Schneier on Security
A blog covering security and security technology.
« Rudy Giuliani on Terrorism Security |
| MySpace and U.S. Attorneys General Agree to Fight Sexual Predators »
January 17, 2008
Fear Is Unhealthy
The New York Times writes about a plausible connection between fear and heart disease:
Which is more of a threat to your health: Al Qaeda or the Department of Homeland Security?
An intriguing new study suggests the answer is not so clear-cut. Although it’s impossible to calculate the pain that terrorist attacks inflict on victims and society, when statisticians look at cold numbers, they have variously estimated the chances of the average person dying in America at the hands of international terrorists to be comparable to the risk of dying from eating peanuts, being struck by an asteroid or drowning in a toilet.
But worrying about terrorism could be taking a toll on the hearts of millions of Americans. The evidence, published last week in the Archives of General Psychiatry, comes from researchers who began tracking the health of a representative sample of more than 2,700 Americans before September 2001. After the attacks of Sept. 11, the scientists monitored people’s fears of terrorism over the next several years and found that the most fearful people were three to five times more likely than the rest to receive diagnoses of new cardiovascular ailments.
After controlling for various factors (age, obesity, smoking, other ailments and stressful life events), the researchers found that the people who were acutely stressed after the 9/11 attacks and continued to worry about terrorism -- about 6 percent of the sample -- were at least three times more likely than the others in the study to be given diagnoses of new heart problems.
If you extrapolate that percentage to the adult population of America, it works out to more than 10 million people. No one knows what fraction of them might consequently die of a stroke or heart attack -- plenty of other factors affect heart disease -- but if it were merely 0.0003 percent, that would be higher than the 9/11 death toll.
Of course, statistics of any sort, even when the numbers are rock solid, don’t mean much to people when they’re assessing threats. Risk researchers have found that even when people know the numbers, they’re less worried about death tolls than about how the deaths occur. They have good reasons -- called "rival rationalities"? -- for fearing catastrophes that kill large numbers at once because these events affect the whole community and damage the social fabric.
It doesn't surprise me that fear of terrorism is more harmful than actual terrorism. That's the whole point of terrorism: an amplification of fear through the mass media.
Refuse to be terrorized:
The point of terrorism is to cause terror, sometimes to further a political goal and sometimes out of sheer hatred. The people terrorists kill are not the targets; they are collateral damage. And blowing up planes, trains, markets or buses is not the goal; those are just tactics. The real targets of terrorism are the rest of us: the billions of us who are not killed but are terrorized because of the killing. The real point of terrorism is not the act itself, but our reaction to the act.
And we're doing exactly what the terrorists want.
The surest defense against terrorism is to refuse to be terrorized. Our job is to recognize that terrorism is just one of the risks we face, and not a particularly common one at that. And our job is to fight those politicians who use fear as an excuse to take away our liberties and promote security theater that wastes money and doesn't make us any safer.
Posted on January 17, 2008 at 7:35 AM
• 29 Comments
To receive these entries once a month by e-mail, sign up for the Crypto-Gram Newsletter.
That is exactly the point. The polititians are puppets playing exactly as terrorists planned.
But statistics alone doesn't mean anything.
Statistics: "Give me enough time and I will prove you anything you want"
I was hoping the last few paragraphs were a continuation of the NYT article. No such luck.
Some day they'll cotton on to it.
It surprises me that "they have variously estimated the chances of the average person dying in America at the hands of international terrorists to be comparable to the risk of dying from eating peanuts, being struck by an asteroid or drowning in a toilet."
It strikes me (ha!) that death-by-asteroid is so unlikely that it is in a league far beyond terrorists, peanuts, or bizarre drownings. I know people who have deathly allergies to peanuts, but I can't recall ever reading of an injury - much less a death - due to an asteroid strike. I find the bizarre drownings a bit more credible based on my recollections from college.
I notice a little bit of statistical playing, too, as they specifically mention "international terrorists", placing the domestic terrorists like Timothy McVeigh and neo-Nazi skinhead thugs outside the sample set. These people are clearly terrorists, but perhaps they look too much like "us" to be easily distinguished as "them".
While there are certainly political connotations to the distinction between "International" and "Domestic" terrorism, the conclusion would stand even if the modifier were elided and the discussion were simply about "terrorism". Throw in McVeigh, skinheads, the KKK, the Atlanta Olympics bomber and other abortion-clinic-hating nutjobs, and other minor cases, and the resulting deaths are still statistical noise compared to auto accidents (50k/year), for example. Averaged over a couple of decades, death by all terrorism is probably within a small factor of death by shark bites.
I suspect that the statistic is supposed to reflect death due to a meteor strike, not being directly hit by a meteor (the former, while pretty unlikely, is far more likely than the later). The idea is that the Earth gets bombarded very infrequently with pretty massive meteor strikes which kill large numbers of creatures, so if you amortize those deaths over several million years, you get a death rate by meteor strike which is close to that of being struck by lightning (but instead of a few deaths per year, it's a lot of deaths every once in a great while).
