The real reason epidemiologists are concerned about swine flu is its novelty. It's possible that it's very far away in phase space from a truly killer flu, but it's also possible that it's only a couple of mutations away from becoming deadly.
Since flu mutates fast, any truly novel strain that reaches pandemic status represents a serious threat. Fortunately this kind of leap only happens every few decades; unfortunately, we have no defense against it.
I would love to see the day when we can calculate the Hamming distance to deadliness for all the common flu strains, so we'll know exactly how alarmed we should be each year.
Of course, that means we'd be living in a future where the DNA sequence for the deadliest possible flu strain is computable. We'd have a pretty good reason to be alarmed every single day.
Once you vaccinate for it, it becomes less deadly. Which means there is a new (computable) "most deadly" flu. And, presumably, there is an upper limit to the number of strains our immune systems can remember and defend against.
Perhaps; on the other hand, once we have the capability to do something as hard as compute most-deadly-flu, we'll almost necessarily have the immunological understanding to vaccinate against very broad categories of diseases, not merely individual strains.
Most viruses just make you sick. It's at least possible that a virus could evolve where the RNA coded for THC or alcohol. Or some other chemical that affects mood. That would make for an entertaining, non-sobering flu season. Can someone estimate the probability of something like that happening spontaneously?
"Can someone estimate the probability of something like that happening spontaneously?"
With a lot of handwaving ...
----
If everything goes perfectly (each random mutation flips the right nucleotide), you would need at least 77.5 years to get alcohol-production-capable RNA in influenza virus solely by spontaneous mutations.
----
Details:
Checking one possible ethanol fermentation pathway [1]: at minimum you would need to synthesize two enzymes - pyruvate decarboxylase (1,691 base pairs [2]) and alcohol dehydrogenase (1,046 base pairs [3]).
Influenza A has 13,558 base pairs [4] and mutates at about 0.0026 mutations per site per year [5] (so you get about 35 changed nucleotides per year).
Before everyone panics, however, I'm pretty sure this kind of statistic is self-reinforcing: if reports of flu are higher, and those reports are broadcast, people are more likely to go to the hospital when they exhibit flu symptoms, making the flu counts higher yet.
The media coverage of the swine flu might very well have set off this cycle.
Addendum: it isn't necessarily a bad thing - according to this hypothesis, if people who would have died of the flu this year are more likely to go to the hospital, and so less people should die. Or it might be that those who die of flu have reasons for not going to the hospital unaffected by media.
There is much truth to that, and you have to expect some exaggeration in numbers. However http://scienceblogs.com/whitecoatunderground/2009/10/whats_t... points out that pediatric deaths from flu this year are pacing about twice a normal year, and are accelerating. So there is a real issue. Similarly in certain areas hospitals are running out of ER beds. That's highly unusual and is objective evidence that there is a real problem under the hype.
When all is said and done, though, I'd expect a death toll somewhere in the 50-100,000 person range. Which sounds horrible, but is a pretty small fraction of the population. For comparison it is on par with traffic fatalities in a 2 year period, or normal flu fatalities over 2-3 years.
But that number could go up sharply if a lethal mutation emerges and spreads as happened in 1918.
Well, that's just it isn't it. More people with flu symptoms are going to the hospital and being diagnosed with flu, whereas in the past, they stayed home and got better.
A women felt ill. Then felt a little better. Then got ill again within a few days and died in the Hospital overnight.
Without even CONFIRMING that it was Swine Flu, they are reporting it like it is, and they are scaring up everyone in town to think that if they have a tickle in their throat they are going to die in 3 days.
In the UK, you're told specifically not to bother going to doctor/hospital etc if you have swine flu (Unless you have complications/other conditions etc).
We had it in July or so. It was kinda like a normal flu :/
The media loves to scare the hell out of everyone - gets them to buy stuff. Sad, but I can't see that ever changing.
Yeah, that makes sense - sometimes I really want to smash the media guys so hard.
But I also wonder if I person dies in a hospital, isn't shouldn't there be a legal course of death on the death certificate? Because then we might have something to compare with.
Trying to draw trends from a single cause of death is much harder than it sounds. Particularly in the case of things like influenza, which rarely kill anyone directly.
What you'd really need are actual electronic medical records, so you could track symptoms, progression, treatments, etc.
Since flu mutates fast, any truly novel strain that reaches pandemic status represents a serious threat. Fortunately this kind of leap only happens every few decades; unfortunately, we have no defense against it.
An interesting article covering this point: http://scienceblogs.com/effectmeasure/2009/04/swine_flu_i_be...