Posted: June 29, 2009
This page is categorized as:    link to Crisis Communication index    link to Pandemic and Other Infectious Diseases index
Hover here for
Article SummaryThe swine flu pandemic started in North America, and by the time the virus was identified it was already widely seeded in the U.S. So the experts judged that it was too late to try to “contain” its U.S. spread; from Day One, the U.S. was focused mostly on coping with the disease, not stopping or even slowing it. Outside North America, on the other hand, an initial containment strategy made public health sense. But containment isn’t just a public health strategy. It is also a risk communication signal of enormous importance. Containment sends a signal that the pandemic can be contained and that it must be contained – that it is stoppable and severe. Instead of countering these misleading signals, the governments of many countries have issued misleading messages to match. This is doing significant damage to the world’s preparedness to cope with the unstoppable (and soon to be pervasive) but so far mild pandemic that is just beginning.

Containment as Signal:
Swine Flu Risk Miscommunication

As the pandemic H1N1 swine flu virus spreads around the world, governments face some fundamental early decisions regarding two response strategies, containment and mitigation:

number 1
Should they try to “contain” the new virus in an effort to stave off the day when it will reach and infect their populace?
number 2
Or should they resign themselves to the inevitability of local transmission, and focus on “mitigating” its effects?
number 3
If they decide to focus on containment at first, when should they throw in the towel and switch to mitigation?

These decisions regarding containment versus mitigation are usually seen – and rightly so – as public health decisions. But they are also risk communication decisions. When a government decides to make strenuous efforts to keep swine flu from crossing its borders, and to keep imported cases from spreading locally, it is signaling something to its public about whether the novel H1N1 virus is stoppable and about whether it is severe.

And there is a fourth decision governments must make that is explicitly about risk communication:

number 4
How should they explain the goals and endpoints of containment to their publics?

Many governments doing intense swine flu containment are signaling that the purpose of containment is to prevent local spread of the virus – not just slow it or reduce its impact, but actually prevent it. Many of these governments explicitly tell their people that this is the goal; some just signal it without actually saying so. In all cases, these signals and messages are misleading, since containment will eventually fail.

Thus governments are setting themselves up to be blamed, and setting their people up to be shocked and unprepared for imminent domestic epidemics.

The misleading signals and messages about containment violate a very basic risk communication principle for helping people cope with bad news that is likely to worsen: anticipatory guidance, link is to a PDF file telling people what to expect.

The “messages” that are signaled by a swine flu containment strategy may or may not be intentional. They may or may not be explicit. But the public nonetheless reads and absorbs these implicit messages. A government that does not intend to send them must work very hard to contradict them in what it actually says about its containment strategy. And a government that says nothing to contradict the signals of containment is usually motivated by the desire to over-reassure link is to a PDF file its public, another violation of basic risk communication principles.

We decided to write a column about containment as signal for two reasons.

First, we believe the combination of the containment strategy itself and the way that strategy is being communicated has had a huge impact on worldwide public understanding of the novel H1N1 pandemic, especially in the developing world. We think the containment signal has:

  • led many publics to expect that swine flu will be prevented from entering or at least from spreading in their country;
  • led many publics to believe that swine flu is currently more severe than it actually is (so far);
  • led many governments to adopt containment measures that experts consider epidemiologically ineffective and economically harmful;
  • led many governments to continue initially sensible containment measures long after they were no longer useful; and
  • led many governments to beg the World Health Organization to delay declaring swine flu a pandemic, in part because they did not want to tell their people that swine flu’s continuing spread was unstoppable.

And when the swine flu containment strategy is ultimately abandoned, the risk communication signals sent by that strategy – if uncontradicted by explicit government communications – can lead many publics to become more mistrustful of their government and more alarmed about swine flu.

Our second reason for writing this column is to explore the dynamics of signaling. So here is a very brief introduction to the concept of “signal,” followed by an in-depth look at how countries have communicated the meaning of swine flu containment.

Signals and signaling

The distinction between words and deeds is a squishy distinction. Some statements are also actions. They’re called “performatives” – for example, “I now pronounce you man and wife.” And some actions are also statements – turning your back on someone, for example.

To make things even more complicated, statements often have meaning beyond the meaning of the words. A phrase like “there’s no need to panic,” for example, is more than a claim that panic is unnecessary. It carries the absurd implication that panic is sometimes necessary, just not right now. More importantly, the statement means, and is taken to mean, that the speaker believes the public is or will soon be overreacting emotionally to the situation at hand, and should calm down forthwith! And thus it is also insulting.