In any case, the chance of being struck by an asteroid is effectively 0, unless you are an astronaut who is, let's say, traveling to an orbit around Jupiter, and your ship has lost any maneuverability and no one bothered to look at your path in advance (in which case, you're probably going to die anyway).
Looks like I'm wrong; perhaps the definition of "asteroid" and "meteor" changed since I last took Astronomy. They are, however, likely referring to Asteroid strikes (as opposed to a direct hit by an Asteroid).
Presumably it's not just fear of terrorism that has this effect. It would be reasonable to expect the same increase in mortality among people extremely stressed about other things, such as global warming.
Refuse to be terrorized.
"drowning in a toilet"
There are no warning labels about this on toilets! Whom do I sue?
Does this mean that security theatre actually saves lives by reducing fear? This could also be used as an argument in favor of the DHS: If people feel less fear because the DHS "does something" about the terrorist threat, then the conclusion is exactly the opposite of the one in the article.
We're not doing what the terrorists want. It's our government that's doing it.
"Does this mean that security theatre actually saves lives by reducing fear?"
Nope. Because "security theatre" does not reduce fear. It continues to promote the old fear to justify itself.
Why do I still have to take my shoes off at the airport?
Why do I have to put my 2oz of toothpaste in a clear plastic baggie?
And watch what happens if any mid-eastern male shows up in a turban.
"Nope. Because "security theatre" does not reduce fear. It continues to promote the old fear to justify itself."
Good point, I should have thought of that. Still, when I see interviews with passengers the general attitude seems to be that the security theatre makes them feel safer. The readers of this blog, including myself, probably don't see it that way, but I don't think we are a representative sample of the whole population. Your point about promoting old fear as justification is still valid, though.
BTW: "It doesn't surprise me that fear of terrorism is more harmful than actual terrorism. That's the whole point of terrorism: an amplification of fear through the mass media."
I hereby propose a new acronym: "FASED" (Fear Amplification by Stimulated Emission of Disinformation)
Why be in Fear when we could be in Terror?
I propose "Terror Amplification by Stimulated Emission of Disinformation", or "TASED".
"The people terrorists kill are not the targets; they are collateral damage. And blowing up planes, trains, markets or buses is not the goal; those are just tactics."
Right. So I'm sure if we just nonchalantly refuse to be terrorized, the terrorists would quit in frustration, and go home. Sounds like a swell strategy.
"The real point of terrorism is not the act itself, but our reaction to the act."
Exactly. So the next time a planeload of people explodes, we should film ourselves laughing hysterically, and post the videos to the Net. When the terrorists see our reaction, which is not the reaction they wanted, they'll stop blowing things up.
It's a wonderful plan!
@We'll Show Them
No need to be so hurtful to the relatives of the dead. All that is needed is a rational response. Start an investigation with the aim of finding the culprits rather than covering some officials back.
Warn people to be vigilant.
Get back to normal as fast as possible.
We used to do this in the UK when the IRA were actually bombing things every week. Sadly since 9/11 we seem to have caught the US invented security theatre meme.
It is a shame that the same logic that goes into the policy of "we don't negotiate with terrorist" seems to follow "we don't allow ourselves to be terrorized," but it is not preached and practiced as such.
> There are no warning labels about this on toilets! Whom do I sue?
I suspect that the cause of this correlation is that the same microbes which cause cardiovascular disease also cause anxiety. The leading suspect for microbial causation of cardiovascular disease is the bacterium Chlamydia pneumoniae; and infections with that germ often involve secondary porphyria, one of the main symptoms of which is anxiety.
Even if you don't buy into this particular germ as the cause, people with one health problem often have others; and health problems often affect the mind.
What you've prescribed sounds quite reasonable to me. But does it go far enough? You say that all that is needed is a rational response. But needed for what? To successfully prevent the next attack? I think maybe two, three, or a dozen rational responses might better protect against the next attack.
When the IRA were bombing things every week in the UK, "getting back to normal" could easily be understood as stoically accepting that next week there's going to be another bombing, so you might as well get used to it and learn to live (unfearfully) with it.
What about those who don't want to live with it?
If "the real point of terrorism is not the act itself, but our reaction to the act", why don't you see terrorists killing puppies en masse, or releasing a video threatening to detonate a dirty bomb in a given city on a given day? They could get quite a reaction from those tactics, yet what they do instead, time and again, is attack airplanes, airports, and crowded markets, killing people. Seems as if they're going for something more than just our reaction. Seems like they're trying to kill people.
To imply that reacting strategically is more important that working to prevent these specific acts of terror, is to give bad advice.
