Nobody knows. There are just lots o’ variables. There’s also a large pool of immunosuppressed - 50 million infected with HIV worldwide.
Thats just not true. The population of the Earth in 1918 (when the last major flu pandemic happened) was 1.8 billion. Today’s population is 6.5 billion. There are 314 cities with over a million people in it. There are a half-dozen cities with more than ten million people in them. These are conditions that nobody would have dreamed of in 1918. We can hop on a plane and visit any one of these cities in less than twenty-four hours. The 1918 flu took 8 weeks to expose about half the world population and kill 2-3% of it before it mutated. How much faster will things go with routine international air travel?
I’m not saying we should panic. But the truth is that deadly viruses and bacteria mutate. Sometimes they mutate into deadly and easily communicable forms. It may not be this year, but it’s going to happen. Science may be able to cure a lot, but it can’t keep deadly forms from emerging. If we can prepare for this, we can keep death tolls down. Imagine if it hadn’t taken us until Americans were dying to figure out AIDS.
So what exactly do you disagree with in my reasoning. When we assess risk, the rational way to do so is to base it on probability of an event, and the damage that will occur if that event transpires. Risk is directly proportional both to probability and potential damage.
If we assess only by probability, then we would not be prepared adequately for unlikely events. That precise sort of reasoning led the U.S. to not build levees that would protect New Orleans from a Cat 4 or 5 hurricane. Extremely unlikey to happen, they said, so why bother? If we assess only by potential damage, then we would have to panic about asteroid impacts wiping out the whole human race. If we assess risk by both factors, we get a reasonable measure of how much effort should be spent preventing a potential disaster.
That the most likely number of deaths to expect is 500,000. That planning should be based on that number.
The probability of half a million deaths is essentially zero. It’s either ten million (P=.05) or none (P=0.95).
If you plan for half a million instead of ten million, you’ll be either uselessly underprepared or the effort will be wasted entirely. If you plan for ten million, and base your actual effort and expenditures on the 1-in-20 probability of that happening in comparison to the probabilities and severities of other demands on your resources occurring, then that’s rational.
I think the notion of expected value is valid. If there is 5% chance of killing 10,000,000 people and 95% chance of killing nobody, then the expected value would indeed be 500,000. Just like if you spend $5 on the lottery that returns 40% of the loot to the players, your expected winnings are $2. Event A with a 5% chance of killing 10,000,000 people is much more significant than Event B with a 100% chance of killing 100 people, even though we know know that the chances are 19 in 20 that Event B will actually kill more people.
Apples and oranges. Let’s say I have a 50% chance of falling off a boat and drowning. Am I best served by having half a life jacket?
If you prepare for a disaster, then you prepare for what that disaster might actually bring. If it costs too much for the level of risk then don’t spend it. Highly unlikey events may not be worth it. Less unlikely events may be worth it. But spending enough to still be unprepared is wasted money entirely.
But on the other hand, people move and travel much more than they did in the 1910s, which could facilitate the pandemia.
The Y2K bug was real. And was adressed at a ungodly cost.
There was an associated bacterial infection, which was thought to have caused the flu. In 1918 they couldn’t see viruses. But I don’t believe this infection was primarily responsible for the deaths - the flu invaded the lungs directly,
BTW, the 1918 flu was called Spanish flu because most countries, on war footing, censored news of the outbreak to avoid giving comfort to the enemy. Spain, being neutral, did not, and got stuck with the name, though I believe it started in Kansas.
The 1918 flu, which thought to be a swine flu, turned out to be avian, so it is another reason to be afraid.
My wife just finished writing a book on flu (very timely!) so I’ve been hearing way more about this than I need to. Interested people can read the books by Gina Kolata and Berry.
Yup, that’s the big deal with it. With people flying between continents every day all the time, if it ever reaches a major city, it’s guaranteed to hit every continent in the world within days. Most likely before it’s ever identified as H5N1. And once it hits an urban area, it’s likely to spread like wildfire.
But then again, this is if it ever mutates into a form that is transferable from human to human, which isn’t guaranteed to ever happen.
The more I think about it, the less I believe the 10,000,000 or zero theory. I think the probability of the number of deaths would be more like a bell shaped curve. Perhaps it is spiked quite close to the y axis and perhaps there is little area under the curve at the 10,000,000 ordinate, but I dare say zero is not terribly likely either.
