Risk: A Practical Guide for Deciding What's Really Safe and What's Really Dangerous in the World Around You by David Ropeik, George Gray

• Table of Contents
• Read an Excerpt

Table of Contents

Read an Excerpt

INTRODUCTION

"I've developed a new philosophy . . . I only dread one day at a time."
— Charlie Brown

We live in a dangerous world. Yet it is also a world far safer in many ways
than it has ever been. Life expectancy is up. Infant mortality is down.
Diseases that only recently were mass killers have been all but eradicated.
Advances in public health, medicine, environmental regulation, food safety,
and worker protection have dramatically reduced many of the major risks
we
faced just a few decades ago.

Yet new risks have arisen. Hazardous waste. Nuclear power.
Genetically modified foods. Mad cow disease. Ozone depletion. Artificial
sweeteners. For all the unquestionable benefits of the modern technological
world and its scientific power, the march of progress that has given us
longer,
healthier lives has subjected us to new perils.
We often react to this conflict, of progress on the one hand and
risk on the other, with fear. Most of us are more afraid than we have ever
been. And not just from any single risk that happens to be grabbing the
headlines at a given point in time, whether it's terrorism or West Nile virus.
We are afraid, cumulatively, of all the new bogeymen to which our modern
existence has exposed us. Many polls find that people feel the world today
is
more dangerous for humans than it has ever been.
It is true that the industrial and information ages have spawned a
whole new range of risks, and raised awareness of those that were lurking
all
the time. But research suggests that our fears may not match the facts.
We
may be tooafraid of lesser risks and not concerned enough about bigger
ones. Polls show a wide gap between what the public and the "experts"
think
is actually dangerous and what is considered relatively safe. Who's right?
There are no simple answers.
But information can help us begin to sort things out. Some basic
facts about the risks we face, or think we face, can help us make more
sense of just what we need
to worry about. The intent of this book is to provide that information. We
want
to empower you to make better judgments about how to protect yourself
and
your family and friends. Our goal is to help you put the risks you face into
perspective.

Risk issues are often emotional. They are contentious. Disagreement is
often
deep and fierce. This is not surprising, given that how we perceive and
respond to risk is, at its core, about nothing less than survival. The
perception of and response to danger is a powerful and fundamental driver of
human behavior, thought, and emotion.
In writing this book, we tried to stay as neutral about these
controversial issues as we could. We think that information devoid of
advocacy is a tough commodity to come by these days, and will be more
useful to you. We do not tell you what you should think. Nor do we make
judgments about whether a risk is big or small for you as an individual. We
offer numbers for society as a whole, but there is no overarching single
conclusion about any risk that can be drawn for each reader. Each of you
has unique circumstances that make any given risk higher or lower for you
than it might be for the next person. Ultimately, how you perceive a given
risk
is a decision for you to make in the context of your own life. We simply
hope
that you are more able to make more informed choices after reading the
information we present. As Arthur Conan Doyle wrote in The Hound of the
Baskervilles, "That which is clearly known hath less terror than that which is
but hinted at and guessed."
We have gathered and analyzed the basic information available on
major risk issues and synthesized from all that research a fair presentation
that you can use to make up your mind about the risks we examine. Of
course we have made judgments along the way, about which risks to
include
or omit, about what information to offer and what information to leave out.
But
we have done so in an honest effort to get to the basic core truths about
each
risk as we see it, in as fair a way as possible. You may well disagree with
some of the judgments we've made. That's a risk we run in taking on a
subject fraught with so much emotion.
We encourage you to use this book in two ways. Reading it all
the way through will let you see how each risk compares with the others
and
will help you put them all in perspective. There are a lot of statistics in this
book. They are provided to give you an idea of how big or small each
individual risk might be. But they will also let you compare similar statistics
for various risks from chapter to chapter. Together, these numbers should
help you gain a larger view of many of the risks you face.
But we also encourage you to use this guide as you would an
encyclopedia, as a reference work you will turn to over time, whenever
there's
something about a particular risk you want to know. Each chapter, for
example, begins with a useful explanation of the specific hazard: What is
radon? How do air bags or nuclear power plants or cell phones work? What
are the most common forms of sexually transmitted or food-borne
diseases?
We hope this book remains valuable to you for some time. Yes,
the numbers of victims for various risks may change from year to year, and
we will certainly learn more about some risks than we know now. But the
nature of the consequences of alcohol consumption or radon exposure will
stay the same. Years from now the use of caffeine, the prevalence of heart
disease, the mechanics of the way radiation or lead or pesticides affects
us,
will all be pretty much the same.
We also hope you find this book useful no matter where you live.
While the numbers and exposure patterns we cite are focused on the
United
States, the details of most of the risks we explain are the same in Europe
or
Asia or South America. The effects of mercury, the science of genetic
modification of food, the persistence of some chemicals in the environment,
the way X rays work, are the same whether you live in Canada or France or
Japan. We recognize that the relative scale of risks varies from place to
place. The public health risk from cigarette smoke, for example, is higher in
Europe, where more people smoke, than in the United States. Firearms
risks
are higher for U.S. residents than citizens of any other country. At the time
of
this writing, mad cow disease is a higher risk in some nations than others.
So the data we use for exposure levels and numbers of victims, based on
statistics for the United States, may well vary for citizens of different
countries. But the general explanations of many of the risks we explore are
applicable for anyone, anywhere.