We have sometimes used the term “meta-message” to describe the meaning of a statement that goes beyond the meaning of the words themselves. Whether a speaker sounds confident or tentative, for example, is a meta-message. Whether the speaker sounds alarmed or unconcerned, respectful or contemptuous of the public, etc. – those are meta-messages too.

A meta-message may confirm the explicit message. Or it may supplement the message, adding information that isn’t in the message itself at all.

Or it may contradict the message. When that happens, the meta-message almost always overpowers the message. (Think of sarcasm, or the veiled hostility of an opponent saying outwardly pleasant things in public.) We are all more responsive to meta-messages than to the actual messages, whether we’re consciously aware of them or not.

It isn’t impossible for messaging to overpower meta-messaging. But it is very difficult. It usually requires acknowledging and explicitly disavowing the meta-message: “I know I must sound like I’m angry at you. I’m not. I’m just upset….”

Since meta-messages are so powerful, risk communicators need to exert conscious control over them. Unintentional meta-messages are always a mistake. We should be deciding whether to sound confident or tentative, alarmed or unconcerned, respectful or contemptuous. (In our family, we often quote Robert Heinlein’s exhortation: “Never insult anyone by accident.”)

The term “signal” has much the same meaning as “meta-message,” but it applies to actions as well as statements.

In short, there are always at least two levels to what we say and do. There is what we said and did. And then there is what we signaled by what we said and did.

The core question addressed in this column, then, is: What are countries signaling when they pursue a strategy of swine flu containment?

Containment as a public health strategy

Here is a representative sample of containment measures various countries (and smaller political units) are adopting in an effort to reduce or slow the spread of swine flu in their jurisdictions:

  • Closing schools under conditions where they would have been kept open during a “normal” flu outbreak, and often keeping them closed longer.
  • Sanitizing the schools while they’re closed (and sanitizing other facilities, such as amusement parks and sports arenas), even though the flu virus doesn’t last very long on most surfaces and the surfaces are recontaminated as soon as infected people start using them again.
  • Canceling events that bring large numbers of people together, including sporting events, plays, concerts, and graduation ceremonies.
  • Testing people with flu symptoms, then tracing their contacts if they test positive, and then quarantining the contacts in case they might have been infected.
  • Hospitalizing infected people with mild symptoms (presumably in order to ensure their isolation – but that is rarely clear in news stories, in which hospitalization signals severe illness).
  • Giving antiviral medications to people with mild symptoms (and no underlying medical conditions), and even to their symptomless contacts – an attempt to ring-fence cases and outbreaks with little “Tamiflu fire blankets.”
  • Treating travelers from affected countries differently from one’s own populace – in particular, screening arriving passengers for symptoms (especially fever), isolating those who test positive, and quarantining their contacts and seatmates.
  • The most extreme examples: closing borders entirely to visitors from countries where swine flu is circulating more broadly than in one’s own country (refusing entry to flights from Mexico, for example); advising against travel to those countries and quarantining travelers when they return; banning imports of pork from those countries; etc.

These measures vary in their individual effectiveness. (Only one, banning pork, has no role at all in the containment effort.) Experts agree that collectively they can help slow the spread of the flu virus, but they cannot stop it. As far as we know, nothing can stop it.

Please pause and take in the words “nothing can stop it.” An influenza pandemic is unstoppable.

A previously unknown flu virus like pandemic H1N1 is predicted to infect between one-third and one-half of the world’s population over the next year or two, not counting those who are successfully vaccinated against it (once a vaccine has been developed). In countries where there is little or no vaccine available, or where most people choose not to be vaccinated, swine flu will be pervasive. Period. That’s not controversial. It’s not even debatable. If it hasn’t sunk in yet in the minds of most people, that’s because it hasn’t been said loudly and frequently enough by official communicators.

According to most experts, the fact that a flu pandemic is unstoppable means that containment is worth doing only at the beginning, when a country (or part of a country) has few or no known cases. The containment strategy should be abandoned, they say, when the number of untraceable local cases increases, demonstrating that the virus is becoming widespread in the community.

(Note: We are talking specifically about containment of pandemic influenza. There are other infectious diseases for which containment is useful, urgently needed, and often successful.)

