Peter Sandman
Comments and Questions
(and Some Answers)
2006

Any suggestions?

1. It’s crucial to acknowledge the yuck factor. Disgust is a close cousin of dread and a well-established component of what I call outrage. And disgust is a nearly universal reaction to mass animal destruction. (It’s a common reaction to normal slaughtering methods too, for that matter.) People’s response to their own disgust is likely to be bimodal: (a) “If you've got to do something that disgusting, it must be really serious!” or (b) “It can’t be right to do something that disgusting – you must be overreacting!” In many cases it would be wisest to ally with (a) – that is, to use disgust at your precaution as evidence of the seriousness of the risk. “Things are so bad we feel we have no choice but to....” Sometimes you need to disavow both (a) and (b), which is harder: “I know this may seem like a very extreme response to a fairly routine outbreak. The photos are so gross! Even I have trouble looking, and I’ve been working on animal health for decades. But this is actually Standard Operating Procedure for situations of this sort....” What’s never wise is to ignore people’s disgust and act as if that weren't our reaction.

2. It’s important to acknowledge that some people question the ethics of mass animal destruction altogether. Animal rights groups rarely encounter a more responsive public than when they oppose these sorts of culls. Those managing the culls should try to acknowledge the existence of opponents even before the opponents show themselves. “The bottom line is that we’re killing animals in order to [protect human health] [prevent the spread of an infectious animal disease] [both]. We’re even killing healthy animals in order to stop the chain of contagion. The science that says this is the most effective way to stop the outbreak is extremely strong. But of course we never, ever use this strategy with humans, and there are some who believe that animals should have much the same rights as humans – including the right not be sacrificed this way for the greater good. We do it anyway – but with real sorrow, real awareness of what we are doing, and real respect for the beliefs of those who think we shouldn’t do it.” (There’s a seesaw at work here. The more respectful the authorities are of the views of animal rights advocates, the likelier the rest of us are to decide, albeit reluctantly, that other priorities must come first.)

3. Actions that appear callous or disrespectful should be avoided if possible – photos reminiscent of concentration camps; use of bulldozers; macabre humor on the part of those administering the cull; etc. If they can’t be avoided they should be acknowledged. “I know this looks awfully callous. It was necessary, but it feels so wrong!” Get on the other side of that seesaw too.

4. A narrower point that’s extremely important right now vis-à-vis bird flu and H5N1-related bird culls: Don’t pretend that what you’re doing is in the farmer’s interest when it isn’t. My wife Jody Lanard and I have done some work (Jody more than I) with international agencies and Asian governments wishing to encourage farmers to cooperate with culls. They often try to claim that an H5N1 outbreak is a serious threat to the health of the farmer and his family. This is simply false. Bird-to-human transmission of H5N1 remains difficult; only a few hundred cases have cropped up in the face of millions of opportunities. Interacting with sick birds is nowhere near as dangerous to the farmer as losing his livelihood. And of course there is no “business case” whatever for a farmer whose birds are healthy to destroy his own livelihood in order to create a cordon sanitaire and stop the spread of infection. That’s wise for the world at large, for reasons of poultry industry prosperity as well as pandemic prevention. But it is surely a net loss – a catastrophe, in fact – for the farmer, his family, and his village. We have advised clients to say exactly that. (It’s not a secret; claims to the contrary are transparently false.) Of course what makes most sense is for the developed world to subsidize compensation to developing world farmers, whose sacrifice helps to protect us all. If compensation isn’t in the cards, there are two other options: appeals to altruism and coercion. Dishonest claims that the farmer ought to want to cull his flock undermine altruism and add insult to the injury of coercion.

That’s what comes immediately to mind. I'll look forward to seeing your article.

When counterterrorism works, it appears excessive and unnecessary. The need for counterterrorism measures is only apparent when they fail. This is a perpetual strategic dilemma that demonstrates why LEGITIMACY is the key issue in all terrorism/counterterrorism efforts. Counterterrorism can only succeed when the people have faith in their government. In fact, promoting government legitimacy is the only strategically decisive goal of counterterrorism. The goal of terrorism is to destroy that bond, not just to kill people or destroy infrastructure. This is hard for even experts to understand, and harder still for them to articulate.

Am I wrong or does this sound like a risk communications challenge?

In fact, Dr. McIntyre’s entire short article framed a risk communication challenge – the challenge of reconciling the threat of violent Islamism with the organizing “narrative” of Western thought (especially, I think, the West’s self-image).

A step in the right direction (the step Dr. McIntyre seems to be trying to take) is dilemma-sharing. We need to be saying much more explicitly that we see an extraordinary conflict between our vision of how differing religions should share the world and our understanding of how to meet the Islamist terrorist threat. We need to say that we’re having enormous trouble reconciling the two. We need to explain – mostly to ourselves – how painful it is to choose between entertaining a much more illiberal vision and denying ourselves some pretty important weapons against Islamist attack.

At least one thing wrong with the Pope’s speech in Regensburg (cited by Dr. McIntyre) was his failure to express any sense of the painful dilemma that might require reinstating the “old narrative” he was advancing. He sounded, perhaps accurately, like the old narrative had always been his preferred narrative, and would have been his preferred narrative even if Islamist terrorism had not become a force to reckon with.

I think President Bush’s advocacy of various civil liberties infringements has been similarly deficient. He certainly leaves me feeling that he doesn’t see the dilemma – that he is quite comfortable abandoning these liberties in order to fight Islamism more effectively, perhaps even that fighting Islamism is merely an excuse for doing what he’d have wanted to do anyway.

Our answer to the question, “Should we grant liberal freedoms to those who deny the validity of those freedoms and would not grant them to us if in charge?” has almost always been a resounding “Yes!” That answer is relatively low-cost when liberal values are securely in charge. Can we still afford to give the same answer? We can’t sensibly address that question unless we acknowledge both that the price seems to be higher than it was and that it may nonetheless be a price worth paying.

We need a debate that starts with consensus about a few points:

• We cherish our liberal “narrative” – our sense of ourselves as essentially tolerant – and are rightly extremely reluctant to give it up.
• That narrative may now be getting in the way of repelling the threat of violent Islamism.
• The crucial question is whether we should move to a new narrative; or whether we must address the Islamist threat handicapped by our deeply felt values; or whether we can somehow forge a compromise or a short-term exception.
• The debate over how best to answer this question is deeply divisive internally, but that debate must nonetheless take place. Those who don’t see some truth and some virtue on both sides of the debate – who don’t see that there is a painful dilemma here – will be little help in finding a resolution.
• This is not an entirely new situation. We had a similar debate with respect to our battle with Communism, for example, though religion was not then at the core. Part of our debate should be about what lessons to draw from history.

It’s going to take more than good risk communication to address this dilemma effectively. But it’s certainly going to take good risk communication.

It appears to suggest an access of good sense before the U.K. goes the way of countries like Sweden, where, in some respects, the population have lost their edge because of over-protection by government and regulation. There is considerable discussion of risk and how it should be managed and who should take responsibility for it – including the radical suggestion that citizens might well be able to deal with more than anxious politicians want to let them.

What do you think? I still visit your website and continue to value your wisdom!

I do think the report would have benefited from some risk communication expertise – especially a better understanding of the reality that people don’t (and in my judgment shouldn’t) judge risks purely in terms of what I call the “hazard.” People’s choices about which risks to try to mitigate are always going to be responsive to “outrage” as well as “hazard.”

In my terms, the BRC wants to reduce regulatory attention to low-hazard high-outrage risks. This is possible, I think, only if the outrage is well-managed. When outrage is ignored and allowed to build, the pressure to over-regulate hazard builds right along with it. Over-regulating a small hazard doesn’t actually mitigate the outrage much; an outrage problem needs an outrage solution. (For a long disquisition on this complicated issue, see “Because People Are Concerned: How Should Public Outrage Affect Application of the Precautionary Principle?”.) But ignoring a big outrage because it’s a small hazard isn’t a solution either.

The problem of our over-regulation of low-hazard high-outrage risks is very different from the “nanny state” problem – which stems mostly from “over-regulation” of high-hazard low-outrage risks, especially those that are voluntary and pleasurable.

New York City, for example, recently proposed outlawing the use of transfats in restaurants. The data that transfats contribute importantly to mortality are pretty unassailable, I think. But so are the data that transfats taste good; the data that many people, given the choice, will take the risk; and the data that those who wish to avoid transfats are able to do so already without further regulation.

I certainly agree with the BRC that encouraging people to make their own risk management decisions is an important piece of progress. But it won’t necessarily make regulations more cost-effective with regard to lives saved per unit of cost (or per unit of bureaucratic interference). If those are the right metrics, the New York City transfat regulation is probably sound.

