Astral Codex Ten

Russian Flu

15 min read

Russian Flu

Article

Russian Flu is a recurring concept in the Astral Codex Ten archive, appearing 2 times across 2 issues between July 30, 2022 and January 01, 2025. The archive places it in contexts such as “1977 influenza pandemic, now called the Russian Flu”; “An H1N1 flu (“Russian flu”) leaked from a biology lab (?) in Russia”. It most often appears alongside Canada, China, US.

Metadata

  • Category: Concepts
  • Mention count: 2
  • Issue count: 2
  • First seen: July 30, 2022
  • Last seen: January 01, 2025

Appears In

Source Context

Recovered passages from the original issue text. When the raw archive preserved outbound links inside the source passage, they are listed directly under the quote.

Your Book Review: Viral
July 30, 2022 · Original source
In 1977 there was an influenza pandemic, now called the Russian Flu, which ultimately killed about 700,000 people. It was discovered that this flu strain was nearly genetically identical to a strain that had previously been common in the 1950s, but had since disappeared, except for samples that were being studied in research labs. It’s now thought that the 1977 Russian Flu was the result of vaccine trials gone awry, in which military recruits became infected after being exposed to live attenuated H1N1 virus.
H5N1: Much More Than You Wanted To Know
January 01, 2025 · Original source
From “Genesis and pathogenesis of the 1918 pandemic H1N1 influenza A virus”, linked above. You may recognize the lead author - Michael Worobey has also been a leading voice on the zoonotic side of the COVID origins debate. The recent history of the flu, as far as I can tell, is: 1918: An H1N1 flu (“Spanish flu”) jumped from birds to humans in America and killed 50 million people worldwide. This replaced all older strains, so most seasonal flus during this era were H1N1. 1957: An H2N2 flu (“Asian flu”) crossed from birds to humans in China, and killed about 2 million people worldwide. It replaced the H1N1 strain, so most seasonal flus during this era were H2N2. 1968: An H3N2 flu (“Hong Kong flu”) crossed from pigs (?) to humans in Hong Kong, and killed another 2 million people worldwide. It replaced the H2N2 strain, so most seasonal flus during this era were H3N2. 1977: An H1N1 flu (“Russian flu”) leaked from a biology lab (?) in Russia (it might have been a strain from the 1940s, which the Russians were trying to make a vaccine for). It didn’t kill that many people, but it stuck around, and from then on, seasonal flus could be either H3N2 or H1N1. 2009: An H1N1 flu (“Mexican flu” until the PC police stepped in; afterwards “swine flu”) took some horrible circuitous route between birds and pigs and back again, crossed over into humans in Mexico, and killed 200,000 people. It outcompeted older strains of H1N1, but couldn’t crowd out H3N2, so seasonal flus are still either H3N2 or H1N1. …which brings us to the present, hopefully illuminating why “new flu strain crosses over from animals into humans” is such an “uh oh” moment. The Bird Flu Technically, all pandemic flus start as bird flus. Influenza A evolved in birds. Sometimes it spreads to other animals, including pigs, cattle, and humans. The most common way for a bird flu to spread to humans is to “reassort” (not exactly virus sex, but close enough, and the real version is less memorable) with a human flu virus (ie one that has already crossed over to humans). The resulting virus has all of the human flu virus’ human adaptations, but borrows enough new antigens from the bird virus to evade the immune system. Pigs can be infected by both human and bird viruses, so they are a common place for this reassortment to take place. If reassortment is sort of like viral sex, pigs are sort of like Tinder. When a bird flu and human flu reassort in pigs, the resulting disease is called a swine flu. At least the 2009 flu pandemic was a swine flu, and a minority opinion thinks the 1918 pandemic was too. There aren’t major epidemiological differences between direct-from-bird flus and swine flus. H5N1 was first noticed in birds - specifically, a flock of chickens in Scotland in 1959 - after which it disappeared for forty years. In 1996, it showed up in geese in China, then gradually increased its market share among birds worldwide. In 2022, it was found in minks; apparently it had learned to infect mammals. By early 2024, it was seen in cows. Now it’s in cow herds in 16 states, and one of them (California) has declared a state of emergency. And in October, H5N1 was found in pigs for the first time. It’s not uncommon for humans to catch an animal disease. This doesn’t mean the disease has “crossed over” to humans. If the virus isn’t suited to human-to-human transmission, it simply dies off (either before or after killing its human host). Thus, chicken farmers have been reporting scattered H5N1 cases since 1997; now that the virus has spread to cattle, cow farmers have started reporting the same. A Metaculus comment on this topic introduced me to the phrase “biocomputational surface”. Every viral replication that takes place in a human gives the virus one more chance to develop the set of mutations that makes it human-transmissible and start the next pandemic. Or, more likely, every viral replication that takes place in a human who has both the H5N1 bird