Why does it make sense to adopt a containment strategy at the outset, even though it will ultimately fail? Part of the answer is that containment measures don’t just delay the inevitable; with luck they can spread it out, so fewer people are sick at the same time, easing such problems as hospital overload and school and work absenteeism. This is particularly important in the southern hemisphere right now, where the onset of swine flu overlaps the start of the regular flu season. Some countries are hoping to delay the former till the latter is past its peak, in order to avoid having to cope with both peaks at once. For instance, the Associated Press reported that New Zealand was trying “to prevent the peak of the annual winter flu epidemic from coinciding with the peak in swine flu infections.”

But the main reason for containment is to buy time. So the value of containment depends on how that time is used. What, exactly, are countries buying time for? Here are some possible answers:

  • Time to manufacture or obtain a swine flu vaccine.
  • Time to prepare healthcare institutions for the onslaught of flu victims and people who are worried they might be flu victims.
  • Time to teach the populace about hygiene and social distancing.
  • Time to teach the populace that this particular pandemic is comparatively mild so far, that most victims recover even without medical help.
  • Time to teach the populace that pandemics are unstoppable, that (absent a vaccine) a typical flu pandemic will ultimately infect one-third to one-half the population.
  • Time to teach the populace that any pandemic virus can turn virulent at any moment, and that’s the main reason experts are so worried.
  • Time to help guide the populace through an adjustment reaction to this unfamiliar and alarming information.

Measured against this list, many countries are not making good use of the time they’re buying with their containment strategy.

Consider the first reason, buying time to manufacture or obtain a vaccine. Developed countries have reasonable expectations of ultimately acquiring enough swine flu vaccine to meet at least some of their needs – so delaying the spread of the disease should mean that fewer people will get sick before the vaccine is ready to administer. Most developing countries, on the other hand, have no such expectations. Although WHO has promised to provide some vaccine doses for the developing world, they will be a drop in the bucket. So it’s hard to argue that these countries have reason to delay the spread of the disease in order to buy time to get their people vaccinated.

The second and third reasons are more universally applicable. Developing countries will obviously find it more difficult than developed countries to ready their healthcare systems for a pandemic and teach their people ways of protecting themselves against the swine flu virus. But all countries can work on these tasks during their containment phase. And most countries are trying.

As for the last four reasons, very few countries – especially in the developing world – have even tried to teach their people that the swine flu pandemic is currently mild, nor that it is unstoppable, nor that it is unpredictable. To the contrary, many countries are letting the containment strategy signal almost the opposite: that swine flu is severe but the government is doing what it takes to protect its people.

Containment as a signal of prevention

Containment is first and foremost a public health strategy. But we are fascinated by its use as a profoundly misleading de facto risk communication strategy.

When governments adopt a swine flu containment strategy, they are sending a signal – intentional or not – that they are actually going to stop the virus, that they can prevent swine flu from infecting their people. In the absence of explicit messages to the contrary, the public absorbs this signal. As a result, the public is over-reassured early on, and more angry, more mistrustful, and less confident in government competence later.

The containment-as-prevention signal is the equivalent of the typical Day One official announcement that “the situation is under control.” Officials who say that really mean: “We’re working on it. Stop bothering us with your worry.” But the clear (and sometimes intentional) over-reassuring implication is: “We’ve got this licked.”

To postpone the day of reckoning, governments that have adopted a containment strategy without explaining that it is short-term are later tempted to stick to that strategy long after it is no longer epidemiologically useful, postponing the admission that containment has failed and the virus is now spreading rapidly in their own country.

Many governments have ended up in this position, and many more will in the near future, paying a price for a poorly explained early strategy of containment.

This isn’t happening only in developing countries. The United Kingdom spent weeks using an algorithm link is to a PDF file for swine flu testing that tested people with influenza-like illnesses only if they had:

Onset within seven days of visiting areas where sustained human-to-human transmission of swine influenza A/H1N1 is occurring … OR contact with a probable or confirmed case.

Having thus ensured that sustained community transmission would not be identified quickly, U.K. health officials kept assuring the public (and WHO) that the U.K. did not yet have sustained community transmission. Finally, Scottish health officials began reporting sustained community transmission in Scotland (which is part of the United Kingdom, of course), while senior U.K. officials were still denying it. Note that the U.K. was one of the countries most insistent in trying to persuade WHO not to declare Phase 6.

Also worth noting: The U.K.’s Catch-22 algorithm was based on recommendations from the European Centres for Disease Control. Several flu experts accused both the U.K. and other European countries of systematically looking only for traveler and contact cases.

Containment as a public health strategy doesn’t have to keep company with an effort to avoid finding community transmission. Containment measures should proceed in parallel with surveillance for community transmission, so health authorities will know when containment has failed (as it eventually must) and should be abandoned.