In other words, I think the BRC report would be a better guide to regulatory reform if it distinguished more carefully the two key problems it addresses:

• Public demand for a governmental response to low-hazard high-outrage risks – which is a misuse of resources insofar as it doesn’t significantly reduce hazard (and may actually increase hazard), but which addresses a public mandate that is genuine and understandable.
• Regulatory inclination to interfere in people’s lives by regulating voluntary behavior that incurs significant hazard, but does so without significant public outrage (indeed, the public may be outraged mostly at the regulation itself).

Of course the lowest-hanging fruit is the over-regulation of low-hazard low-outrage risks – where bureaucrats run amok and do something that addresses neither a serious hazard nor a serious outrage. An example in the BRC report: a regulation requiring people to remove doormats from their entryways because of a hypothetical tripping risk.

But once you’ve gotten rid of such totally useless regulations, you’re left with two real policy dilemmas: what to do when people are really upset about a tiny threat to their health, and what to do when people are perfectly happy to endanger themselves seriously.

If you were writing a memo to Rupert Murdoch in the wake of the O.J. Simpson book/TV special controversy, what communications recommendations would you give for both Rupert and News Corp. moving forward?

But the outrage at News Corp. for its role in the deal, and the insistence that it back out, surprised me.

Though it surprised me, the outrage certainly didn’t offend me – as any attempt at government censorship would have, or as a violent response like the response to the Muhammad cartoons would have. It’s not “censorship” for people to react to a publication they dislike by shunning and shaming those involved, and thus costing them reputation and profitability. And I approve of corporate “self-censorship” in response to public outrage. That’s exactly how the system is supposed to work. When stakeholders disapprove of something a company has decided to do, the company gets to choose whether to do it anyway and endure the costs of the disapproval or to defer to its stakeholders’ values and change its decision. Of course the choice is tougher when stakeholders disagree with each other, or when management’s strongly felt values are in conflict with the mood of the society. On occasion there are good reasons not to be responsive to public outrage. I don’t believe the O.J. Simpson controversy is such an occasion.

Obviously, I wouldn’t have been canny enough to advise News Corp. to steer away from the deal. I would have advised the company that the deal would certainly make it look pretty cheesy, willing to profit from just about anything that’s not illegal. And I’d have advised it to acknowledge that this was so, and to have arrangements in place (or at least offers on the record) from the outset to share their profits with victims’ organizations and the specific victims’ families. But I wouldn’t have known to warn that the deal would probably cost News Corp. more money than it would make.

As the outrage built, I certainly would have advised Judith Regan not to disavow the financial motives for her involvement in the project. I could believe she saw the book as a sort-of confession; I could even swallow her rhetoric about “closure.” But I couldn’t buy that she wasn’t interested in the money too, or that she saw the project as a way of healing victims of domestic violence everywhere.

And I would have urged Rupert Murdoch attribute his decision to pull the plug on the project more to the public’s outrage and less to his own (belated) outrage. I suppose it’s possible Murdoch wasn’t paying attention to what Regan was up to until people started revealing how disgusted they were. But to me this smacks of the childish way companies typically react to being forced by others to do something they don’t want to do. Like a six-year-old forced to clean her room, companies imagine the consolation prize is that they get to pretend it’s voluntary: “I’m not cleaning my room because you made me. I’m cleaning my room because I want to!” When you’re bowing to pressure, it’s more honest, more gracious, and ultimately more profitable to say so.

What advice would I give now that the furor is pretty much over? The main thing, I think, is to keep mentioning it in the months ahead, maybe even the years ahead – however long it takes until the public (not the company’s management) is sick of hearing about it. In the aftermath of a bad mistake, people want to know that you know it was a bad mistake. They want to know that you feel chastened and have learned some lessons. That may mean announcing some changes in company policy or personnel. It may mean putting a retrospective on the incident in the company’s annual report. The next time the company decides not to move forward on some project because it could be offensive, it would be wise to say so, and to point out specifically that it doesn’t want to make again the mistake it made with O.J. Simpson.

And the next time the company decides to move forward on a project that some could find offensive, it should certainly mention Simpson before its critics do, explaining that it thought long and hard before deciding to go ahead anyway ... and why it sees this situation as different from the Simpson situation.

If risk is only the possibility of loss, then why would anyone ever take a risk? I think risk is more than the possibility of loss. Risk is the possibility of gain plus the possibility of loss. We take the risk because we believe the possible gain exceeds the possible loss.

If the possible gain exceeds the possible loss, the risk is positive and more likely to be taken. If the possible loss exceeds the possible gain, the risk is negative and less likely to be taken.

Positive Risk = the possibility of reward > the possibility of loss.

Negative Risk = the possibility of loss > the possibility of reward.

But I think for most people “risk” is about bad outcomes. In some definitions the risk is the bad outcome itself. In some definitions it’s the probability of the bad outcome (how likely it is), or the magnitude of the bad outcome (how bad it is). In quantitative risk assessment it’s the probability multiplied by the magnitude.

I don’t know of an occupation that routinely includes good outcomes in what it means by risk. People who invest money in a highly speculative stock, for example, don’t normally say it isn’t very risky because the potential gain is so high. They say it’s worth the risk because the potential gain is so high.

You’re certainly right that we judge the wisdom of taking a particular risk by balancing the probability and magnitude of bad outcomes against the probability and magnitude of good outcomes. But we don’t usually include the good outcomes in our definition of the risk itself.

That said, there is a profound sense in which you’re right and the conventional definitions are wrong.

People define risk in terms of bad outcomes and then try to balance risks (and costs) against benefits to reach a decision. That’s what we mean to do. But empirical studies of actual decision-making show that very often that’s not what we actually do. Instead, we systematically misperceive high-risk high-benefit situations. If we’re focused on the risk side, we underestimate the benefits. If we’re focused on the benefit side, we underestimate the risks.

For obvious reasons, advocates of specific outcomes do this in their rhetoric. Supporters of nuclear power tend to claim the risks are lower than they really are, while anti-nuke activists tend to claim the benefits are lower than they really are. But endless evidence says ordinary people accomplish a similar distortion in their private thoughts.

Here’s why. It’s hard to balance risks against benefits – not just intellectually hard, but emotionally hard. The act of balancing forces us to recognize that we can’t have it all, that we have to choose between giving up on benefits we covet and bearing risks we fear. We avoid the discomfort of a tough choice by misperceiving one of the two factors, telling ourselves either that the benefits aren’t really all that terrific or that the risks aren’t really all that horrific. We think we’re balancing risks against benefits. We keep them distinct in our definitions. And then we muddle them in our perceptions in order to make tough decisions feel easier.

This risk-versus-benefit confusion is one of three such confusions that play crucial roles in risk perception ... and therefore in risk management and risk communication. The other two:

• The magnitude-versus-probability confusion. It is very hard for people to perceive high-magnitude low-probability risks correctly. It’s uncomfortable to notice that a risk is really horrible (so take precautions) but really unlikely (so shrug it off). We tend to make the problem go away by either underestimating risk magnitude or overestimating risk probability. See my column on “Worst Case Scenarios.”
• The hazard-versus-outrage confusion. Decades ago, I proposed redefining the word “risk” so it includes not only the probability and magnitude of bad outcomes (which I want to call “hazard”) but also the factors like control, trust, dread, and familiarity that make a risk upsetting (which I call “outrage”). Whether you like my definitions or not, the evidence is clear that outrage has a huge impact on hazard perception. A risk that is really upsetting is likely to be seen as very dangerous, whether it is or not. But if a risk doesn’t have the characteristics that make it upsetting, it’s hard to notice that it’s dangerous.

Unlike the other two, this hazard-versus-outrage muddle isn’t symmetrical. Hazard doesn’t lead people to misperceive outrage. But outrage routinely leads people to misperceive hazard. Whether we think a risk is dangerous depends systematically on whether it’s upsetting – that is, on whether it’s controlled by others, managed in untrustworthy ways, dreaded, unfamiliar, etc. For more on my “Risk = Hazard + Outrage” concept, start with “Risk Communication: Facing Public Outrage,” and then look at some of the articles listed in my Outrage Management Index.

My effort to untangle the hazard-versus-outrage muddle led me to propose including what I call outrage in the definition of risk. So I can’t say you’re way off base wanting to include benefits in the definition of risk. You’re going against the tide, but so was I.

However we decide to define “risk,” what matters is that we stay alert to these three distinctions: (a) Hazard versus outrage; (b) Magnitude versus probability; and (c) Benefit versus risk (or, if you win your battle, “positive risk” versus “negative risk”).

From my background in medical ethics, law, and human rights, I have plunged into pandemic issues to bring the “preparedness” rubric to the public, to local communities. This is extremely important. However, I have discovered significant cultural, political, social barriers to this that I want to share with you in the hope that you might have some insight or wisdom to offer.