WHAT IS RISK?
Of all the wonders that I yet have heard,
It seems to me most strange that men should fear;
Seeing that death, a necessary end,
Will come when it will come.
— William Shakespeare, Julius Caesar

An anonymous writer once observed, "To risk living is to risk dying." Risk is,
indeed, inescapable. But just what is risk? How do you define it? To a
stockbroker it means the prospect of losing, or making, money. Same thing
for a person at the racetrack or at a blackjack table. For a skier or a bungee
jumper or a skydiver, on the other hand, risk has more to do with physical
than fiscal health. To the person taking a pill with known side effects, it's
about choice. To the person eating food with potentially harmful ingredients
that aren't listed on the label, it's about no choice.
At it's simplest, risk is the idea that something might happen,
usually something bad. But within that simple notion are some important
components that you need to understand in order to have a better basis on
which to make your personal risk judgments.
You may be hoping that this book answers the common question
we all have about most risks: "What are the chances that . . . ?" If you are
like most people, you think that risk means probability, the likelihood that
something will happen, as in "Your risk of dying from X is one in a million."
But there is more to risk than just calculating the statistical chances of a
certain outcome.
There is also the issue of consequences, as in "The likelihood of a
nuclear plant meltdown may be low, but it's a risk because it's disastrous if
it
does happen." A full definition of risk must take into account not just the
probability of an outcome, but its severity. Generally, risk involves an
outcome that is negative. You might say, "The odds of winning the lottery
are . . ." but you wouldn't say that winning the lottery is a risk. And the
more
severe the outcome, the higher we judge the risk to be.
A complete definition of risk must also include the presence of a
hazard, as in "That compound is a risk. It causes cancer in lab animals." If
something to which we're exposed isn't hazardous, it isn't a risk. We're all
exposed to a lot of cotton in the clothes we wear. So what.
Which brings up the fourth major component of risk, exposure, as
in "Flooding isn't a risk. I live on a hilltop." If a substance is harmful to test
subjects, but we're never exposed to it, it doesn't pose a risk. The risk of
being eaten by a shark doesn't exist in Kansas. A hazard can't do you any
harm if you are out of harm's way.
So a more complete way of thinking about risk might read: Risk is
the probability that exposure to a hazard will lead to a negative
consequence.
It's helpful to keep all these elements in mind when thinking about
risk. Take out any one of those components, and the definition is
incomplete.
Each one involves characteristics that help you understand risks more
completely and keep them in clearer perspective.
As an illustration, let's consider that dreaded common risk:
ketchup. If we are exposed to ketchup, that exposure alone doesn't make it
a
risk. As far as we know, ketchup isn't a hazard, except for the chance of
spilling some on your clothes.
But let's say somebody discovers that ketchup is hazardous. It
still isn't much of a risk if the consequence of exposure to this hazard is,
say, an increased taste for pickles on your hamburger. The nature and
severity of the consequence has a lot to do with judging whether a risk is
big
or small.
But let's say that you're allergic to pickles, so anything that
entices you to eat them could indeed be dangerous to your health. Ketchup
still isn't much of a risk if the probability of its leading to increased pickle
consumption is one in a million. You may have exposure to a hazard, but
the
level of risk still depends on the likelihood, the chance, that a negative
consequence might occur.
In other words, we can make better judgments about how to think
about risks if we keep in mind the ideas of hazard, exposure, consequence,
and probability. These characteristics help to define and explain the risks in
this book. Accordingly, most chapters are laid out as follows:
The Hazard: Just what is the agent we're talking about? (What is
asbestos?) How does this hazard come to be in the world around us? (How
does mercury get into our fish?) What is the biological or physical
mechanism by which the hazard does its supposed harm? (How does
radiation affect us?)
The Range of Exposures: How are we exposed to this risk?
Where? When? How do exposures vary over time, by location, or by
population subgroup?
The Range of Consequences: How much harm does the hazard
do? In what ways? To how many people? To what kinds of people? Who is
most at risk? Is the harm short-term or long-term, fatal or not? What is the
probability of harm? How many people are injured or killed by the risk?
Reducing Your Risk: In this section we offer some general
suggestions about what you can do to minimize the risk we're discussing.
For More Information: Each chapter ends with a list of resources
to provide you with more information.
Perhaps the biggest risk we take as authors is offering our
perspective and judgment of whether the risk is big or small, with visual
guides at the beginning of each chapter. This estimate is our best effort to
synthesize what we've learned on your behalf and to give you our opinion.
You will find two "risk meters" in each chapter. One will offer our
assessment
of the general likelihood of exposure to hazardous levels, taking into
account
the factors of exposure and hazard from our definition. The other meter will
indicate our assessment of the risk's consequences—including severity
and
number of people affected.
Here are a few examples of what you will see in each chapter. At
the beginning of Chapter 1, "Accidents," the first risk meter will look like
this:

<illustration>

The upper bar indicates that the likelihood of exposure to accidents in a
way
that will probably cause harm is high. The lower bar indicates that the
consequences of the risk of accidents—the severity of the outcome and
how
many people suffer these consequences—are also high. But not quite as
high as the first meter, because the majority of accidents are not fatal, so
the
severity of the consequences brings the rating down a bit.
Here's what the meter will look like for Chapter 35, "Radon."

<illustration>

The likelihood of exposure to levels that will probably cause harm is, in
general, pretty low. Lots of people are exposed, but the levels in most
cases
are fortunately not usually enough to cause harm. So there is exposure, but
not to levels that present a hazard in a lot of cases. But since the
consequences of radon exposure at levels high enough to do harm are
potentially severe, and several thousand Americans a year suffer those
consequences, the lower bar for radon takes account of both those factors.
These meters require several cautions. First, they refer to the
population as a whole. Your risk is almost certainly different from that of the
general population because of your age, gender, genetics, income,
education, location, and other factors that make you unique. These risk
meters offer only a general reference to where we think the risk falls on the
high-low scale. Second, these are estimates. They are not scientific. They
are the result of our analysis of the information we've collected and are not
statements of fact and truth. And since there is a lot of uncertainty about
many of the risks in this book, the meters are ballpark estimates that offer
only a general range of where we think the risk falls. That's why we don't
give
our ratings specific numbers between 0 and 10. (In Appendix 2 we discuss
our thinking behind each of the ratings. The appendix does offer our ratings
numerically, though some of them are given as a range rather than as one
specific number.)
Further, these risk gauges don't take into account the benefits
that come from the hazard being discussed. Air bags can be harmful, for
example, but clearly they save many more lives than they take. Some
people
suffer serious side effects from vaccines, but vaccination's benefits far
outweigh the risks. We leave that risk-benefit accounting out of our
judgment.
Since our definition or risk presumes that most people think of risk as
resulting in a negative consequence, that's what we rate.
Not every risk meter offers a ranking. Instead, a question mark
acknowledges that some risks are too new or poorly studied to rate. As an
example, for endocrine disrupters, the meter looks like this:

<illustration>

In a few cases the risk meter will be blank. That's because there
is no specific hazard, so the first bar is irrelevant. A few of the risks we
discuss, like cancer and heart disease, are really outcomes. That is, you're
not exposed to cancer or heart disease. You just end up with these
illnesses
as a result of other processes. But because they kill so many people, and
are so often the result of exposure to many of the hazards we discuss, we
think they deserve explanation in a book about risk.

WHERE DO "THE FACTS" ABOUT RISK COME FROM?
"There is something fascinating about science. One gets such wholesale
returns of conjecture out of such a trifling investment of fact."
— Mark Twain

As we try to judge what's risky and what's not, we look to science for
answers. But even with all the facts that science can provide, much
uncertainty remains, for a number of reasons. First, the sciences by which
risk is investigated—toxicology, epidemiology, and statistical analysis—are
inherently imprecise. Second, there are a lot of risk questions science
simply
hasn't asked yet. New risks like using a cell phone while driving or eating
genetically modified food haven't been studied nearly enough for us to have
all the answers. And third, even for risks that have been studied, the facts
as
we know them are constantly changing as scientific answers to one set of
questions reveal more questions.
A lot of information in this book comes from the findings of the
three major risk sciences. It's important for anyone trying to make informed
judgments about risk to understand what these sciences can, and cannot,
tell us.

Toxicology
Most simply described, toxicology is the study of poisons. But because of
that very definition, you can understand why toxicologists usually can't test
the agent they're investigating on human subjects. So animals are used as
surrogates. But toxicologists admit that they can't say for sure what a
compound will do in humans based on evidence of what it does in animals.
As one toxicologist says, "With stuff that might kill people, animal testing,
as
imprecise as it is, is the best we can do. But despite what you might think
of
your boss or some people you don't like, humans aren't rats." Toxicologists
don't know which lab animal species serve as the best indicators of what
would happen in people, nor do they know which species are better
indicators for which kinds of hazards. So extrapolating from lab animals to
humans is imprecise. As one example, cyclamate, an artificial sweetener,
causes one type of liver tumor in only one species of rat, and then only in
males, and doesn't cause it in any other test animals. Yet test data from
the
experiments on those rats caused the food additive to be banned for human
consumption.
Another imprecision from toxicology arises because testing of lab
animals often involves subjecting the animals to massive doses of an agent.
In testing for carcinogenicity, animals routinely get doses, each day, far
greater than you would be exposed to in your entire lifetime. Toxicologists
call this dose the MTD, for "maximum tolerated dose." They use this
technique when testing for cancer in order to maximize the chance that
they'll find any effect that might occur and that might not show up from a
milder dose.
Using these MTDs, toxicologists presume that if the substance
they're testing causes an effect at a high dose, it might cause the same
effect at a lower dose. This approach seems like a rational way to deal with
potentially dangerous chemicals and other agents; if high doses cause
harm,
assume that low doses might too. But sometimes the size of the dose is
what's really causing the harm. Think of aspirin, for example. One or two
aspirin are fine. Too many will kill you. The standard toxicological approach
of
subjecting lab animals to high doses of a test compound can reveal subtle
effects, but it can also produce misleading results.
(Toxicological tests for noncancer health problems, like
developmental hazard or cognitive impairment, don't rely on the MTD
approach. For these outcomes, scientists assume that higher levels cause
worse effects and lower levels cause weaker effects, and below a certain
level
the hazard might not cause an effect at all. So they subject the test
animals
to varying doses to find the lowest one at which an effect occurs. Also, for
these other health problems, scientists believe that once exposure to the
agent stops the effect usually goes away. But for cancer, since just a one-
time mutation of a single gene can create permanent changes to the DNA
that causes the disease, toxicologists use the MTD method, a more
conservative and precautionary approach.)
A further imprecision arises in toxicology because in vivo tests in
living lab animals, or in vitro tests of cells in a lab dish or beaker, isolate
and
test just one compound at a time. That's a smart way to find out with
precision whether that particular agent is hazardous. But in the real world
we're exposed to a stew of agents, and the mix can lead to different
outcomes than exposure to any individual component. (Radon and
smoking,
for instance, apparently work synergistically and increase the risk of lung
cancer more than the sum of one risk plus the other risk.) In addition, while
the environment in the lab is stable and uniform, the real world is full of
variables such as our environment, our health, our food, our emotional
states,
and our genetic makeup from one generation to the next and from one
person
to the next. These factors and many others affect how we react to a
compound or circumstance.
In short, while toxicology can tell us a lot about the biological
hazard of a particular chemical or element or compound, it can't tell us with
absolute accuracy just what the substance being tested—at high doses to
another species in a controlled lab—will do at lower doses to humans in the
complicated real world.

