing in the media, virtually all of which emanates from the Chinese Communist Party. Accordingly, I think we know virtually nothing as to the kinetics of this infection.

    I believe that in principle, given known kinetics of some particular impending pandemic, the idea has merit – the threshold question being: is the infectivity rate and doubling time such that a pandemic is certain. If so, then one might have a basis to model the kinetics of intentional infections of selected individuals at a series of separated locations beyond the event horizon of the pandemic. All this, is, as has been pointed out, completely dependent upon whether or not survival confers reliable immunity. Absent this set of ascertained facts, for this bug, the idea is a non-starter.

    For myself, I am definitely worried that the population of mechanical ventilators in hospitals may be overwhelmed, even if there are enough people well to run them. In this event, expect triage. Old folks like me & my wife will be left to die by unappealable decisions in local hospitals (expect Bernie’s minions to be venting their notions of social justice “helping” to make such decisions in hospitals, at least temporarily). I have been planning to find ways to care for my family. I have stocked up on foods & potentially-helpful medicines (potent steroids, antibiotics, IV fluids& administration equipment, several general anti-virals, etc). I am also in the process of acquiring either oxygen-concentrating machine(s) and or large tanks of oxygen – which, in itself, can be life-saving in incipient respiratory failure (the usual cause of death). As well, I may well purchase a bi-pap machine (used for sleep apnea) and fit it with the supplemental oxygen. This provides a non-invasive form of assisted ventilation (=moving air in and out of lungs which are stiffened so as to assist muscles which are weakened), without having to place an endotracheal tube, which would be intolerable absent heavy sedation (part of the skills needed to ventilate people in the hospital [and not kill them in the effort]). I’ll try to update you all if and when this all comes together. For me, this is round three of disaster preparedness. 1. Y2K, 2. 9/11 3. whatever this actually turns out to be. [passionate typing – no time to proof, sorry for typos and/or general stupidity of impending decrepitude and of end-stage American 20-21st century paranoia of deplorability].

    4+
    avataravataravataravatar
  10. civil westman:
    I just read Midnight in Chernobyl, which, among much other information, revealed the astounding dishonesty of officials at every level of the Soviet Union. By analogy, I therefore believe nothing in the media, virtually all of which emanates from the Chinese Communist Party. Accordingly, I think we know virtually nothing as to the kinetics of this infection.

    This is true but data from countries besides China are beginning to emerge. Though there are reasons to be skeptical of the credibility of any government data, I’m giving other governments the benefit of the doubt. The current doubling time for confirmed cases outside China is 6.5 days, in agreement with Hansen’s value for deaths. Time will tell if this trend persists.

    source: Johns Hopkins CSSE

    2+
    avataravatar
  11. civil westman:
    I have been planning to find ways to care for my family. I have stocked up on foods & potentially-helpful medicines (potent steroids, antibiotics, IV fluids& administration equipment, several general anti-virals, etc).

    Everyone should be thinking along these lines because when it all hits the fan, it’ll be too late. The level of preparedness will differ according to one’s circumstances but the answer should never be no preparation.

    3+
    avataravataravatar
  12. drlorentz:

    civil westman:
    I just read Midnight in Chernobyl, which, among much other information, revealed the astounding dishonesty of officials at every level of the Soviet Union. By analogy, I therefore believe nothing in the media, virtually all of which emanates from the Chinese Communist Party. Accordingly, I think we know virtually nothing as to the kinetics of this infection.

    This is true but data from countries besides China are beginning to emerge. Though there are reasons to be skeptical of the credibility of any government data, I’m giving other governments the benefit of the doubt. The current doubling time for confirmed cases outside China is 6.5 days, in agreement with Hansen’s value for deaths. Time will tell if this trend persists.

    source: Johns Hopkins CSSE

    God Bless Log/Lin.

    2+
    avataravatar
  13. Actually, when you think about it, this is exactly what vaccination with attenuated stock does – and in a sense ALL vaccine. You undergo the process of developing resistance by exposure – to live, attenuated organisms or dead particles that are immunologically active. So this whole theory isn’t all that “out of the box”. Many people undergo mild forms of the “disease” when they receive a vaccine.