And both efforts should proceed in parallel with good risk communication. Governments should announce from the outset that containment is a short-term strategy aimed to slow the inevitable spread of the disease, explain why that is important, and forewarn the public that the strategy will be abandoned once community transmission is established. They should detail how they are looking for evidence of community transmission. And they should specify how current swine flu policies are likely to change when containment is no longer the goal, particularly:

  • future changes in who will be tested;
  • future changes in who should seek immediate medical care; and
  • future changes in who is likely to be offered antiviral treatment.

The core risk communication principle here is anticipatory guidance, link is to a PDF file telling people what to expect. It is especially important when the situation is likely to get worse.

Here are three good examples of how it can sound when a country candidly tells people what to expect about the transition from swine flu containment to swine flu mitigation:

From the Philippines:

Health Secretary Francisco T. Duque III reported Monday that the total number of cases of Influenza A (H1N1) in the country has now reached 46 after confirming 13 more cases over the weekend.

“We have reached 46 and although it’s a big number we are fortunate that all of these are just mild cases which are consistent with most of the cases of A (H1N1) we are seeing in other affected countries,” Duque said….

He announced that with the more complete profile of the A(H1N1) virus seen in local cases, which is mostly mild in nature, the DOH is set to shift its control strategy from containment to mitigation.

“This means that we only do early detection and aggressive treatment of cases as they come especially for those with high-risk preexisting conditions. Those manifesting with mild symptoms can be managed and monitored at home so that only patients at risk who develop complications will have to be hospitalized,” he explained.

With the mitigation strategy, Duque said that government will also veer away from contact tracing and instead focus on controlling the disease as is being done for other emerging and reemerging diseases in the country.

From New Zealand:

At the moment we are continuing our containment efforts, as every week we buy is important for delaying the peak of infection out past winter when hospitals are under most pressure, and giving families, schools and businesses a chance to prepare.

The way the virus has spread in Australia is what we’re likely to see here – the numbers will increase and at some stage we will have community spread.

At that point we will increase our emphasis on managing cases in the community and treat it like all flu – the difference is that many more people will be affected because the population is not immune to this new virus. While swine flu has been relatively mild so far, it will be serious for some.

From Singapore:

Countries were advised [by WHO] to assess their specific situation and make a timely transition from focusing national efforts on containment to focusing on mitigation measures…. In terms of control measures, we will be in containment mode for as long as it is practical to do so, and for as long as there is no community spread here in Singapore…. However, with a level 6 pandemic, the shift to mitigation measures may be inevitable. H1N1 has so far been reported in 74 countries and it may be a matter of time that we have community spread in Singapore. To prepare ourselves, we will gradually pace the shift of our control strategy from containment, i.e. “stopping the spread of the virus” to mitigation, i.e. “caring for those who fall ill.”

In the absence of clear communications like these, the implicit assumption of any precaution is that it is intended to succeed. Unless communicators aggressively assert that the containment strategy is foreordained to fail, and explain why it is worth doing anyway – to buy time for specific reasons – the public will naturally get the impression from the strategy itself that it is actually supposed to work: that it will keep the locals from getting swine flu and giving it to each other.

A containment strategy, in short, automatically signals the erroneous expectation that containment will succeed. Only a few countries (including the Philippines, New Zealand, and Singapore) have worked hard to countermand that signal by teaching their public that it is only a short-term strategy to buy time.

(Added: July 5, 2009)

“Containment” is a word that seeds its own failure.
How about: “Deceleration”?

The word “containment” implies an attempt to stop rather than slow the spread of disease. So it is intrinsically misleading with regard to the initial response to a novel influenza virus that has already “seeded” (another term of art) numerous locations and therefore cannot be stopped.

After reading this article, one reader wrote us:

About containment, it always struck me as the wrong word, because it either works or not. And it wasn't ever likely to work. So the name itself was a problem – indicating a success or failure.

Another reader suggested “deceleration” instead:

Squeezed in between containment and mitigation is deceleration. Slowing the arrival of a threat provides value in extended preparation time, and possible mitigation of the adjustment reaction…. It's a matter of setting expectations, discussing them transparently, and then marshalling support for the proactive plan with everyone aware of the reasonable goals for the recommended course of action. That disclosure [of expectations] should be made forthrightly by governments, to counter the “signal” implied by a dramatic containment display.