Epidemiology
When we can't test a substance or hazard on people but we want to know
whether it might be a threat to public health, we look around for
circumstances in which people might already have been exposed. Studying
what has happened, or is currently happening, to real populations in the real
world, and trying to make sense of which hazards and exposures might be
associated with which consequences, is the essence of epidemiology.
Like toxicologists, epidemiologists readily acknowledge that their
science is imprecise. Epidemiology can usually provide only associations,
not absolute proof, that some particular exposure may be what's causing
some particular consequence. For example, in one kind of study
epidemiologists investigate a specific small group of people who get sick.
The book and movie A Civil Action, for instance, made famous the polluted
drinking water in Woburn, Massachusetts.
A higher-than-expected number of cases of childhood leukemia
showed up in just a few years in a small neighborhood. Epidemiologists
investigated to find out what sources of exposure to potential hazards the
neighbors shared. They discovered that one thing the neighbors had in
common was that those who drank from a certain water supply had a higher
rate of illness. Therefore, something about the water was the likely cause of
the leukemia. They tested the water for chemicals suspected to cause that
illness and estimated how much of the water people drank, for how long,
and
how polluted it was when people drank it. In the end, a peer-reviewed
epidemiological study showed an association between how much of the
well
water pregnant mothers drank and the frequency of childhood leukemia in
their offspring. The more they drank, the more likely it was that their
children
developed leukemia.
But that's not proof. Perhaps a couple of the neighbors were
exposed to something else the researchers didn't ask about. Maybe the
researchers never detected something else in the well water. These other
factors are known as "confounders," hidden clues that can muddy the
epidemiological waters and lead to an inaccurate assumption that A caused
B. Hidden confounders can never be completely ruled out.
Epidemiologists can also do a different type of study, not just
looking back in time at a small local group of people over just a few years
but
tracking a much larger population forward over longer periods of time. The
famous Framingham Heart Study is an example of this kind of
epidemiology,
following an entire community over decades. Again, the researchers
examine
these wider populations for patterns in illnesses and exposures that
suggest
an association between the two. For instance, many epidemiological
studies
show that there is a strong association between air pollution increasing one
day and hospital admissions for respiratory problems going up over the
following several days, which suggests that the pollution is probably
causing
the respiratory problems. But again, that apparent association is not the
smoking gun of absolute proof. Only when many long-term studies of
different
large populations repeatedly show the same thing, as with tobacco
smoking
and lung cancer, can epidemiology confidently say A causes B.
This isn't to suggest that the findings of epidemiology are weak or
of little use in judging risks. In good epidemiological studies, researchers
give
the research subjects in-depth questionnaires about their health, their
lifestyle, their diet, their social and economic characteristics, even their
residential history (where they have lived and when), trying to rule out all
confounders. They compare a group of people suffering some kind of health
problem, like those families in Woburn, with other "control" groups,
populations of similar size and socioeconomic status somewhere else, who
presumably were not exposed to the same things. For the bigger long-term
population studies, epidemiologists carry out multiple research programs in
different places at different times to see if their results agree. With such
techniques, epidemiologists can rule out every other possibility they can
think of. They can become more and more certain of the associations they
find.
But, like toxicologists, they can rarely be completely sure.

Statistical Analysis
In addition to the findings of toxicology and epidemiology, risk analysts also
look for their clues among large sets of statistics. Those data collections
are
compilations of real-world information, on either morbidity (nonfatal health
problems) or mortality (deaths). These databases can offer rich details, like
how many people were injured or killed in motor vehicle accidents,
categorized by speed, vehicle size, whether the victim was male, female,
old,
young, wearing a seat belt or not, and so on. There are data sets on
hundreds of risks that offer information on the age, gender, and race of the
affected population and the circumstances that led to the death or illness,
such as the number of food poisoning cases connected with restaurants, or
the number of workers murdered on the job. Other data collections provide
risk analysts with information about hazardous materials emissions, local
water or air pollution levels, or the presence of harmful chemicals in our
blood
or the food we eat. These details all offer insights about the hazard,
exposure, consequence, and probability of various risks.
But the numbers in these data collections usually suffer from
some imprecision. Not everybody who suffers food poisoning after dining at
a
restaurant, for example, actually goes to a doctor to report his illness. Not
every police officer fills out every last detail on every accident report. Not
every factory keeps accurate, or honest, records of its emissions. And not
every government information collection system gathers the information and
enters it into its database accurately.
Numbers are also subject to interpretation. Here's an example.
According to national motor vehicle crash statistics, drivers 75 years old or
over are involved in four times as many fatal crashes as the average of all
other age groups. But does that mean that elderly drivers are killing other
people, or just that because of frail health they're more likely to die
themselves whenever they're in a crash? You can't tell by that statistic. The
numbers don't tell you everything you need to know. As Mark Twain
said, "There are three kinds of lies—lies, damned lies, and statistics."
Finally, no matter how precise and narrow statistical categories
are, they lump everybody in that category together. For example, federal
motor vehicle crash statistics group data by age, gender, the day and time
of
crashes, and the kind of vehicle involved. So you can determine how many
15-
to 24-year-old males were involved in crashes on Sundays at 5 p.m. in
pickup trucks. As narrow as that seems, that's still a large group of people
and not everyone in it is the same. Individuals within that group have all
sorts
of differences in health, lifestyle, education, genetics, body size and shape,
and on and on.
Risk statistics are generalities, and by definition cannot
specifically answer the question we all want answered: "What is the risk to
me?" You will read a lot of numbers in this book. As we've stated, because
you are unique none of those numbers will accurately and precisely answer
your question. Risk numbers can be only a general guide. They give you a
sense of which risks are bigger and which ones are smaller, and
sometimes
they can tell you which risks are higher or lower for the demographic groups
to which you belong. But even risk numbers that define the categories as
narrowly as possible still can't calculate the risk for each unique individual.
In sum, the sciences that supply the facts about risk, while
growing more and more powerful, are still imprecise. They can provide us
with
valuable insights. But their results are uncertain and open to interpretation.
There are very few unequivocal answers when it comes to defining and
quantifying the risks we face. That's why in this guide our approach is to
offer
information in ranges: the range of exposures, the range of consequences,
and so forth.

In addition to this scientific imprecision, sometimes we can't tell whether a
risk is big or small, or real at all, simply because it's too new and hasn't
been studied enough. Our modern world presents us with many new
technologies (cell phones) or processes genetic modification of food) or
compounds (statin drugs to reduce cholesterol) that have profound benefits,
but which also come with risks. Sometimes we are exposed to these
technologies or processes or compounds before the risks have been
adequately studied.
In the professional and policymaking world of people who deal
with
risk, how we should handle this uncertainty is a hotly debated issue. Some
people argue that we should thoroughly study anything that might pose a
risk
before we start to use it.
The people on this side of the argument heed the advice of the
eighteenth-century British politician Edmund Burke, who said, "Early and
provident fear is the mother of safety." They suggest that we should adopt
as
a matter of law the "Precautionary Principle," the academic term for what
most people think of as "Better Safe Than Sorry." These advocates argue
that
the best way to protect human and environmental health is to treat new
compounds or technologies as guilty until proven innocent. They say that
while we do this with some things, like new drugs, we don't do it with
others,
like new industrial chemicals. Advocates of the Precautionary Principle say
that we must apply this careful approach across the board.
Others might subscribe to the advice of the American essayist
Randolph Bourne, who wrote in his 1913 book Youth and Life, "We can
easily
become as much slaves to precaution as we can to fear. Although we can
never rivet our fortune so tight as to make it impregnable, we may by our
excessive prudence squeeze out of the life that we are guarding so
anxiously
all the adventurous quality that makes it worth living." These opponents of a
sweeping Precautionary Principle argue that it would deny society many of
the benefits of new technologies for years, even decades, until thorough
scientific study can be completed. Those that argue against the
Precautionary Principle also point out that almost anything carries some
risk.
Under the most rigorous application of the Precautionary Principle, these
people claim, it would be hard to approve such things as motor vehicles or
prescription drugs or vaccines. They argue that while it makes common
sense to err on the side of caution, we should assess risks on a case-by-
case basis, rationally weighing them against benefits. They say a blanket
Precautionary Principle might deny society a public health advance that
could save lives before all the scientific answers are in.
There are also times when we think science has come up
with "the" answer, and we're reasonably certain about just how
precautionary
to be. And then things change. Even for risks that have been well studied,
the
facts are always evolving. We learn more and more every day. Between the
time this book goes to press and the time you read it, our knowledge of the
effects of hormone-disrupting chemicals will change. We'll certainly know
more about human genetics and the risk of some diseases. We'll probably
know more about the actual levels of particle pollution in the air we breathe.
The statistical trends on established risks will have almost certainly shifted.
Further, our world of rapid technological development means that
new risks are being created, and new solutions are being found, at an
accelerating rate. Many of the risks in this book were largely unknown just
a
few years ago. Researchers are constantly developing new technologies or
drugs that reduce some risks while potentially creating others. And science
itself changes and grows more powerful. We can detect chemicals in our
blood or in the air at levels much lower than we could just a decade ago.
Risks that were always out there are just now being revealed. And all of
these discoveries and new risks and new solutions interact in highly
unpredictable ways. Methyl tertiary-butyl ether, or MTBE, was added to
gasoline to improve air quality. But that policy led to the pollution of drinking
water. The certainty we want in order to know how to judge risks is a tough
ideal to achieve in our modern, dynamic world.
In short, our effort in this book to give you an accurate, reliable
explanation of many of the risks you face is tantamount to shooting at a
moving target. Which makes trying to score a bull's-eye with everything we
present here a very risky proposition.
In an additional effort to ensure that what we present is accurate,
thorough, and balanced, we asked experts to review each chapter. These
reviewers were academics, scientists, doctors, government officials,
engineers, risk assessors, and members of advocacy groups like the
American Cancer Society and the Union of Concerned Scientists. We
asked
them to check our facts, to correct mistakes, to point out omissions, to
clarify—and for feedback on whether we had provided a thorough and
balanced overview of the issue. Their input was immensely valuable. But we
take full responsibility for the final product. And the reviewers had nothing at
all to do with the risk ratings in each chapter. We offer the names and
backgrounds of the reviewers at the end of each chapter.