@Andy, Carlo Graziani
"While there are certainly political connotations to the distinction between "International" and "Domestic" terrorism, the conclusion would stand even if the modifier were elided and the discussion were simply about "terrorism"."
From my reviews of the data available at:
http://www.start.umd.edu/data/gtd/ and http://www.nsc.org/lrs/statinfo/odds.htm
your observation is spot on. Yes, there are some questions as to how you count, and what. But even if you are generous on the side of allowing event to called terror, the "usual" events still outweigh the death by terrorism by a huge margin (Heart disease kills 1 million to 1 more than terrorism, even with broad terror definitions.)
In defense of sharks, when I first started my reviewing I suspected sharks and terrorists were equal too, but I was surprised to find that deaths by shark are very rare, even more so than deaths by terrorism. On the Auto and Heart disease scale, the difference is minor, but notable. According to http://www.flmnh.ufl.edu/fish/sharks/attacks/... the risk of fatal shark attack is about 1 in 3.75 million, while lifetime odds of dying to terror have been placed around 1 in 1.5 million or 1 in 1.9 million (depending on source and method used).
Heart disease still ranks top at 1 in 5 for lifetime odds.
so, from a public health perspective, anyone who calls in about a "suspicious package" in a public space should probably be given a certificate for a cardiovascular health screening. as a "reward."
While agreeing that the effect of the 'War on Terror' could easily have a bad effect upon the physical and mental health of a nation, I'm surprised that nobody seems to be talking about the effects of being subjected to arbitrary and pointless bureacratic dictats, whatever the source may be.
The fact is that being subjected to petty, unpleasant rules by police, TSA, government, bank and utility companies can easily raise stress levels and that is unlikely to be good for you.
I suggest that the 'terror' angle is being overblown here. If our governments and utility companies respected its citizens/sutomers and tried to minimize the daily grind of modern life, we would all be better off physically and mentally.
Thanks for the numbers, they are worth squirreling away. I must say, however, that at 1:5M versus 1:3.75M, the lifetime death risk from terror and from shark attack are actually closer than I thought they'd be -- I would have been happy with a relative factor of 10 or less, whereas the actual factor is apparently about 2.5.
What about fear for our jobs and livelihood? That's become a greater and greater stressor for many people as companies try to squeeze out more and more costs to keep Wall St happy, hasn't it?
@Jack C Lipton
I agree. Terror in our public schools is a good example. Osama has nothing over what kids have to confront every day.
The pharmaceutical industry has been a generous donor to Republican campaigns. So the damage to people's health resulting from the War on Terror could only be beneficial for an administration dedicated to providing its loyal donors with the maximum return on their investment.
Bullshit statistic alert!
Suppose for the sake of argument that the study does show a significant correlation between heart disease and fear of terrorism (I haven't read it).
The question you should ask yourself is, "does this prove that fear of terrorism causes heart disease?"
I think it's obvious that, while a causative connection in that direction is plausible, the study shows no such thing. It may be that fear of terrorism causes heart disease (plausible, since we know that stress can have that effect). It may be that heart disease causes fear of terrorism (although I don't know what the mechanism would be, so I doubt that somewhat).
It may be that there is a common cause of both phenomena. Norman Yarvin already suggested one mechanism. The mechanism I would suggest here is that any tendency to anxiety will (a) cause you to be more likely to fear terrorism, and also (b) cause you to be more subject to heart attack, due to increased stress and anxiety about *everything*, not just terrorism.
To attribute to terrorism the full effects on cardiac health of a broad-based tendency to anxiety is to massively overestimate the effects of fear of terrorism. An easy mistake to make if you're campaigning against fear-mongering anyway, and looking for empirical support for your position, but still a mistake.
> After controlling for various factors
> (age, obesity, smoking, other
> ailments and stressful life events)
Sounds like they didn't control for "fear of a stock market crash", "fear of a Republican win in 2008", or "fear of being imprisoned for a crime they didn't commit". So my "general fearfulness of everything" hypothesis is not ruled out, and there is not yet any evidence that terrorism-related scare-mongering has anything to do with the results.
The most direct way to prove causation, is a randomised controlled trial (for medical interventions you want it to be double-blinded too, because the placebo effect is so strong. But here that's unfortunately not possible, so you will also need an expert to help design your trial and interpret your results).
So, you take a few hundred people (cardiac disease is very common, so a small sample should give statistically significant results if there really is an effect of the magnitude suggested above), and randomly select a group which will be made unafraid of terrorism, and a group which will be made afraid of terrorism. Then record incidence of heart disease and crunch the numbers.
If there's a difference, *then* you've proved that the intervention is the cause of the difference. If there's no difference in a randomised controlled trial, despite a correlation in the general population, then you're looking for a common cause (or causation the other way).
And don't forget about the other HIGH profile domestic terrorists: the CIA & the US Government.
Schneier.com is a personal website. Opinions expressed are not necessarily those of Co3 Systems, Inc.