“Containment” is a standard word in epidemiology, and we never thought to question it. So we found ourselves recommending that leaders tell their publics: “Containment will ultimately fail. We know that, but it's still worth doing because….”

If leaders had a word that implied “delay,” they could say, “Our goal is to slow the initial spread of the pandemic a bit. Widespread community transmission is almost inevitable. But even a short postponement will give us some time to do X and Y to get ready.” Leaders could talk prospectively about their deceleration efforts, instead of warning about the inevitable failure of containment (the advice we gave) or giving the impression the swine flu virus can be stopped (the practice we criticized).

As a Time magazine reporter wrote early on in the SARS outbreaks of 2003:

The language of epidemiology, the study and prevention of the spread of infectious diseases, is steeped in the metaphor of blockade. Doctors and scientists develop “barrier” nursing techniques and try to erect “ring fences” within society to corral a dangerous microbe, preventing it from jumping between people and countries.

There are times, early on in some outbreaks of some diseases, when containment by way of isolation, quarantine, “ring-fence” prophylaxis or vaccination, and other measures, can stop an outbreak in its tracks. As far as flu is concerned, the World Health Organization has an experimental rapid response plan for trying to stomp out the earliest outbreak caused by a novel flu strain, in the unlikely event that it is promptly identified in one discrete location, such as a rural Thai village. No one is very hopeful that this would work (it has never been tried), but what a gift to the world if it did!

The conditions required to try that kind of containment were not present by the time Pandemic Influenza Virus A (H1N1) 2009 made itself known. Deceleration, but not containment, was a reasonable initial goal – and still is as we write this, in the diminishing number of countries that don’t yet have sustained community transmission. It would help if “deceleration” were the term we all used.

By contrast, many countries have gone beyond merely letting containment itself signal that containment will work. They have implied this in their messaging as well – and in some cases have even explicitly promised containment success.

From Qatar:

Dr al-Thani [Public Health Director Dr Mohamed al-Thani] claimed that the precautionary measures being taken by Qatar could not be matched anywhere else, even in the US where the swine flu virus is already confirmed.

He said that the authorities had put in place measures to check if any passenger has fever or flu-like symptoms and necessary procedures are being followed to protect other passengers.

“Swine flu is not pandemic here, so we can’t prevent any person from travelling out of Doha and the best we can do is to advise them not to travel if they have fever,” he said while reassuring the public not to panic, adding that all necessary precautions are being taken to ensure that the flu does not get here.

From Malaysia:

The public has been urged not to panic over the country’s first locally transmitted influenza A (H1N1) case….

The authorities have so far quarantined 20 of the girl’s family members and friends with whom she had come in contact with.

Liow [Health Minister Datuk Seri Liow Tiong Lai] also urged the public against panicking as talk of implementing “social distancing” in Petaling Jaya loomed.

“The ministry is taking steps to ensure the infection does not spread further. We are taking all necessary steps. The situation is under control.”

From Bulgaria:

All measures have been implemented by the health ministry and the national emergency pandemic plan to contain the strain and to maintain vigilance and effective surveillance at airports, ensuring that the pandemic will not spread beyond control, the report says.

The most explicit, excessive, and repeated over-reassurances about containment have come from India.

From The Hindu, June 11, 2009:

Delhi Principal Secretary (Health and Family Welfare) J.P. Singh said: “We are following a set protocol and care is being taken to ensure that the general public is kept at zero risk.”

From “Times Now TV,” June 12, 2009:

‘H1N1 is under control in India’

Reacting to the panic caused by the World Health Organisation’s (WHO) late night decision to increase the H1N1 virus alert to 6 (the highest), the Health Minister Ghulam Nabi Azad on Friday (June 12) spoke exclusively to TIMES NOW saying the Indian government was well prepared to deal with any outbreak of swine flu and that adequate measures were being taken to halt the possible spread of the virus in the country….

“There is no need to panic over WHO’s decision to raise virus alert. We are fully prepared to deal with the H1N1 virus,” said Azad today, adding “Proper screening stations are being set up across the country, at all airports to keep a check on the flu….”

The Health Minister also said, “There is absolutely no need for people to panic as the virus is a 100 per cent curable.”

And from Asian News International, June 18, 2009:

Swine flu totally under control in India, says Azad

Union Health Minister Ghulam Nabi Azad on Thursday said that the H1N1 ’swine flu’ virus is totally under control in India as, out of 35 cases reported, 12 have already been cured.

Azad said that due to the effective precautionary measures taken by the Union Government, extend of H1N1 virus has been brought under control.

