about Epidemiology & the department

Epidemiology academic information

Epidemiology faculty

Epidemilogy resources

sites of interest to Epidemiology professionals

Last Updated

10 Mar 2003

Source: Los Angeles Times, March 10, 2003


Fear Factor

Truth is, sunbathing is probably more dangerous than terrorism

By Cass Sunstein, Cass Sunstein teaches law at the University of Chicago and is the author of "Risk and Reason" (Cambridge University Press, 2002).

Risks are everywhere. There are plane crashes to worry about and random abductions, not to mention shark attacks, anthrax scares and snipers. These days, there are also orange alerts and yellow alerts and chilling warnings from the FBI that are making many Americans anxious, even hysterical.

But before reaching for the plastic sheeting and duct tape, shouldn't we be thinking carefully about just how real the risks are?

The truth is that, when it comes to risk, people often think poorly. Research shows that much of the time we fixate on bad outcomes without stopping to assess the probability that we will actually be harmed.

Sure, we want to be "safe" and "protected," but safety and protection are inevitably matters of degree. Often, we neglect the size of the risk altogether.

Consider the astonishing finding, from University of Pennsylvania economist Howard Kunreuther and his colleagues, that many people will pay the same amount for insurance against risks of 1 in 100,000, 1 in 1 million and 1 in 10 million. We don't have much experience in thinking about low probabilities like these, and so we pay no attention to differences that really should matter.

When our emotions are engaged, our judgment gets even more muddled. We focus on what looks like the worst case, giving no thought to the likelihood that it will occur. Vivid, dramatic images of harm -- hazardous waste sites, nuclear accidents, terrorist attacks -- can lead us to excessive fear of highly improbable risks. Social scientists find that when people discuss such risks, their concern usually rises, even if the discussion consists mostly of trustworthy assurances that the likelihood of harm is tiny.

But when we lack vivid images -- as in the case, say, of obesity or sun exposure -- we often treat the risk as if it were zero. The result is that we badly overestimate some risks and underestimate others.

Studies by psychologist Paul Slovic prove the point. Because grisly accidents are more dramatic than deaths from disease, most people think that accidents kill more people than disease. But the opposite is true.

Similarly, people mistakenly believe that more people die from homicides than from suicides. They exaggerate the annual number of deaths from dramatic and sensational causes, such as tornadoes and botulism. Remarkably, they perceive the risks from pesticides and herbicides to be higher than the risks from motor vehicles, alcoholic beverages and sunbathing.

Of course many of us try as best we can to factor in the likelihood of harm. But we aren't statisticians. So we tend to assess the question of probability by seeing if we can recall instances in which the harm actually materialized. That's why, as Slovic has also shown, in the aftermath of an earthquake, the number of people who buy earthquake insurance rises sharply, but that number declines steadily as vivid memories recede.

Then there are the risks that really can't be calculated. We don't know the probability of terrorists attacking Los Angeles, Chicago or New York. But when our behavior is driven by vivid images, we often make serious errors.

Imagine that you are about to fly to Paris and you are asked how much you would pay for an insurance policy covering losses "from all causes." Now imagine that you are about to take the same flight and you are asked how much you would pay for an insurance policy covering any losses "from terrorism." A moment's reflection should show that the first policy is better. It covers losses from terrorism and everything else. But as demonstrated by Columbia University's Eric Johnson and his colleagues, most people will pay more for the second policy because the word "terrorism" conjures up actual incidents of disaster.

If our behavior is affected by vivid images, we inevitably will be fearful of some trivial risks and neglect others that are actually serious.

There is a warning here. When newspapers, magazines and news programs stress highly improbable risks, people's concerns will be out of proportion to reality. And if government responds to unjustified concerns, we will spend too much of our time and money on pointless precautions. We might even take steps that increase the risks we face.

A tragic example comes from the aftermath of a train crash in Britain in 2000. As a result of the publicized crash, many people were fearful about railway travel and used the highways instead. The increase in car traffic is estimated to have led to five additional deaths in the first 30 days after the crash, nearly equal to the total number of deaths from train accidents in the previous 30 years.

For those interested in lengthening their lives, the lesson is simple: Focus on the actual likelihood of harm occurring. The millions of Americans who are devoting time and energy to duct tape and emergency supplies would be far better advised to lose weight, stay out of the sun, drive carefully and stop smoking.