flu and a normal human flu - or in a pig which has both viruses - gives the virus one extra chance to reassort in a way that produces a bird-antigen-fortified human-adapted flu virus. This doesn’t mean H5N1 will definitely become human-transmissible soon. Many viruses hang out on the borders of transmissibility for decades. Some, for unclear reasons, never cross over at all. But all of this is compatible with the virus becoming transmissible soon. So: What Is The Chance Of A Pandemic? The prediction markets on this topic ask a question about “10,000 cases in the United States”. Does this necessarily mean “pandemic”? Might it be possible to get to 10,000 cases just from the scattered chicken and cow farmers, with no human-to-human transmission? Despite many chicken and cow infections this year, there have only been 60 - 70 recorded human cases. Unless there is a phase change in screening methods, it seems hard for this number to increase to 10,000 off farmers alone. I think it’s fair to treat this question as operationalizing “what is the chance of a pandemic”? By this definition, Manifold estimates a 40% chance of an H5N1 pandemic in 2025. Metaculus estimates a 5% chance. You can see below whether that’s changed since I wrote this essay: 5% versus 40% is a big difference! Who do we trust? I trust Metaculus. Metaculus has beaten Manifold in both of the two head-to-head comparisons that I know of (Jeremiah Johnson’s and mine). Manifold’s number swings by a factor of two from week to week; Metaculus has been steady. But also, Metaculus hosts a CDC-sponsored respiratory disease forecasting tournament which has enriched them in epidemiological expertise. And if you look at the quality of comments on both sites, it’s pretty obvious where the people with more intellectual chops are hanging out. The Manifold comments are mostly single sentences, or occasionally just links to an article about new cases. The Metaculus comments look more like this one by dimaklenchin: Despite the panic propaganda, H5N1 is unlikely to be "just a single mutation away from switching host preference": 1) It normally takes a lot more than a single mutation to switch hosts. E.g., there are at least five different reasons why SIV (monkey equivalent of HIV) is not infectious to humans. Heck, a variant of SIV that bears HIV's receptor-recognizing surface protein (SHIV) is still not infectious to humans. HIV most certainly evolved from SIV but, almost as certainly, it took a very long time to get there. Not that all viruses are the same and things can't turn out differently with flu, but I don't subscribe to the idea that a mere change of receptor specificity (something that can take 1-2 mutations) will be sufficient. 2) We have data. Lots of human infections with other varieties of bird flu in the past - all those viruses ultimately went nowhere. Why would H5N1 be radically different? E.g., the "Canadian teen", despite what sounds like a prolonged exposure, failed to infect anyone around him. Since I am at 18% for the h-2-h H5N1 detection in 2025, I am arbitrarily going ~ an order of magnitude lower than that for something as unprecedented as 10K human infections. Maybe should be much lower but hedging for the time being and will allow another couple months of observations. And Sergio: I'm currently at 20% on the question of reported human-to-human transmission of highly pathogenic avian influenza H5N1 globally before 2026. However, this question is only about the US, and is more general about all subtypes of H5. But H5N1 very strongly appears to be the most important subtype to consider in this time period. And, given the current situation in the US with H5N1 human cases derived from exposure to poultry or cattle (with cattle(mammals) being more worrisome), h2h transmission seems quite more likely to arise in North America than elsewhere before 2026. Conditioning on h2h transmission in the US (and also trying to consider, with lower probability, a start in Canada), I want to estimate the chances that it becomes sustained and out of control (in which case, if it starts in Canada, I largely expect it to spread to the US). The (6) past events of probable h2h transmission of avian H5(N1), none of which were sustained, could serve as a base rate, although I'm a bit wary of giving much weight to this precedent, since the last event was quite a while ago (2007), and also because reporting and testing standards may have improved considerably since then (so perhaps they might not have been classified as h2h transmission events if they had occurred more recently). The current situation in the US, and events such as the Canadian teen who got sick with H5N1, do suggest a higher background level of risk than normal (which would be reduced if a vaccine for cattle is licensed soon), but I'm wary of overupdating. Conditioned on sustained h2h transmission, reaching over 10k cases in a few months seems likely, although perhaps very strong monitoring and surveillance could contain the situation in time (at the very least to moderate the growth rate). Trying to combine all these factors somewhat haphazardly, I'm currently at 3.5% for this question. That’s before 2026. What about longer-term? Manifold gives a ~50% chance before 2030; Metaculus uses a more complicated method but it says about 25% chance before 2030. H5N1 may cross to humans, but it could take a while. Superforecaster Juan Cambeiro at The Institute For Progress estimated a 4% chance of a “worse than COVID” H5N1 pandemic in “the next year”, but their estimate was made in 2023, without the benefit of the Metaculus estimates or most of our current knowledge. This feels high now - Metaculus says 5% total for H5N1 pandemic, and most pandemic flus are not worse than COVID. IFP also seem to be expecting a case fatality rate greater than 10%, which I find unlikely for the reasons mentioned above. I trust their estimate less than Metaculus’ current ones. I conclude that the most plausible estimate for the chance of an H5N1 pandemic in the next year is 5%. Interestingly, 5% is about the base rate for pandemic flus per year: five in the past century = one per twenty years = 5% chance per year. Isn’t it surprising that we’re still at the base rate when we can see a dangerous-looking flu virus spreading through the types of animals that have caused pandemic flus in the past? Part of the answer is that we’re not - in addition to the 5% chance of H5N1, we have to add the chance of some other pandemic flu. This probably isn’t 5% on its own; scientists monitor flu strains closely, and they haven’t found any others which are giving off as many red flags as H5N1. Still, something could always come out of left field. Maybe we should add a 2.5% chance of some other strain, for a total of 7.5% chance of a flu pandemic (ie beyond normal seasonal flu) next year. But still, isn’t it surprising that we’re so close to the base rate? One way to think about this: the base rate represents how concerned we should be if there was no epidemiological monitoring at all. In that case, we would estimate a probability distribution across different epidemiological landscapes, most of which contain some concerning-looking flu strains. Since we are doing the epidemiological monitoring, we can collapse that distribution into a single picture: one flu strain, H5N1, is in fact pretty concerning, and other strains mostly aren’t. This is enough to move our prior from 5% to 7.5%, but no more. The forecasters I talked to raised one other point of uncertainty: does the flu work more like a dice roll, or like a bus? Dice rolls are uncorrelated with their predecessors; even if it’s been a hundred rolls since you last rolled a 6, your chance this time is still 1/6. But buses come at fixed intervals; if the buses are hourly, and you haven’t seen a bus in the past 59 minutes, then your chance of seeing a bus in the next minute is very high. It’s been 16 years since the last flu pandemic; these pandemics come (on average) every 20 years. I don’t think anyone has a good sense of how to think about this. But it was 40 years between the Spanish and Hong Kong flus, so the twenty year number is at best a rule of thumb. The 5% number feels very low to me (and, apparently, to the average Manifold forecaster). Isn’t H5N1 spreading to cows and pigs and all sorts of other mammals? Isn’t it in the news all the time? I trust Metaculus a lot, but I agree that this is a surprising update, and I’m taking it on faith rather than feeling it in my bones. What Would The Fatality Rate Be For An H5N1 Pandemic? There are four basic stories you could tell about likely H5N1 mortality. First, maybe mortality would be 50%. The argument here is that official statistics report this mortality rate in the chicken farmers who have been infected with H5N1 so far. Several news sources and even some scientists have raised the specter of a pandemic version of H5N1 pandemic with this same death rate, which could kill a quarter to a third of the world population. THIS IS EXTREMELY FAKE. The official statistics only report fatality rate in the infections we know about. Bird flu is rare, there’s no mass testing, and we only learn that somebody had it if they’re in a hospital and the doctors are worried enough to test for rare conditions. Of Americans who got bird flu in the past year, 0 out of 61 have died. Probably this is mostly because America upped its detection game and is now finding milder cases; we also can’t rule out the virus mutating to become less virulent. Metaculus estimates the current true mortality rate as 1.25%. …but leaves a wide 90% confidence interval, from 0.5% to 7%. Second, maybe mortality would be somewhere around 1.25%. The argument here is that Metaculus uses this as its central estimate of US mortality. But Sentinel discusses some reasons to be skeptical of broad inferences from the US numbers: Scientists have been puzzled by the apparently low H5N1 case fatality rate in humans in the US. They offer a number of hypotheses: “The way in which the virus is being transmitted — along with the amount of virus exposure — is limiting the severity of disease.”
Inline links: was found in pigs, Jeremiah Johnson’s, mine, a CDC-sponsored respiratory disease forecasting tournament, dimaklenchin, Sergio, reported human-to-human transmission of highly pathogenic avian influenza H5N1 globally before 2026, human cases, past events, vaccine for cattle is licensed soon, https://substackcdn.com/image/fetch/$s_!fL7J!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fd6a70fa6-b356-422d-ba9a-5db431e5a056_751x471.png, The Institute For Progress estimated, news, sources, scientists, Sentinel discusses