First of all, we live in a time when fear-mongering has been rampant and people’s capacity for additional threat perception is extremely – and justifiably – limited. There is what I call “Y2K syndrome” (a hugely over-hyped allegation of dire disruptions that never materialized) as well as “phantom menace syndrome” (the politically motivated scare tactics surrounding supposed disease threats from smallpox and anthrax – neither of which ever materialized and never truly existed anyway). Among those who do pay attention to the news media, therefore, there is justified skepticism about this new disease threat. Once the global context for this is introduced, they can generally be persuaded – but bringing people out to a town hall on pandemic influenza is a challenge in itself!

Given the general lack of education in this country with respect to how the U.S. connects with the rest of the world, there is the additional barrier of ignorance regarding globalization, recent worldwide changes in agricultural practices, the nature of viruses, the ecology of infectious diseases, etc. Even among clinicians, there is an astonishing lack of awareness of the context for disease outbreaks or the relationship between globalization and infectious diseases, or human rights and the spread of diseases throughout the world.

In addition to these barriers, we have ongoing cultural distractions ranging from sports to sitcoms and an Orwellian public discourse that has turned off most intelligent observers and utterly confused everyone else. Trying to bring complex fact patterns to an exhausted middle class is proving extremely difficult. (Forget those at or below the poverty level – the recommendation to “wash hands” and “stockpile food” is so grossly wide of the mark that it borders on the grotesque.)

Meanwhile, for whatever reason (turkey-consuming holidays? editorial fatigue?), our own national media outlets are noticeably “ho-hum” about avian influenza outbreaks ongoing in Asia, or the ongoing threat to humans by simple processes of evolution and ecology.

The concept of “planning” and preparedness, moreover, is flawed at its core. It is disconnected from local conditions, and offers generic “all hazard” mantras about “incident command” and “first responders” that bear no relation to the actual dimensions of a pandemic disease process. The public is simply left out of all of this. How are they supposed to develop the capacity to stay home for a month when their jobs are not secure? How will they obtain extra supplies of insulin for their diabetic child when insurance companies won’t pay the cost?

Why isn’t anyone in a leadership position addressing these issues? These are the risks that no one seems to be communicating – the risks of an exhausted public, a brainwashed and misled population that is both undereducated and overstimulated and can no longer distinguish the real from the fabricated, the authentic from the artificial, news from entertainment, real threats from politically motivated fear-mongering. Any suggestions?

• There is now a cohort of “pandemic fanatics” that didn’t exist a few years ago – people who are following the issues, preparing their households, and harassing their neighbors and employers. If a pandemic comes, they will be ready to lead.
• Public awareness in the mainstream has improved. In fact, it has improved so much that we have a new reason for frustration when there’s a lull (as there is right now) and it appears to be slipping.
• Civil society is taking notice. More and more companies, even small ones, are beginning the process of developing their own pandemic plans. So are many hospitals (at last!), schools, and soup kitchens. The ones that haven’t started yet are at least thinking maybe they should.
• Realism is taking hold. News stories (though scarcer) are showing more clarity about the distinction between bird flu in birds and pandemic flu in humans. Government pronouncements contain less knee-jerk over-reassurance. There is a better balance between medical preparedness and infrastructure preparedness.

Alerting a complex society to a risk it hasn’t previously focused on much is a slog. It was a slog for those who came before us too – those who pushed for wider acceptance of hardhats or seatbelts or vaccinations; those who are still pushing for wider acceptance of radon testing or carbon monoxide monitoring. Progress isn’t guaranteed; sometimes society shrugs off a risk and refuses to take precautions. What’s guaranteed is that progress, if it comes, comes slowly. If and when a severe pandemic materializes, of course, people will climb the learning curve very, very quickly. Getting people to climb the learning curve beforehand can save lives and reduce suffering. We can only hope there’s sufficient time, and keep slogging.

Several of the barriers you raise deserve discussion.

1. Has people’s ability to respond to the pandemic threat been undermined by their exposure to too many other threats?

It is certainly true that fear arousal is a competition. Jerry Falwell wants people to be more afraid of gay marriage; Greenpeace wants people to be more afraid of genetically modified foods; you and I want people to be more afraid of a pandemic. People have only so much fearfulness to allocate. And people have only so much time, energy, and money for precautions. So our success must come at the expense of Falwell, Greenpeace, and the advocates of awareness and precaution-taking with respect to thousands of other risks.

This understanding can greatly ameliorate officials’ widespread “fear of fear.” When we try to frighten people into pandemic preparedness, we are not turning them into more frightened people; we’re just competing for our slice of the fearfulness pie.

What isn’t true, I think, is that people are more threatened – or feel more threatened – than in the past. It’s pretty obvious that most people don’t face greater actual risks than their grandparents; if anything, we face fewer and less dangerous risks than they faced, freeing us to obsess over threats that didn’t make the cut for them. Nor do I believe that people are more bewildered or more overwhelmed than their grandparents were by the panoply of risks they face, and thus less able to take a new risk onboard. It has always been difficult to persuade people to take a new risk onboard before it was imminent and obvious. And those trying to do the persuading have probably always imagined that the task must have been easier in past generations.

2. Have people become so skeptical about all those phantom threats that it is harder and harder to warn them about a real threat?

I love your phrase “phantom menace syndrome.” And I think you’re probably right that people are more skeptical than they used to be, more aware that the current scare of the week may not turn out to be such a big deal after all.

But I see very little persuasive evidence that this skepticism translates into any kind of immunity. Sophisticated skepticism (even cynicism) seems to coexist with vulnerability. People know that you and I and Jerry Falwell and Greenpeace are all trying to arouse them to action against one or another risk. They know they shouldn’t take our word for it. They know that some of the risks they’re being warned against probably are phantoms, and some of the people doing the warning probably have hidden agendas. They may know all this more clearly than their grandparents knew it. But like their grandparents, they have a quantum of fearfulness to allocate. And like their grandparents, they keep on reallocating, ignoring some new candidates for their attention and taking others onboard.

It’s interesting to read in your comment the oscillation between these two concerns. Are people so overwhelmed by so many threats that they can’t think straight about how dangerous pandemics really are? Or are people so disenchanted with all those efforts to frighten them that they can no longer take pandemic warnings or any warnings seriously? Neither, I think – at least no more so than their grandparents.

3. Have “politically motivated scare tactics” and indiscriminate fear-mongering undermined the legitimacy of our efforts to frighten people?

I encounter this objection a lot when I urge clients to be more willing to frighten people into pandemic preparedness. Some point out that Hitler mobilized fear to justify the Holocaust. Others point out that President Bush mobilized fear to justify the invasion of Iraq. (I am not suggesting that these are equivalent sins.) How can it be right for us to do the same thing to justify pandemic precautions?

In the final analysis, ethics are personal, and I would never urge a client to do something that felt unethical to that client. But for myself, at least, I see fear-arousal as an ethically neutral tool – good when deployed on behalf of good goals and evil when deployed on behalf of evil goals. (I’d add that it may be evil even when deployed on behalf of good goals if it’s deployed hypocritically; whatever you’re trying to accomplish, you should think twice before frightening people about something you don’t actually consider scary.)

The term “fear-mongering,” of course, begs the issue. We apply the term only when we believe that a particular effort to frighten people is illegitimate. We all agree it applies to Hitler’s rants against Jews and others. Opponents of the war in Iraq would apply it to President Bush’s warnings that Saddam Hussein might have Weapons of Mass Destruction. It sounds like you’d probably be comfortable applying it to Y2K, smallpox, and anthrax warnings. I’m inclined to apply it to Falwell’s and Greenpeace’s warnings. Industry applies it to activist anti-pollution campaigns. Anti-pollution activists apply it to industry efforts to warn about job loss and economic stagnation. Lots of people apply it to our efforts on behalf of pandemic preparedness.

The most effective way we know to motivate preparedness and precautions is to frighten people. Excessive fear is immobilizing, of course, but so is insufficient fear – and insufficient fear is by far the more common of the two problems. If we think pandemics are worth preparing for, we ought to want to arouse some proportionate fear about pandemics. I see no reason to be deterred by the fact that other communicators are working to arouse fear about other issues in ways we consider disproportionate.

4. Is ignorance a major barrier to pandemic precaution advocacy?

Yes and no. There are certainly a lot of things we need people to know about pandemics and pandemic preparedness.

But the relationship among information, attitudes, and behavior isn’t what we usually imagine. Educators are particularly prone to assume that the relationship looks like this:

^{InformationAttitude changeBehavior change}

But Leon Festinger’s “cognitive dissonance theory” established some 50 years ago that behavior change usually comes first, often motivated by fear or other emotions. Then people seek out information to justify the new behavior and thus reduce the “cognitive dissonance.” Finally, they evolve attitudes that integrate the new information, stabilizing the new behavior. Thus:

^{Emotional arousalBehavior changeCognitive dissonance
Information-seekingInformationAttitude ChangeBehavior stability}

(You may want to look at a book chapter by Brian A. Day and Martha C. Monroe that describes my approach to these issues in greater detail.)

5. Is the concept of planning “flawed at its core”?

I agree with you fervently that planning that doesn’t involve the public is flawed at its core. People don’t necessarily have a perfect understanding of how they are likely to respond to a particular threat or a particular recommended precaution. But their understanding of their own likely responses is hugely better than the understanding of professional planners. So unless we consult with them as we plan, our plans are pretty close to useless. (Collaborative planning has other advantages as well – among them that people are likelier to trust and comply with a plan they helped develop.) As you point out, the discontinuities between planners’ expectations and people’s actual responses are greatest when the planners are planning for people most unlike themselves – for poor people and ethnic minorities, for example.

For a landmark study demonstrating these realities for two case studies, a smallpox attack and a dirty bomb attack, see the “Redefining Readiness” report by Roz Lasker and others at the New York Academy of Medicine.

I’m not sure whether I agree with you that the “all hazards” approach is flawed at its core. I do see lots of differences between pandemics and other sorts of emergencies – but I also see the inefficiency, maybe even the futility, of preparing independently for an endless list of possible scenarios. Ideally, we would pursue all-hazards preparedness as far as it can take us, and then would focus real attention on what’s unique about the hazards that worry us most.

Certainly one of the most important things about a pandemic is that people, families, and neighborhoods would be forced to rely largely on their own resources, rather than waiting to be helped by governments. That’s not unique to pandemics, but it certainly distinguishes pandemics from many other kinds of emergencies. It makes citizen involvement in pandemic planning all the more important. And it makes the “incident command” concept all the more peripheral.

It is, I think, a healthy response to the current political climate to be wary of assertions that “we” (the public) should be ready to trade off our civil liberties in the face of vague and unspecified threats to our “security.” The challenge is to provide accurate and up-to-date factual information that differentiates the pandemic from other kinds of threats. I agree with you that realistic fear is an appropriate and necessary motivator.

Finally, about the “incident command structure” – it seems unlikely that this decision-making and coordinating rubric will have much to offer after a couple of weeks of a pandemic, much less a couple of months or years. My concern is that the overall responsibility, having been handed to homeland security agencies rather than health-related agencies, may too easily become militarized.

In all of this, the core focus needs to be public awareness and engagement at some level of the preparedness process. Without knowing the extraordinary dimensions of a pandemic, both in time and in geographic reach, people will not grasp what they need to do that is different from preparing for floods or fires. In this sense, the “all hazards” approach is inadequate, even for first responders.

I have been worried that when health departments are in charge of pandemic preparedness, they tend to focus on a fairly mild pandemic paradigm – because in a mild pandemic (like 1957 and 1968) the main issues really are medical issues: hospital surge capacity, ventilator sufficiency, antiviral supply, etc. In a severe pandemic (like 1918), on the other hand, the main issues are likely to be infrastructure survival: energy supply, food supply, water supply, and the like. These non-medical concerns tend to get more attention when pandemic planning is in the hands of emergency response agencies.

We’re best off, obviously, when both sorts of departments have a hand in pandemic planning – along with human rights experts, risk communication experts, a wide range of other sorts of experts, and, above all, normal people whose chief expertise is the sort of ordinary self-knowledge we experts tend to lack.

It seems to me that there are similarities between the two, and based on your Risk=Hazard+Outrage equation, that the outrage components would influence risk perception.

I would appreciate it if you could clarify this for me.

I try to steer away from the perception formulation because it leads so easily to the view that people are making a mistake when they take voluntariness, control, trust, and the rest seriously. “Perception” gets interpreted as “misperception” – and the lesson seems to be that clear thinkers should avoid contaminating their judgments about risk with all this extraneous psychological chaff.

In my formulation, “outrage” and “hazard” are both part of what risk means. A situation is thus riskier if it’s less under your control (for example), every bit as much as it is riskier if it’s likelier to kill you. If I make the situation less deadly, I’m making your risk smaller. If I give you some control, I’m also making your risk smaller.

So where does perception fit in? Outrage has a huge impact on hazard perception – a bigger impact, in fact, than hazard has. If I give you more control over the situation, you are likely to perceive the hazard as smaller, even though the hazard hasn’t actually changed.

But even though outrage greatly influences hazard perception, it’s a mistake to think they’re the same thing. How much control you have is how much control you have. How likely you are to get killed is how likely you are to get killed. (Both have to be perceived, by the way; perception is the only means humans have for inputting data.) So yes, how much control you think you have will affect (that is, distort) how endangered you think you are. But independent of its effect on your hazard perception, how much control you think you have will directly affect how serious you consider the risk. And that part isn’t a distortion or a misperception at all. It’s part of what people mean when they talk about a risk being big or small, serious or trivial.

Think about a controversy over a factory’s emissions of dimethylmeatloaf. Narrow the controversy to two issues: the dose-response curve of dimethylmeatloaf for pancreatic cancer (a traditional hazard issue) and the frequency with which the plant manager has misled the neighborhood about how much dimethylmeatloaf is emitted (a traditional outrage issue). Notice three truths:

1. We “know” the dose-response curve and the integrity of the plant manager only through perception. We may perceive these two variables accurately or inaccurately. We probably have better data on the integrity of the plant manager.

2. If the plant manager has been consistently candid, we are likelier to see the dose-response curve as fairly benign. If the plant manager has been dishonest, on the other hand, we are likelier to suppose that the dimethylmeatloaf must be really dangerous.

3. Even though the plant manager’s integrity affects our perception of the dose-response curve, they’re not the same thing. If we discover that dimethylmeatloaf cannot cause pancreatic cancer – thus correcting a misperception – we will nonetheless continue to consider it a serious risk to live near a factory with a dishonest management.

Equally useful, or perhaps even more useful, would be a “Pandemic Flu Good Communication Example File.” In my experience teaching risk communication, people seem to perk up when I tell them “good example stories” – it gives them hope that it is possible to learn strategies for doing a better job.

The first training exercise that seems to give my workshop “students” hope is when they practice converting over-reassuring statements into reassuring but cautionary statements. Here is a typical “before” and “after” for such a makeover:

I'll try to prime the pump on this when I get back from my current posting in the Philippines, but maybe someone on the Flu Wiki will get it started sooner than that!

I fervently agree that good examples are at least as powerful a learning tool as bad examples. So a repository of good pandemic news stories (good reporting and good sourcing) would be really useful – especially if it were annotated to explain what makes it good. For a comparable annotation of a wonderful pandemic speech given in May 2005 by Australian Health Minister Tony Abbott, see “Superb Flu Pandemic Risk Communication: A Role Model from Australia.”

I am a student of political sciences and social psychology in Frankfurt/Main Germany. Right now I am doing an internship in a German company in Beijing, China for three months. In Frankfurt I worked with Edelman as a freelancer for the last year during my studies, focusing on crisis communications.

Have you written any paper on the relationship between CSR [corporate social responsibility] and crisis communications, risk communications, and risk management? Or could you tell me any author that has written on this relationship?

I would like to write my diploma on this topic, focused on the Chinese market. (Actually there is a relevant crisis in China right now, the SK2 scandal, involving some shampoo that contained toxics.)

I think the relationship exists and a good CSR management could keep customers from ruining a company’s reputation. Do you see this as I do?

I have written on what I see as one of the key differences: Whereas CSR is preoccupied with “responsibility,” risk communication is much more interested in “responsiveness.” See “Responsible or Responsive?” for a dialogue I had with John Elkington of SustainAbility on this distinction. I feel even more strongly about this distinction today than I did when this exchange took place. I don’t much want corporate CEOs deciding what’s good for the world. I want the world deciding, in an ideological free-for-all in which CEOs play no greater role than the rest of us. And then I want CEOs knuckling under to what the world has decided, because that’s the path of profitability and survival for their companies.

I also comment routinely, though not usually in writing, on the distinction between trying to do good/look good on the one hand and trying to avoid doing evil/looking evil on the other hand. I am convinced that the latter plays a much more important role than the former in corporate reputation and profitability; the former is, of course, at least as central to CSR people. There is a transcript on the Web of a presentation I gave at a 2004 corporate responsibility conference run by the Boston College Center for Corporate Citizenship. The very beginning of the transcript addresses this distinction between the ways of improving a company’s reputation: trying to be loved more (which is central to CSR) versus trying to be hated less (which is at the core of the outrage management side of risk communication).

Neither of these distinctions, of course, means that risk communication and CSR are antithetical. They are allied. But they are far from identical, I think!

I didn’t know if you had seen this thread yet. It has some interesting examples: http://www.fluwikie2.com/pmwiki.php?n=Forum.PandemicFluMisinformationHallOfShame

Jody (my wife/partner Jody Lanard) and I have our own collection, to which we keep adding. The errors fall into distinct categories:

• Confusion of bird flu with pandemic flu (as if a pandemic were going to reach “here” via sick birds rather than sick people)
• Over-optimism about local preparedness (as if mass inoculation drills were going to help when there’s nothing yet to put into the syringes)
• Excessive focus on medical issues (as if overstretched hospitals were going to be our biggest problem and we didn’t need to worry about a pandemic’s impact on supply lines, infrastructure, and where our food and energy would come from)
• Insufficient mobilization of public involvement and public preparedness (as if governments could do the job themselves, and as if people would necessarily comply with government pandemic plans they hadn’t helped write)
• Confusion of current risk with potential risk (as if the fact that H5N1 has infected very few people so far meant that it’s silly to start worrying until after the virus learns efficient human-to-human transmission)

And so forth. Maybe someone will use the Flu Wiki list to develop a typology of kinds of pandemic misinformation.

I don’t blame journalists for pandemic misinformation in the media as much as I blame their sources. There are only a handful of “pandemic flu reporters” in the world — science or medical writers who have become really expert on this particular story. Most science and medical writers know surprisingly little about the pandemic risk, and of course general assignment reporters know far less.

It is useful to teach journalists more about pandemics. (I will be participating in a December conference at Harvard University that will try.) But teaching pandemic experts more about how to talk to journalists will probably help more. And more crucial than either is teaching local health officers and public health professionals — the principal local news sources and the principal local pandemic planners — more about the realities of pandemic preparedness.

When I see a piece of misinformation in the media, I often can’t tell if the source got it wrong, or if the source explained it badly so the reporter got it wrong. Either way, I believe better sourcing is a higher priority than better reporting. But I’ll settle for either, and work for both.

I have read with interest a number of your case studies, but have a specific interest in your perspective on how to deal with communities when a new/controversial industry may be coming to their community.

There is a tactic to tell the locals as little as possible, but I was always taught that properly informing the public – answering their questions as opposed to being evasive – leads to much more trust. It’s just human nature.

Thank you in advance for any comments and/or previous writings you can direct me to that deal with heavy industry introduction to new communities.

There’s no question that telling the locals as little as possible makes the siting process move faster in its early stages. But the price is high: When the community finally learns what you’re up to, people’s suspicions about the wisdom of tolerating the facility are validated by your failure to deal straight with them. So after an efficient, secretive start the siting process typically grinds to a halt.

There are two reasons why it’s smarter to have your public debate early rather than late. First, the debate can focus on the pros and cons of the facility, rather than the disingenuous process you used to get it sited – so it’s a fairer, less hostile debate. And second, you get to find out what you’re up against sooner; it’s not really in your interests to proceed blithely for years before you learn that you may not be able to get your facility sited after all.

The counterargument is that there’s not much opponents can do about a fait accompli. So supporters of secret siting say that the further along you are before the locals know enough to try to stop you, the likelier you are to be able to succeed over their objections. This may have made some sense in years past, and may still make some sense in the developing world (unfortunately). But public involvement is now a settled feature of LULU siting in western countries. (“LULU” is a wonderful acronym invented by Frank Popper: “locally unwanted land use.”) You probably can’t get all the way through the siting process without facing the music – so getting three-quarters of the way through it isn’t very smart. And even if you can sustain your secrecy strategy long enough that it’s no longer practical for opponents to make you change course, do you really want to operate a facility whose neighbors will remember and resent for decades that you snuck in when their backs were turned?

For more on this timing issue – though not in the siting context – see “When to Release Risk Information: Early – But Expect Criticism Anyway.”

Three other pointers on risk communication aspects of siting an industrial facility:

In addition to letting the cat out of the bag early in the siting process, do so with appropriate emphasis on the downsides of the facility you are proposing to build.

Presumably you have a case to make that community benefits outweigh community costs and risks. Make your case. But don’t just talk about the benefits. Concede the costs and risks. Better yet, announce them before opponents have a chance to do so. Acknowledge that the burden of proof is on you to show that the community stands to gain more than it loses. Then meet that burden of proof. And work hard to do something for the people – there are usually at least a few – for whom the benefits genuinely don’t outweigh the costs and risks.

In talking about these risk-benefit tradeoffs, moreover, pay more attention to what you’re doing to mitigate the risks than to how great the benefits are. Even though people do tolerate more risk when they have more to gain, they don’t like thinking in terms of tradeoffs. When benefits get higher, we are inclined to see risks as lower – not just more acceptable, but actually lower. So make sure people know about the benefits, and then talk mostly about what you can do to keep the risks low.

Go beyond candor to responsiveness and accountability.

It’s certainly a good start to design a proposal whose community benefits genuinely outweigh its community costs and risks, and then to communicate promptly and straightforwardly about your proposal. But that’s no longer always enough. In addition, your new neighbors are likely to want some say in the design of the facility. They’d rather you came to them with a rough draft and a lot of options to discuss; not unlike your senior management, they want to see their own handiwork in the final product. Benefits and precautions that they have demanded and secured from you are a lot more reassuring than benefits and precautions you offered them from the get-go.

And your new neighbors are likely to want some oversight as well. They’ll want to be able to check for themselves that the facility’s benefits, risks, and costs are what you say they are. I’m working with a client right now that offers to negotiate a formal contract with each community in which the company wants to put a factory. The contract outlines enforceable commitments – including what happens if something goes wrong: how the community will be able to tell and what recourse it will have.

Make sure you’re asking, not telling. And if possible, consider a bunch of communities at once.

The facility siting process that is least likely to succeed is the one governments often use for really controversial facilities. They spend years delineating complex criteria for finding the “best” site for their facility. Then they crank through a site selection process and come up with a single target site. Then, having clearly established in everyone’s mind that the facility is so incredibly dangerous that only the very safest site on the continent will do, they shake their heads in wonderment when the targeted community isn’t wildly enthusiastic about having won this particular lottery.

Ideally, facility siting should be voluntary on both sides. The community should feel like it has a right to say no, and should initially be asked to say maybe, not yes. And the community should know that your company is looking at other options too. “We have identified several possible sites that could work well for us. We’re exploring each of them, looking for a community that thinks hosting our facility will work well for our neighbors too. Let’s sit down together and figure out if we’re a good match for each other.”

For more on risk communication aspects of facility siting, see:

• “Siting Controversial Facilities: Some Principles, Paradoxes, and Heresies”
• “Getting to Maybe: Some Communications Aspects of Siting Hazardous Waste Facilities”
• “Not in Our Back Yard”

The article’s title says it all: “Disaster Preparedness: Ending the Exceptionalism.” Here’s an excerpt:

Even with the inaccuracies in public health spending figures, these numbers suggest that between 6% and 8% of all annual public health expenditures go toward public health emergency-preparedness efforts. The ... numbers also raise the question as to whether this level of expenditure is necessary. If we follow McGinnis and Foege’s exhortation to focus on “those factors that represent the root determinants of death and disability,” we are spending too much on public health emergency preparedness – that is, an amount not proportionate to the impact of public health emergencies on health.

What are your thoughts on this article?

I have worked with a lot of health departments that fervently agree. Many times in recent years I have been brought to a state or local health department to give a seminar on crisis communication. Almost always, federal money paid for the seminar – usually terrorism or bioterrorism money, sometimes pandemic preparedness money. Almost always, most of my audience would have rather spent the money (and their time) on less hypothetical, more imminent health threats.

It’s worth noting that a national or global health catastrophe like the 1918 flu pandemic could alter the statistics significantly. Horrific though they were, 9/11 and Katrina were health disasters only locally. National preparedness for a possible local catastrophe is likely to fail Dr. Lipsman’s cost-effectiveness test. Preparedness for a possible national or international catastrophe has a sounder economic rationale. Nobody knows the probability of another severe influenza pandemic like 1918’s, but given its very high magnitude, it doesn’t have to be all that probable to be worth getting ready for.

Still, Dr. Lipsman is almost certainly statistically in the right. On a dollar-for-dollar basis, disaster preparedness saves fewer lives than going after the main U.S. sources of preventable death would save.

Of course if economic efficiency is the criterion, going after the main U.S. sources of preventable death isn’t much of a bargain either. The least expensive way to save lives is by fighting hunger, poverty, and chronic infectious diseases in the developing world.

The real flaw in Dr. Lipsman’s argument is that nobody (probably including Dr. Lipsman himself) really wants to focus on the least expensive way to save lives.

What are the risks Dr. Lipsman wants to address with money now squandered on preparing for pandemics, natural disasters, and terrorist attacks? He lists them for us. Roughly 800,000 Americans, he says, die annually from smoking, eating too much, and exercising too little. Alcohol kills another 85,000. The remaining biggies: endemic microbial diseases like flu, pneumonia, and tuberculosis; pollution; motor vehicles; firearms; sexual behavior; and illicit drug use.

