Talking about a flu pandemic
worst case scenario

Unfortunately, many people have trouble thinking simultaneously that a risk is horrific and that it’s unlikely. There’s nothing intellectually incompatible about these two risk characteristics; most horrific risks are unlikely. The problem is that the two characteristics are emotionally incompatible. The lesson of horrific is “take precautions.” The lesson of unlikely is “don’t worry about it.” So we tend to pick one. Either we focus on how horrific a severe pandemic could be and imagine that it’s a pretty likely scenario, or we focus on how unlikely a severe pandemic is and imagine that it wouldn’t really be so bad.

Unfortunately, many senior corporate executives also have trouble thinking simultaneously that a risk is unlikely and that it’s worth planning for. Or at least many business continuity managers believe their senior executives are going to have trouble keeping these two in mind at the same time. Whenever I advise my clients to acknowledge the low probability of a severe pandemic, they worry that doing so might undermine top management’s willingness to allocate time, effort, and money to pandemic preparedness.

Paradoxically, my clients are often equally reluctant to trumpet the severity of the pandemic worst-case scenario, for fear of frightening people too much, or seeming too fearful themselves. The result is a lot of pandemic risk communication that understates severity and overstates probability – that makes a pandemic sound almost inevitable but pretty tame.

This is exactly the wrong way to talk about worst-case scenarios. For the sake of credibility (and honesty), we need to tell people that the pandemic worst case is unlikely. And for the sake of impact (and honesty), we need to tell people that the pandemic worst case is horrific.

How unlikely? We don’t know. In the 1918 pandemic, guesstimates are that 30%+ of the world population got the flu and 2.5%+ of those who got the flu died. That was the worst flu pandemic in recorded history. So if you judge by history, a flu pandemic as bad as 1918 is very unlikely, and a flu pandemic worse than 1918, as far as we know, has never happened. On the other hand, the H5N1 virus has already launched the worst bird flu outbreak in the recorded history of the poultry industry. And as political scientist Scott Sagan has said, things that have never happened before happen all the time.

How horrific? We don’t know that either. So far, H5N1 has killed over 60% of the known human cases – 24 times the estimated case-fatality rate in 1918! Try to imagine a pandemic that infects 30% of your neighbors and kills 60% of those it infects. That means 18% of the population would die from the flu. Nobody can predict how many more would die from the collapse of society – from riots, perhaps, from other diseases (with hospitals unable to function and medicines unattainable), or from shortages of food, energy, or potable water.

Here are some tips for warning people about this low-probability, high-magnitude pandemic worst case:

Don’t claim or imply that the worst case is likely.

This is the risk-communication mistake that President George W. Bush made in the run-up to his invasion of Iraq. He told the public he was virtually certain that Saddam Hussein had weapons of mass destruction. Bush should have said that he couldn’t confirm this, but he considered even a low-probability “Saddam with nukes” scenario an unacceptable risk. I believe many environmentalists are making exactly the same mistake right now in giving the impression that global warming worst-case scenarios are almost a sure bet.

Don’t understate how awful the worst case could be.

Use dramatic imagery. Tell stories, even though they’ll have to be hypothetical ones. Paint a picture that makes the worst-case scenario real. For two stunning examples, see “The Flu Pandemic: Were We Ready?” Nature reporter Declan Butler’s fictional blog of freelance writer Sally O’Reilly, caught in a pandemic in Washington, DC, and “H5N1: My Town – a Projected Epidemic,” a novella by Susan Smith, aka CanadaSue.

Don’t rely too much on numbers to convey your worst case.

For senior management it might be about the numbers, but certainly not for employees or most other stakeholders. Most people are sufficiently innumerate that their response to a number depends almost entirely on how it’s framed. A pandemic that “could kill as many as 2 to 7 million people worldwide” sounds really serious; one that “probably would kill only 2 to 7 million people worldwide” sounds much milder. Get the numbers right – 2 to 7 million is a mild pandemic, not a worst case – but don’t expect the numbers to tell your story.

Put your worst-case scenario in context.

The mistake of overstating the worst case is less common than understating it, but it’s still a mistake. There is no such thing as a literal worst-case scenario. However horrific your scenario, I can always make it worse: “Not only that, but at the same moment the Martians invade!” Anything too far out will sound like a reductio ad absurdum and cost you credibility. For more context, remind people to consider mild scenarios, too. Higher-probability, lower-magnitude pandemics also deserve preparations. And you might want to point out that in a pandemic with a 2.5% case-fatality rate, the vast majority (97.5%) of people who get the flu will recover.

Remind business audiences that there’s nothing new about hedging.

A few months ago I was talking with the business continuity manager of a financial services company. She told me her management was skeptical about the need to prepare for anything as unlikely as a severe pandemic. I reminded her that spending money to make unlikely cataclysms less likely or less cataclysmic is standard business procedure. It’s the essence of all hedging strategies. It’s why investors diversify. It’s why farmers and mining companies support the futures markets. It’s why we all buy insurance, and why insurance companies buy reinsurance.

Copyright © 2007 Regents of the University of Minnesota. Originally published on
the CIDRAP Business Source website, March 8, 2007. Reproduced with permission.

For more on infectious diseases risk communication: