Flu, Scott Alexander reminds us, is a disease caused by a family of related influenza viruses:
Pandemic flu is always caused by the influenza A virus. Influenza A has two surface antigen proteins, hemagglutinin (18 flavors) and neuraminidase (11 flavors). A particular flu strain is named after which flavors of these two proteins it has – for example, H3N2, or H5N1.
Influenza A evolved in birds, and stayed there for at least thousands of years. It crossed to humans later, maybe during historic times – different sources give suggest dates as early as 500 BC or as late as 1500 AD. It probably crossed over multiple times. Maybe it died out in humans after some crossovers, stuck around in birds, and crossed over from birds to humans again later.
During historic times, the flu has followed a pattern of big pandemics once every few decades, plus small seasonal epidemics each winter. The big pandemics happen when a new strain of flu crosses from animals into humans. Then the new strain sticks around, undergoes normal gradual mutation, and once a year immune response decays enough / mutations accumulate enough to cause another small seasonal epidemic (Why is this synced to the calendar year? See here for more).
Let’s digress to visit his piece on “diseasonality”:
The most common theories for disease seasonality are:
- Pathogens like the cold
- Pathogens like low humidity
- People are cramped indoors during the winter
- People have low vitamin D during the winter, and vitamin D helps fight pathogens
None of these are really satisfactory on their own.
[…]
Deprived of seasons, a place doesn’t just have a slow burn of flu cases all year. It has a big epidemic, then dies down for a while, then has another big epidemic. In retrospect, this is an obvious consequence of how diseases work (eg the SIR model of transmission). Some people get the disease, it spreads exponentially until lots of people are immune, and then it stops until something changes.
[…]
So in the tropics, Florida, and Alaska, epidemics “want” to follow a cycle of coming approximately once a year. In the tropics, nothing is giving them that cycle, so they come at a random time once a year, or twice a year, or whatever. In Florida, UV light, temperature, etc provide that cycle, and they come once a year. In Alaska, UV light, temperature, etc also provide that cycle, and they come once a year independent of what’s going on in Florida.
Back to the original piece:
The severity of any given flu epidemic depends both on the innate severity of the virus, and on how closely the human population’s circulating flu antibodies match the epidemic strain. People usually have good antibodies to the seasonal flu, because it’s only slightly different from last year’s seasonal flu. For the big new animal crossovers, the level of protection provided by existing antibodies is unpredictable. Older people may have antibodies left over from the last time that particular flu crossed over from animals to humans; younger people probably won’t. In some cases, people’s immune systems will be permanently synced to the first flu they encounter, with less protection against subsequent versions.
So for example, the Spanish Flu of 1918 was an H1N1 strain that killed about 2% of the world population. But the exact mortality pattern was surprising; people between 18 and 28 were especially likely to die, and people older than 88 especially likely to survive. Why? Because an H1N1 flu went pandemic in 1830; anyone who first encountered the flu around then had an immune system synced to H1N1. But an H3N8 flu went pandemic between 1890 and 1900; anyone who first encountered the flu then had an immune system synced to that strain and was unprepared for H1N1.
[…]
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.
[…]
Third, maybe mortality would be between 0.01% and 0.2%. The argument here is looking at normal (ie not 1918) flu pandemics of the past century. The least bad among these, the 2009 swine flu, had CFR of 0.01%. The worst, Hong Kong flu, was somewhere around 0.2%. If H5N1 is a normal pandemic flu — and right now there’s not much that differentiates it — it will probably be somewhere in that range.
Fourth, maybe mortality will be 2-10%. This was the mortality rate of the 1918 Spanish Flu. It seems to be an outlier: as far as we know, no other flu in the past 500 years was nearly as bad.