It’s interesting that Dr. Lipsman puts pollution on his list, because my corporate clients frequently make the same argument about pollution that he is making about disaster preparedness. They complain that factory emissions (for example) are orders of magnitude less deadly than smoking a cigarette and not wearing your seatbelt on your way to a meeting to talk about factory emissions. They’re statistically in the right too.

What they (and Dr. Lipsman) are missing is that living next door to a polluting factory and being attacked by terrorists are both higher-outrage risks than smoking, overeating, and the rest. People don’t prioritize risks according to how cost-effective they are to avert. We prioritize risks according to how upset they make us feel.

This isn’t just a fundamental risk perception principle – though it is that. It is also a fundamental human value. The risk perception part is that upsetting risks are likely to be perceived as more dangerous than they really are, while risks that aren’t very upsetting are perceived as less dangerous than they really are. Understanding that is essential to risk communication. But equally essential is understanding that even when the misperception is corrected, people still hang onto their preference for spending more to mitigate high-outrage risks than low-outrage risks. That’s the part that’s about human values.

I sometimes tell my clients this hypothetical story:

Imagine that every ten years or so a sniper climbs up onto an overpass with a high-powered rifle and shoots and kills a passing motorist. Then he’s good for another decade. Finally, after 30 years and three deaths, he is caught and brought to trial. Here’s his defense: “During the 30 years during which I shot and killed three passing motorists, thousands of people died on our nation’s highways as a result of drunk driving, not wearing their seatbelts, poor highway design, and poor automotive design. Sniping is an infinitesimal part of the highway death toll. In picking on me, a mere sniper, the government is distorting the public’s understanding of the real priorities of highway safety. The money the government is spending catching me, trying me, and imprisoning me could save far more lives if the government made the rational risk management decision to let me continue killing a mere one person per decade, and reallocated the money to repainting the lane markers on highways.”

Like Dr. Lipsman, my hypothetical sniper is right on the data. Nonetheless, no jury would vote to acquit. Even after they study the data, normal people support spending more money per life saved catching and punishing snipers than repainting lane markers.

Two outrage components are especially relevant to disaster preparedness versus smoking, obesity, exercise, and the like: chronic versus catastrophic risk, and voluntary versus coerced risk. A number of other components – moral relevance, dread, familiarity, etc. – also come into play, but let me focus on the two main ones..

First, and most directly relevant, is the distinction between chronic and catastrophic risk. Consider these examples:

• Driving is more dangerous than flying in terms of deaths per passenger mile. But airline crashes are occasional catastrophes, which makes them a source of considerable anxiety (and media interest). Car crashes are routine, and therefore much easier to shrug off.
• Nuclear power is statistically safer than fossil fuel power. Like a severe pandemic, a major nuclear disaster could alter the statistical picture. But even without a major disaster, the hypothetical risk of a catastrophic nuclear accident generates a lot more outrage than the actual but much more diffuse deaths attributable to burning oil and coal.
• Smoking kills roughly 400,000 American a year. Imagine that they all had to die on November 13 in Chicago. On November 14, we would outlaw smoking. But they die spread out over the year, and spread out over the country, in the privacy of their pain. That’s much more acceptable – much lower-outrage.

As with my sniping example, the greater outrage that society attaches to catastrophic as opposed to chronic risk isn’t just a perceptual distortion. When individuals die, the social fabric is relatively undamaged; we bury the dead, we comfort the bereaved, and life goes on. Recovery is much harder when disaster strikes a community – or a country. Disasters rip the social fabric in ways that individual deaths simply don’t. Our willingness to “overspend” on disaster preparedness embeds not just misperception of the statistical risk, but also wisdom about how societies work.

In fact, you can make a pretty good case that we spend too little on disaster preparedness. There are two normal reactions to a high-magnitude low-probability risk – that is, a possible catastrophe. Either we imagine its probability to be higher than it really is, and therefore over-prepare (at least as judged by the cost-effectiveness criterion). Or we go into denial, imagine the probability to be zero, and therefore feel fine about not preparing at all. In response to 9/11 and Katrina, we may be doing a bit more of the former than usual. But we are still doing plenty of the latter. Look at the statistics on how many local hospitals (not to mention power plants, water treatment facilities, and families) have done next-to-nothing to get ready for a possible pandemic. Look at the unpopularity of flood insurance, even though it is subsidized by the federal government. Ask yourself why the levees in New Orleans were inadequate, why so many people didn’t evacuate, why so few brought food with them to the Superdome.

Preparedness is always something of a tough sell, because it’s so emotionally unsatisfying. Preparedness assumes a bad thing may well happen and works to reduce its consequences. By contrast, prevention assumes there are ways to keep it from happening. Both in individual psychology and in political sociology, prevention is much more attractive. We will pay more than it’s worth to take a risk off the table entirely, and less than it’s worth to reduce the damage. The late Aaron Wildavsky argued persuasively that too much of the national safety budget went to prevention and too little to preparedness. A society that focuses too much on preventing disasters, he also argued, will lack resiliency when forced to cope with one it couldn’t prevent. (For more on these issues, see my column on “Worst Case Scenarios.”)

The other outrage component that Dr. Lipsman's article brings to mind is the voluntary-versus-coerced distinction. Most of the risks on Dr. Lipsman’s list are voluntary risks. People choose to smoke, overeat, and live sedentary lives. And the vast majority of Americans who smoke, overeat, or live sedentary lives do so in spite of knowing it’s bad for them.

Leave aside the question of how effective medical intervention – or any intervention – can be in changing the behavior of people who already know better but have so far been unable or unwilling to change. Assume that a dollar spent on smoking cessation clinics will do more good than a dollar spent on bioterrorism early warning systems. This important question remains: Isn’t it more important to protect people from evildoers than from themselves?

Terrorists, like snipers, strike fear into our hearts in ways that our own bad habits cannot match. Once again, this isn’t just a misperception of the statistical risk. Most people know they are unlikely to die in a terrorist attack. Their main worry is that others will die, and they will have to watch it on television, explain it to their children, and go on living in the more miserable world a terrorist attack creates. Overspending on terrorism preparedness isn’t irrational. Neither is a judgment that we don’t want to do everything possible to deter people from their dangerous bad habits. Maybe this is selfishness, a reluctance to expend tax money on other people’s foolishness. Maybe it is tolerance, respect for the life choices of others, and a disinclination to become more of a nanny state than we already are.

Of course natural disasters and infectious disease pandemics aren’t intentional acts of evil. But they aren’t voluntary either. They’re in the middle. And so they generate an intermediate level of outrage, and thus of willingness to prepare.

The bottom line for me: Dr. Lipsman is probably right that we are spending more per life saved on preparing for public health disasters than on persuading people to quit smoking, eat less, and exercise more. But there are reasons why we make this choice. Those reasons are debatable, and worth debating. Cost-effectiveness statistics should be part of the debate, but the statistics shouldn’t be allowed to preempt the debate. Fortunately, given how fundamental outrage is to our psychological makeup and our societal values, there is no real risk of that happening.

Like you, I am a strong believer in having all points of view, even the strongest opponents, on the CAP. However, I’m getting a lot of pushback about having the PETA-backed opposition group involved. Is there ever a time where an opponent is so radical there is no benefit to involving them?

It’s not that putting a PETA-backed activist group on your CAP is necessarily a benefit. Let’s assume that having a media-savvy extremist animal rights organization anywhere in the vicinity isn’t your client’s cup of tea. But it’s not up to your client whether the PETA affiliate is in the vicinity; that’s up to the PETA affiliate. All your client gets to decide is whether it would rather have the group attending its CAP meetings or protesting its CAP meetings. Given that limited choice, Lyndon Johnson’s advice is on target.

Of course a PETA affiliate may very well decline any invitation to join an advisory panel for a lab that tests drugs on animals. In fact, I’d say it will probably decline. But it’s hard to tell for sure. Almost invariably, an activist group has two sets of constituencies. Its radical stakeholders – key volunteers, for example – are likely to take offense if the group “compromises itself” by giving courteous advice to the enemy. They want it to stay pure and embattled. But moderate stakeholders – including large numbers of small-money donors – may want to see it settle for half a loaf, accomplishing something concrete even at the expense of purity.

PETA is certainly aware of this dynamic. It knows its leaders and foot soldiers are predominantly animal rights extremists while its financial contributors are mostly animal welfare moderates. It deliberately decides when to zig and when to zag. But it doesn’t aspire to become the World Wildlife Fund or the Humane Society. Opportunities to cement its radical image are usually more tempting than opportunities to soften that image.