    I, for one, always got the flu when I got the vaccine, but in a lessor form. For a while, since I worked in an ED, so was exposed to all manner of bugs and never got sick for very long, decided to forego the vaccine and take my chances on getting the flu. Worked until the hospital Nazis required everyone to get the vaccine every year. BUT the up side has been that I have not gotten a “reaction” to the vaccine for quite a few years. Perhaps that was partly the effect of being in the military, where you are vaccinated for everything known or unknown to man. I believe I was the only guy in the ED that had been vaccinated for the plague when 09/11 hit and we were working out contingency plans for a possible bio weapon attack on the city.

    2+
    avataravatar
  14. So Much for that idea:

     

    https://www.zerohedge.com/health/hubei-doctors-warn-even-deadlier-coronavirus-reinfection-causing-sudden-heart-attacks

     

    It’s highly possible to get infected a second time. A few people recovered from the first time by their own immune system, but the meds they use are damaging their heart tissue, and when they get it the second time, the antibody doesn’t help but makes it worse, and they die a sudden death from heart failure,” reads a message forwarded to Taiwan News from a relative of one of the doctors living in the United Kingdom.

    The source also said the virus has “outsmarted all of us,” as it can hide symptoms for up to 24 days. This assertion has been made independently elsewhere, with Chinese pulmonologist Zhong Nanshan (鍾南山) saying the average incubation period is three days, but it can take as little as one day and up to 24 days to develop symptoms.

     

    Also, the source said that false negative tests for the virus are fairly common. “It can fool the test kit – there were cases that they found, the CT scan shows both lungs are fully infected but the test came back negative four times. The fifth test came back positive.” –Taiwan Times

    Notably, one of the ways coronaviruses cripple the immune system is via an HIV-like attachment to white blood cells, which triggers a ‘cytokine storm‘ – a term popularized during the avian H5N1 influenza outbreak – in which an uncontrolled release of inflammatory ‘cytokines’ target various organs, often leading to failure and in many cases death.

    The cytokine storm is best exemplified by severe lung infections, in which local inflammation spills over into the systemic circulation, producing systemic sepsis, as defined by persistent hypotension, hyper- or hypothermia, leukocytosis or leukopenia, and often thrombocytopenia.

    In addition to lung infections, the cytokine storm is a consequence of severe infections in the gastrointestinal tract, urinary tract, central nervous system, skin, joint spaces, and other sites. (Tisoncik, et. al, Into the Eye of the Cytokine Storm)(2012)

    2+
    avataravatar
  15. Haakon Dahl:
    I don’t find ZH credible, or even comfortable.  Maybe a picture will beat a thousand words:

    I don’t have quite as low an opinion of Zero Hedge as that but the graphic is really funny and not far off the mark.

    5+
    avataravataravataravataravatar
  16. If, indeed, infection and survival does not confer immunity, and re-infection causes higher mortality, we are in a new situation compared to all infectious diseases in human history.

    3+
    avataravataravatar
  17. John Walker:

    If, indeed, infection and survival does not confer immunity, and re-infection causes higher mortality, we are in a new situation compared to all infectious diseases in human history.

    I read this out to Snooks, and she pointed out that pneumonia is like that.  It weakens your lungs so that if you get it again, it is worse.

    4+
    avataravataravataravatar
  18. MJBubba, pneumonia may be worse after the first time, but it is not contagious.

    It is extremely rare for recovering from a virus to not give at least some temporary immunity. In fact, that is what it means to recover: for a period your immune system wins against the infection. Typically immunity is more than a few months, and that is enough to make it worth spreading out who gets infected over a similar time period.

    3+
    avataravataravatar
  19. Robin Hanson:
    drlorentz complained that I tried to extrapolate early doubling times, but that rate has in fact continued to apply to non-China cases for the last six weeks

    We agree. Please see comment #13, when I made this exact point ten days ago.

    I don’t understand his/her claims that spreading out cases over time doesn’t free up medical resources.

    Obviously it would. Please cite the passage where I made such a claim. I think you’ll find I was saying something quite different.

    0

  20. John Walker:

    If, indeed, infection and survival does not confer immunity, and re-infection causes higher mortality, we are in a new situation compared to all infectious diseases in human history.

    I thought the reinfection was getting hit too soon after your recovery so the body overreacts. I think SARS has been known to do this.

    1+
    avatar

Leave a Reply