Bob,
He was using made up numbers and made up probabilities. No one knows the real odds. Estimates are really WAGs because so much is unknown. What is known is what previous pandemics did and that international travel will allow this one, if it happens, to spread very quickly. Worst case is that it reassorts without losing much of it lethalness but gaining human to human infectivity and having people get ill just slowly enough to be able to travel first, without much in the way of preparation done. No stockpiles of Tamilflu by the WHO or by individual governments. No rapid response plan. No infrastructure ready to produce vaccines rapidly. In such a case 1918 would look like a fizzle. Not enough meds, Tamiflu or antibiotics, to go around. Not enough healthcare workers even if most of them didn’t get ill themselves, which would be likely. Resources overwhelmed. Hundreds of millions dead. Economies would be trashed worldwide. Is that likely? Hardly, since the world is responding and preparing. And again this bug may not be able to shift like that or may lose its lethality as it does. When a pandemic hits it might just look like '68. Or anywhere in between.
The kind of analysis he was playing with would be useful (presuming if one had real numbers to play with) if one needed to decide what project to fund. For example, do you fund a project to detect and destroy that asteroid hit on Earth which has, let’s pretend, a one out of 500,000 annual risk but the consequence of, say 3 billion dead, or flu preparation with the numbers he made up, using the same limited resource? Sure, the catestrophe of an asteroid hit is very severe but the annual risk calculates to only 600 lives. Clearly flu preparedness is more important with those fictional figures. But it doesn’t mean that planning for an asteroid strike (should we decide to do so anyway) should consider 600 dead as a result.
Anybody that wants to understand why this particular strain is so worrisome should read the WHO’s primer on it -Avian Influenza: Assessing the Pandemic Threat. published in January. It’s really fascinating and very good at explaining the fairly complicated science and history in lay terms. Also it talks in detail exactly about the risk analysis we’ve been discussing.
Other resources for the interested reader- a 2003 Science article calling attention to the problem (over two years ago). And a current article from the same rag on reconstruction of the 1918 virus demonstrating that some of the virulance of that bug was for very particular features intrinsic to it, including its ability to reproduce in other than lung cells, vigourously reproduce in bronchial cells, and cause much direct inflammation. Also that it came from an avian source.
Granted, DSeid, but I was merely refuting the “It’s gotta be either 10 million or zero” analysis. I believe it will almost certainly be something in between. IF (and I’m not sure that you can) you could plot the probability as a bell shaped curve, prudent planning would be to prepare for that number that would correspond to say 95% of the area under the curve.
Wouldn’t the 60 deaths already attributed to Avian flu put it squarely between 0 and X? The probability just relates to things like will it go “pandemic” versus being an isolated (and efficient killer). It’s already jumped from birds to humans; that we know. What we don’t know if whether it will mutate to jump person-to-person. If so, then we’re on the road to “outbreak”. If it does that AND jumps efficiently it’s on the road to pandemic. If we get a pandemic and the mutation doesn’t affect the high mortality rate, then we’re on the road to disaster. That’s scary but it’s still a lot of “ifs”.
It was oversimplified, sure - but it’s still based on the idea that the virus will either mutate into something that will infect anad kill humans very efficiently, or it won’t affect us much at all. The epidemiologists haven’t offered the possibility that it will mutate into just another flu strain no worse or no milder than most others. They could be wrong about that, probably no one this board can say, but if they’re right, the probability distribution is still bipolar and not Gaussian. One pole is near zero, not at it, sure, but the other is at 10e7 and anything in between is down in the noise.
We’re not talking about halves of lifes saved. If there were two people on the boat each with the same 50% chance of drowning, would you say that having only one life jacket was useless?
If the actual number killed is expected to be 10 million, and you have only prepared for 500,000, the 500,000 saved will be pretty happy there was money spent on preparation even though it was inadequate to save everyone.
Part of the NOAA’s job is to issue warnings about hurricanes. 
But they did say that this year was expected to be unusually bad in terms of storms, and days before it hit, they were saying Katrina was going to head for New Orleans and cause a lot of damage. When you get a warning like that…
Meanwhile, this thread makes me more concerned about the possibilities of this illness. I’d heard about the avian flu before, but I tend to assume these things are overhyped. I’m a young guy in good health, we’ll see if that helps.
Yes, and the relatives of the other 9.5 million will be justifiably upset, because you knew that 10 million people could be killed and prepared inadequately. If the options are 10 million and 0, it makes no sense to prepare for some other figure unless being ready for 10 million is just impossible.
Actually those are possibilities discussed by the experts as well. It could be anywhere in between. It could keep all of its lethality as it gains human to human transmissibility or it could lose some degree of it. Experts merely say that we better be ready for the worst case. Or at least realize that the worst case is a distinct possibility unless we take measures to prevent it. Have the smoke detectors installed, fire extinguishers at the ready, and an escape plan practiced. May you never need it. But if you do, having it ready will prevent one unavoidably frayed wire from burning the house down.
I’d say it was unconscionable not to have two life jackets.