Whether or not the local group accepts your invitation to join the CAP, issuing the invitation is in your client’s interests. It’s not in the group’s interests. If the group’s leaders can con the lab into excluding them, they get a free ride: They don’t have to choose between offending their left wing by accepting and offending their right wing by declining. They can stay comfortably outside demanding to be let in. They can make up pretty much any story they want about what probably went on at meetings they weren’t allowed to attend. They can ask archly what the lab is afraid they might find out.

Although your client is better off issuing an invitation that’s declined than not issuing one at all, it is even better off if its invitation is accepted. Among the many reasons why this is so: It’s much harder for activists to invent dastardly deeds to accuse lab management of perpetrating if they have undisputed access to the real evidence. (Of course if the real evidence is pretty dastardly, your client’s options are much less attractive. I’m assuming the lab is already taking animal welfare seriously.)

Another good reason to hope the group says yes is the opportunity to build a record of accountability. Even a well-run animal testing laboratory needs to ratchet up its animal welfare policies periodically. Paradoxically, improvements that opponents have demanded yield more outrage reduction than improvements that the lab has instituted unilaterally. So it’s handy to have a PETA surrogate in the room pushing the envelope.

And a third reason for wanting the group on the CAP: It will do wonders for the demeanor and demands of more moderate groups. If the PETA allies aren’t present, the moderates who are present will feel some pressure to stand in for them, to show you and show the world that they’re no patsies. But if the PETA allies are there and vocal, the moderates may well feel a different sort of pressure: Pressure to show that “we’re not like them.” In one pretty common CAP dynamic, the extremists kill a lot of the group’s time with repetitive off-the-wall accusations. Other CAP members are rolling their eyes. The company, however, continues to take the group’s claims very, very seriously. If your CAP follows this scenario, the lab could end up allied with the PETA affiliate on behalf of the importance of not muzzling anybody. The extremism of the PETA people and the responsiveness of the lab people combine to make everyone else on the CAP that much more fervently moderate.

Of course there are downsides. For one thing, the CAP will accomplish less with extremists on board. I’m guessing that concrete accomplishments are less important to your client than providing a venue where anti-lab outrage can be safely vented and accommodated – but it’s still important to make sure there are some real accomplishments. Another disadvantage: The CAP will be potentially more fragile. Moderates may get frustrated at the lack of concrete progress and quit ... or just stop coming. Or the extremists themselves may realize that their presence is helping the lab’s cause more than their own, and look for a pretext to quit in protest.

Still, on balance the lab has more to gain than to lose by getting the PETA allies there and keeping them there. So it should look for ways to make its invitation easier for a PETA-backed group to accept, and harder or costlier to decline.

For example, I remember working with a cement company that was seeking permits to burn certain types of waste in its cement kiln, essentially turning the kiln into an incinerator. After painful reflection, it decided to invite a local Greenpeace affiliate to join its CAP. The group initially declined, on the grounds that it opposed all incinerators and wasn’t about to give up its opposition. My client said something like this:

Of course not! We’re not asking you to stop opposing incinerators. We assume you’ll oppose ours, and we certainly aren’t making support for our incinerator a condition for sitting on our CAP. But it’s possible the incinerator will be approved despite your opposition. If that happens, you would certainly want to make sure it’s as safe as an incinerator can be. So we’re asking you to pursue two parallel tracks. Work to stop the incinerator – and just in case you fail, work to push us to make sure we design, build, and operate the safest incinerator possible. Given how deeply concerned you are about incinerator safety, how could you do any less?

The company’s invitation to the Greenpeace affiliate led to a succession of private negotiations and public appeals. Eventually, the group decided it couldn’t afford to look insincere about incinerator safety, and it reluctantly joined the CAP. The company never got the permit. But at least the Greenpeace affiliate was inside the tent pissing out.

We’re a public health agency working on bird flu news release templates. We know “bird” experts use the high-path/low-path terminology. But do those terms necessarily mean something to the everyday person?

We don’t necessarily see low-path H5N1 as new news, but it does need to be addressed as birds continue to test positive in the U.S. When bird flu is affecting birds, distinguishing between the two is useful. But what we’re truly concerned about is the bad one. If bird flu affects people (not talking pandemic), the high-path/low-path description becomes a moot point because we’re only talking high-path.

So what about using “deadly bird flu” to describe high-path? Would that make things more meaningful to the public without having to go into all the low-path versus high-path explanations? Then what would you call low-path? Or would this confuse people even more?

1. There are lots of different strains of bird flu, just as there are lots of different strains of human flu.
2. Categorized in terms of ability to inflict harm on poultry, bird flu comes in two “strengths.” “High-pathogenic avian influenza” (often abbreviated as HPAI) can be deadly to poultry. “Low-pathogenic avian influenza” (LPAI) causes very mild symptoms in poultry, or sometimes no observable symptoms at all.
3. Some bird flu strains – not all – have been found in both HPAI and LPAI versions. This is true of the H5N1 strain that is currently getting so much attention.
4. The bird flu strain that is currently devastating some bird populations, that has managed to spread to a few hundred people, and that many believe threatens a possible human pandemic is HPAI H5N1.
5. LPAI H5N1 occurs from time to time in North America. (In fact, it's often called “North American H5N1,” as opposed to the “Asian” HPAI H5N1.) In its current form, LPAI H5N1 is no threat at all to human health – at least to the best of scientists’ knowledge thus far.
6. HPAI strains differ in their ability to cause serious disease when they spread from a bird to a human. HPAI H5N1 is deadly to humans (though so far very hard for humans to catch). But a recent high-path H7N3 strain, which devastated domestic chicken flocks in Canada two years ago, caused only a mild eye infection when it spread to some poultry workers.
7. Laboratory tests for bird flu give you some information fairly quickly, while other information takes longer. You usually get the H before the N. You usually get both before the definitive high-path or low-path determination (though you may have a strong high-path/low-path hint from clinical observation – birds are dying or they aren't).
8. Some strains of bird flu can be low-path in most wild bird species and high-path in domestic poultry. Some strains can be low-path but then mutate (or not) to high-path after they spread to domestic poultry. H5N1 has both of these traits. So if H5N1 is found in wild birds that are perfectly healthy, it is still possible that it’s HPAI as far as poultry are concerned. And it’s also possible that it’s LPAI but will become HPAI later.

All that is what you ideally want people to understand – though it’s a lot for laypeople (and even for many officials) to absorb. You want people to recognize that not all cases of bird flu are H5N1, and that not all cases of H5N1 are high-path. You want people to recognize that vanishingly few cases of high-path H5N1 in birds have so far spread to a human victim, and that not one of those rare human cases has so far launched a pandemic. And you want people to recognize that if a pandemic ever does start, the human risk will no longer be from birds, but from each other.

At a minimum, you want to avoid giving people misunderstandings. Any story is a misleading story if it creates the impression that there is already a bird flu pandemic somewhere else that fortunately hasn’t reached Our Town yet. Any story is a misleading story if it creates the impression that we are safe from a pandemic as long as our birds aren’t sick. Any story is a misleading story if it creates the impression that every case of bird flu represents an imminent threat to human health. And any story is a misleading story if it creates the impression that so long as a wild bird is healthy it doesn’t matter what bird flu strain it has.

The only way to make this all clear is to say it, and keep saying it. And acknowledge that it is confusing. And admit that it took you a while to understand it too. And apologize for sometimes explaining it in ways that add to the confusion.

• When you find a case of low-path bird flu, you need to announce that that’s what you found – and explain that this is a potentially significant economic problem if it gets into poultry flocks, but not a problem at all for human health (as far as we know).
• When you find a case of high-path bird flu, you need to announce that that’s what you found – and explain that this is a very big economic deal for poultry farmers. Depending on what high-path strain it is, it may also be a potential health risk to poultry workers, and a possible longer-range threat to public health because of pandemic potential.
• When you find certain kinds of low-path bird flu – H5 or H7 – you need to explain that these types could have the potential to become high-path over time. That’s why farmers are culling many chickens that aren’t terribly sick, trying to eliminate the virus before it has a chance to develop into a devastating high-path strain.
• And when you find a healthy wild bird that has a bird flu strain that’s low-path in wild birds but high-path in poultry, you need to take the time to explain why it’s not a big deal (yet) nor are you entirely out of the woods (yet).

I vote for using the terms “high-path” and “low-path” – and explaining that “path” is short for “pathogenic,” which means “disease-causing.” (I’d also point out that these terms are used only to describe the effects of various flu strains on poultry; they’re not used for people.) The abbreviations “HPAI” and “LPAI” probably come across as off-puttingly technical. And phrases like “deadly bird flu” are likely to cause more problems than they solve. A high-path bird flu that threatens to devastate poultry flocks isn’t necessarily deadly to wild birds – or to humans, unless they’re unlucky enough to catch it. A headline about “deadly bird flu” is likely to send a far more alarming signal than you intended.

But the vocabulary issue isn’t the main issue. The main issue is telling the public what kind of avian influenza you have found, and what you think it means (and doesn’t mean). If you are candid and precise, reporters and the public will have a better chance of eventually getting it right.

The biggest dilemma is what to do with incomplete information. I’m a strong supporter of announcing everything you know as soon as you know it, even when there’s more you don’t yet know. That’s the best way to give the public a fair shot at understanding not just the high-path/low-path distinction, but also the fact that it isn’t an ironclad distinction. It’s also the best way to teach the public to trust that you’re not going to withhold potentially frightening information while you wait and hope that new data will prove it isn’t frightening after all.

Waiting and hoping that new data will show a possibly high-path outbreak is blessedly low-path is the high-stakes poker game some veterinary and health officials are now playing.

Last April in New Jersey, for example, authorities discovered bird flu during routine surveillance testing in an urban live poultry market. It was presumably a low-path strain, since the birds were healthy. But see #8 above: Some low-path bird flus can mutate to high-path. State officials said nothing whatever about the initial discovery until they had some fairly advanced lab tests back. Then they announced only that it wasn’t N1. They very likely knew first what the H was, since H lab results normally come in quicker. But they never said. (An apparently knowledgeable source told a reporter from the prestigious science journal Nature that it was H5, but New Jersey officials still haven’t confirmed or denied the claim.) As for the N, at least they said what it wasn’t, but not what it was. As far as I am aware, no other state has steadfastly refused to reveal what strain of bird flu it found – not just while awaiting further data, but even after all the data were in.

Here’s another example, this one from Michigan. State officials announced that routine surveillance had found low-path H5N1 in a pair of wild swans. Fine – I don’t doubt that’s what they found. But odds are they found out first that the virus was H5, then that it was N1, and then that it was low-path. (The fact that the swans weren’t sick wasn’t any help in the high-path/low-path diagnosis; H5N1 can be high-path in poultry and low-path in swans.) So for some period of time Michigan authorities kept to themselves the news that they had some swans that might or might not have the “deadly” Asian high-path H5N1, the one everyone’s worrying about. The possibility they didn’t announce turned out not to be the case, so no doubt they felt validated in their decision to suppress their partial information until they had it all.

But eventually authorities somewhere in the U.S. will almost certainly find high-path H5N1 in birds. The information will again come piecemeal, and the authorities will in all probability have suppressed the early data while waiting hopefully, and futilely, for reassuring final results. When they get around to announcing the bad news, they will be rightly pilloried for not having warned us sooner. This is normal bureaucratic behavior, but it isn’t good risk communication.

Is there any known modeling on escalation of alert levels from 3 to 4 to 5 to 6? Has anyone done any planning around the speed of each escalation, or is this not possible?

I accept that WHO internally may be using a different system (not available to the public) and probably plays down the actual risk, but if we are to truly believe the alert levels, has any appropriate modeling been done – perhaps using previous examples, for example SARS?

You write: “I accept that WHO internally may be using a different system (not available to the public) and probably plays down the actual risk....”

We have no reason to think that WHO is using a different system, or that WHO is playing down the risk. Everyone – including WHO officials – wishes there were certainty about the predictive value of each cluster, and of each new situation that arises with H5N1 in humans. When WHO investigates a cluster or other new development, it is looking for evidence of progress toward a possible pandemic. If it finds such evidence, it needs to decide – without anything approaching certainty – whether the evidence justifies notching up to the next pandemic phase.

The phases are merely ways of describing a progression of the H5N1 virus in the direction of more efficient human-to-human transmission. The actual cut-off points between each phase and the next were judiciously but arbitrarily defined; WHO could just as easily have divided the progression into more or fewer phases. These cut-offs are not nearly as crisp and definitive as we all wish they could be – and as we all sometimes imply they are. Even if the facts of a particular situation are known and undisputed, there may be room for debate over whether that situation does or doesn’t meet the criteria for Phase 4.

Nor can anybody predict how long H5N1 (or any flu virus) might linger in one phase before moving on to the next. We could move to Phase 4 and stay there for years – which would probably lead many people to conclude we shouldn’t have moved to 4 at all. Or we could move to Phase 4 and then move almost instantly to 5 and 6 – which would probably lead many people to conclude we should have moved to 4 earlier.

The phases are approximations that WHO experts hope will provide at least a little guidance to countries that are deciding how to respond to a changing situation.

Remember: This has never been tried before. We have no experience predicting pandemics – no experience with successful predictions, and no experience with failed predictions.

Everyone worries that some countries may try to hide what is happening, or may not be aware of what is happening in some remote areas. But some people worry also that WHO may try to hide what is happening. You can find entire threads on Flu Wiki and other flu-related websites devoted to this latter worry.

It’s a worry we personally do not share. Here’s why:  The more you know the people who work at and with WHO, the more you realize how deeply committed they are, and the more you understand how much they struggle against the technical and political constraints they work under. If or when the situation appears more urgent, and knowable data support that impression, we are convinced that WHO will warn the world. WHO did so with SARS, which to us is the clearest publicly available evidence that it will do so again when it thinks a warning is necessary.

(However, there are important differences between the emergence of SARS – a disease that appeared in humans before any animal source was known – and the emergence of a pandemic virus from an existing known animal source. When SARS appeared, it looked for all the world like the start of a flu pandemic – more precisely, it looked like one of the ways a flu pandemic might look when it starts. SARS turned out to be far less transmissible than flu, and it had a much longer incubation period. But in the first weeks, no one knew this, and it raised the specter of The Next Pandemic.)

As it monitors H5N1 outbreaks in animals and humans, WHO is explicitly worried about errors in both directions. If WHO raises the phase at the very first rumor of increased human-to-human transmission, or larger clusters, it may be responsible for a false alarm. It’s not just that this could lead to serious economic repercussions; it would also impair WHO’s credibility and thus its ability to warn the world effectively the next time it announced a phase change. But if WHO waits until it has evaluated every possible lab test and traced every conceivable source of exposure in a given cluster, its warning may come very late in the evolution of the virus – reducing the time available for urgent efforts to respond to the threat.

WHO knows, almost for sure, that it will end up catching some blame one way or the other. If a severe pandemic devastates the world, by definition WHO will not have warned us early enough or aggressively enough. If years go by with no severe pandemic, it will have made mountains out of molehills. Already, many who expect a severe pandemic are critical of WHO’s reticence, while many who doubt that a severe pandemic is imminent are critical of its stridency. Not knowing which set of critics will turn out right, WHO tries to chart a scientifically defensible middle course and let the blame fall where it may.

None of this means that WHO’s middle course is necessarily the right course. There are grounds for debate over whether WHO should worry more about false positives or false negatives – whether it should be more deliberative (eventually confirming a phase change that others will have announced already) or more urgent (quickly announcing a phase change while others are still skeptical). That's a different debate – a more useful debate, we think – than the one over whether WHO is suppressing facts it knows to be alarming.

Ultimately, if the virus starts causing Phase 4-level damage, WHO officials will probably be slower to announce it than we and they will wish. But it will not have been because of a conspiracy or a cover-up. It will be a delay during which officials assess the new data and debate whether they justify a change in phase.

Below are some responses to other comments and questions in your email. As communication experts who are not flu experts, we want to emphasize that our technical answers below may not be exactly correct, so please do not rely on them as definitive.

1. You write: “There seems to be some concern re the WHO alert system. Currently we are at Phase 3, although there is concern we should be Phase 4 re: Qinghai (July ’05) and the Turkey/Iraq cluster (Jan. ’06).”

We are not aware of any human cases related to the dead wild birds in the Qinghai Lake area, so from our limited knowledge, we don’t see how the Qinghai avian outbreak would have influenced the WHO pandemic phase level.

In Turkey and Iraq, there did not appear to be any human-to-human transmission, or large clusters. Here is a nice WHO description of the relevance of occasional and very limited human-to-human transmission, dated December 5, 2005, just before the human cases in Turkey started to emerge:

What is the significance of limited human-to-human transmission?

Though rare, instances of limited human-to-human transmission of H5N1 and other avian influenza viruses have occurred in association with outbreaks in poultry and should not be a cause for alarm. In no instance has the virus spread beyond a first generation of close contacts or caused illness in the general community. Data from these incidents suggest that transmission requires very close contact with an ill person. Such incidents must be thoroughly investigated but – provided the investigation indicates that transmission from person to person is very limited – such incidents will not change the WHO overall assessment of the pandemic risk. There have been a number of instances of avian influenza infection occurring among close family members. It is often impossible to determine if human-to-human transmission has occurred since the family members are exposed to the same animal and environmental sources as well as to one another. (http://www.who.int/csr/disease/avian_influenza/avian_faqs/en/index.html#isit)

A Phase 4 declaration would require “evidence of increased human-to-human transmission.” The worst known situation thus far is one instance of probably three generations of spread – huma