WHERE DO OUR FEARS ABOUT RISK COME FROM?
"People are disturbed, not by things, but by the view they take of them."
— Epictetus

As we wrote earlier, the facts about risk are only part of the matter.
Ultimately we react to risk with more emotion than reason. We take the
information about a risk, combine it with the general information we have
about the world, and then filter those facts through the psychological prism
of
risk perception. What often results are judgments about risk far more
informed by fear than by facts.
The terrorist attacks on the World Trade Center in New York and
on the Pentagon and the subsequent anthrax attacks in the fall of 2001 are
an example. Many of us were afraid, and rightly so. But some people
responded by driving to a distant destination rather than flying, even though
the facts clearly showed that flying remained the far safer mode of
transportation, even after September 11. Some people bought guns, raising
their risks from firearms accidents far more than reducing their risk of being
attacked by a terrorist. Many people took broad-spectrum antibiotics even
though they had no evidence that they had been exposed to anthrax—but
they didn't get an annual flu shot.
Do these judgments make sense? Are they rational? Not based
simply on the facts. But this is how humans respond to risk . . . with our
hearts as well as our heads.
The psychological study of this phenomenon, known as "risk perception,"
explains why our fears often don't match the facts. It is perhaps the biggest
reason why writing this book is a risky affair.
We're confident that as you read this guide, your interpretation of
what we say about various risks will differ from ours, and from some of your
friends or family or neighbors. Same facts. Different interpretations. Despite
our efforts to be neutral, many of the issues we write about are highly
emotional and trigger powerful risk perception responses that all but
guarantee that you might not like, or agree with, all of what you're about to
read. We think it's valuable to understand what researchers have learned
about risk perception because it might help you understand your own
reactions to risk a little better.

Risk Perception
Humans tend to fear similar things, for similar reasons. Scientists studying
human behavior have discovered psychological patterns in the
subconscious
ways we "decide" what to be afraid of and how afraid we should be.
Essentially, any given risk has a set of identifiable characteristics that help
predict what emotional responses that risk will trigger. Here are a few
examples of what are sometimes called "risk perception factors."

• Most people are more afraid of risks that are new than those they've lived
with for a while. In the summer of 1999, New Yorkers were extremely afraid
of
West Nile virus, a mosquito-borne infection that killed several people and
that
had never been seen in the United States. By the summer of 2001, though
the virus continued to show up and make a few people sick, the fear had
abated. The risk was still there, but New Yorkers had lived with it for a
while.
Their familiarity with it helped them see it differently.

• Most people are less afraid of risks that are natural than those that are
human-made. Many people are more afraid of radiation from nuclear waste,
or
cell phones, than they are of radiation from the sun, a far greater risk.

• Most people are less afraid of a risk they choose to take than of a risk
imposed on them. Smokers are less afraid of smoking than they are of
asbestos and other indoor air pollution in their workplace, which is
something
over which they have little choice.

• Most people are less afraid of risks if the risk also confers some benefits
they want. People risk injury or death in an earthquake by living in San
Francisco or Los Angeles because they like those areas, or they can find
work there.

• Most people are more afraid of risks that can kill them in particularly awful
ways, like being eaten by a shark, than they are of the risk of dying in less
awful ways, like heart disease—the leading killer in America.

• Most people are less afraid of a risk they feel they have some control over,
like driving, and more afraid of a risk they don't control, like flying, or sitting
in
the passenger seat while somebody else drives.

• Most people are less afraid of risks that come from places, people,
corporations, or governments they trust, and more afraid if the risk comes
from a source they don't trust.
Imagine being offered two glasses of clear liquid. You have to drink one.
One
comes from Oprah Winfrey. The other comes from a chemical company.
Most people would choose Oprah's, even though they have no facts at all
about what's in either glass.

•We are more afraid of risks that we are more aware of and less afraid of
risks that we are less aware of. In the fall of 2001, awareness of terrorism
was so high that fear was rampant, while fear of street crime and global
climate change and other risks was low, not because those risks were
gone,
but because awareness was down.

•We are much more afraid of risks when uncertainty is high, and less afraid
when we know more, which explains why we meet many new technologies
with high initial concern.

• Adults are much more afraid of risks to their children than risks to
themselves. Most people are more afraid of asbestos in their kids' school
than asbestos in their own workplace.

• You will generally be more afraid of a risk that could directly affect you
than
a risk that threatens others. U.S. citizens were less afraid of terrorism
before
September 11,
2001, because up till then the Americans who had been the targets of
terrorist attacks were almost always overseas. But suddenly on September
11, the risk became personal. When that happens, fear goes up, even
though
the statistical reality of the risk may still be very low.

People who first learn about these risk perception patterns often
remark on how much sense they seem to make. It's little wonder. These
are
deeply ingrained patterns, probably ancient behaviors imprinted in us over
millions of years of evolution. Long before we had our modern thinking brain,
long before humans or primates even developed, only organisms that could
recognize and successfully respond to danger survived and evolved. In
Darwinian terms, these affective, "irrational" ways of protecting ourselves are
adaptive. They help us preserve the species. Evolution selects for this type
of
behavior. That belief is supported by the fact that these patterns of risk
perception cross cultures, age groups, genders, and other demographic
groupings. There are some variations among individuals. Those variations
make sense too because different people have different lives, different jobs,
different family circumstances, different sets of experiences, different sets
of
values, and so on. Fearing a risk more if it involves children, for example,
means parents will react differently from, say, teenagers. What is
frightening
to you might not be to your friend. Neither of you is right or wrong. You just
each have a unique perspective on the same statistics and facts. But risk
perception research shows that underneath our individual differences, we
share certain patterns of risk response.
As we've written, this way of protecting yourself can be
dangerous. What feels safe might actually be dangerous: driving instead of
flying, antibiotics against anthrax instead of flu shots, arming yourself
against
a phantom risk. We explain risk perception, therefore, to help you
understand
the psychological roots of how we all respond to risk. That might help you
understand your own concerns and put the risk issues in your life into
clearer
perspective.

THE RISKS WE INCLUDE, AND THOSE WE DON'T
This book does not include or omit risks based on whether they are "real."
As
we have mentioned, whether a risk is real is ultimately something you will
decide for yourself.
Rather, what follows in this book are many of the major risks you might
want
to learn about. In selecting what to include, we made no distinction among
those that have been largely debunked (radiation from microwave ovens),
those that are new (cancer from cell phones), those that remain poorly
understood (hormone-disrupting chemicals), or those that have been well
studied (electrical and magnetic fields from electricity lines and
appliances).We include risks with high likelihoods, such as food poisoning,
and low likelihoods, including cancer from pesticides on food.
That said, we do not include other risks. For example, we do not
explore risks to the environment that don't have direct implications for
human
health. While risks like climate change, acid rain, and destruction of
wetlands or forests all impact the biosphere on which we depend, they do
not
have a direct and immediate connection to human health.
We also do not include many of the risks that arise in detailed
medical care. We do discuss medical errors in general as a category of
risk,
and we also explore some broad medical issues, such as vaccines or
antibiotic resistance. We write about some health risks, like heart disease,
cancer, and obesity. But we do not explain risks from drug reactions, the
relative risk of one form of medical treatment over another, or many other
specific medical risks. These risks are so unique to each individual that to
discuss them in a book about risk in general might in fact be dangerous to
the reader.
Nor do we discuss the risks of developing health problems
because of genetic predisposition, which has more to do with susceptibility
than cause. We don't include many things that you might think of as risks,
but that are actually outcomes. Stroke and diabetes, for instance, are end
results, the outcomes of natural biological processes or, in some cases, of
exposure to a hazard. This book deals with the hazards to which we are
exposed, because our actions have bearing on these parts of the risk
equation. By the time we're faced with an outcome, it's too late. Still, we do
include a few outcomes—heart disease, cancer, obesity—because they are
such major killers and such common outcomes to many of the hazards we
write about that we felt a general explanation of these issues would help.
We also don't go into detail about the risks of crime, a complex
and unique set of issues. Some basic crime statistics are included in
Appendix 1 in the back of the book. We leave for this appendix those risks
that, while of interest, really don't need much of an explanation, like the risk
of being hit by lightning, being killed in a plane crash, of snakebite or bee
sting or shark attack. We also list in that appendix the statistics for some
common causes of death and injury, such as stroke or homicide, diabetes
or
drowning, asthma or Alzheimer's disease. This appendix is simply a
numerical listing of how many Americans suffer these outcomes each year.
It
is not a chart of your individual risk.
As the historian Edward Gibbon wrote, "The laws of probability. So true in
general. So fallacious in particular." Your individual risk depends on dozens
of
factors unique to your lifestyle, genetics, socioeconomic characteristics,
and
so on.

In the end, we hope we came up with a list, and an approach, that offers a
review of what is known, and what is not known, about most of the risks
that
most people care about.
We have tried to offer information in a neutral way, keeping matters simple
by
culling the essentials from the mountains of information on each risk. At the
same time, we have rigorously pursued accuracy, and we've tried to offer
some context and richness of detail. We hope this guide provides useful
information that will help put the risks in your life in perspective. We hope it
helps you lead a healthier, safer, less worried life.

Copyright © 2002 by David Ropeik and George Gray.
Reprinted by permission of Houghton Mifflin Company.