Following reports of the dangerous virus, screening of international passengers had been put in place at all entry points to the country, including airports and sea ports.

“Swine flu is totally under control. With the size and population of our country, 35 cases are negligible and at the moment 12 have already been discharged and the rest 23 are undergoing treatment at various hospitals,” said Azad.

India faces stiff competition for the worst-in-class award from Thailand. Typical is this paragraph from a June 14 Bangkok Post article:

The [Thai health] minister told the people not to panic as there was no domestic pandemic of the virus and that all of the patients … contracted the swine flu from abroad.

Earlier, the Thai government was caught trying to cover up its first known swine flu cases. Then the health ministry explained that it covered up the cases out of a fear that people would panic and tourism would decline. The health minister was clearly frustrated that the cover-up was “necessary“ in the first place, and that its discovery was newsworthy. His comment on the cover-up: “Can’t the public just accept that a highly competent government is right on top of the problem and there is absolutely nothing to worry about?”

Whether a country relies on the signal of containment itself, or bolsters that signal with explicit promises (and perhaps by hiding cases), the result is the same: People “learn” (mislearn) that containment can keep swine flu from the door. And once your people have been led to expect containment to work, you have reason to expect that if sustained community transmission is uncovered, your people will find the news frightening and will blame you for the failure of your containment strategy. Thus you have pretty much set yourself up to delay finding and reporting sustained community transmission.

It’s almost unnecessary to point out that this is awful risk communication. It trades short-term reassurance (“we will protect you from swine flu”) for long-term fear, anger, and mistrust. It has more collateral disadvantages than we want to list – for example, making it harder for governments to rebut false rumors of local outbreaks.

Perhaps the most important collateral disadvantage: Once a country has implied that containment will work, it is much harder for health officials to resist public and political pressure for escalating containment efforts. Close all the schools? Quarantine every visitor? Shut down the airports, even? (There have been no airport shutdowns yet.) Well, if that’s what it takes….

A public that understands containment as an earnest attempt to delay the inevitable spread of the disease will want to know how much delay they’re buying and what they’re paying for it – in money, inconvenience, lost freedom, lost education, etc. Even more important: A public that understands the containment goal will be less shocked, and more supportive, when containment has run its course and government policy shifts to mitigation – trying to manage the disease instead of trying to stop it.

A public that thinks containment is an all-out effort to defeat the novel H1N1 virus is much likelier to demand excessive containment.

Containment as a signal of severity

Just as containment efforts automatically send a signal that containment is expected to work, they also automatically send a signal that it is extremely important for containment to work. The more containment measures a government undertakes, and the more burdensome those measures, the more clearly the government is signaling that keeping everyone from getting swine flu is a very high priority. It follows that getting “the dreaded swine flu” must be really, really bad.

Governments could explicitly countermand this signal, pointing out that swine flu is in fact pretty mild so far. Then of course they would need to explain why they are investing so heavily in containing a disease that is mild.

Some governments that have been clear about the “buying time” goal of containment and the current mildness of pandemic H1N1 have also done a good job of answering the obvious follow-up question: Why buy time by slowing the spread of a virus that will cause mostly mild disease when it gets here? Some of the answers:

  • Even a relatively mild flu isn’t a nonevent for the individual. So we need time to teach people ways to reduce their chances of getting it.
  • When a lot of people are sick at the same time, even if their illness is mild, the societal burden of widespread absenteeism and overwhelmed healthcare systems can be significant. So it is useful to try to “flatten the epidemic curve” so the inevitable illnesses are more spread out over time.
  • This currently mostly mild pandemic includes some severe cases, especially in people with other medical problems. So we need time to try to prepare healthcare facilities to cope with the increased number of patients.
  • Pandemic flu viruses are unstable. This one could mutate or reassort into a more severe strain at any time. So we need time for everyone to prepare for that scenario to the extent that they can.

But many governments doing intense containment are not asserting that the swine flu is currently mild, and then explaining why it still merits a containment strategy. Instead, they are trumpeting the containment strategy, and signaling that swine flu is currently severe.

This is especially typical in the developing world, where many governments have tried to hospitalize every suspected swine flu case (particularly inbound travelers and their subsequent sniffling contacts), while almost never mentioning that the vast majority of cases elsewhere – and the few in their own country – are relatively mild.

Swine flu hospitalizations in countries with only a few cases are almost always for purposes of isolation, not treatment. They are part of the containment strategy. Yet official statements and media coverage quite often leave the impression – without saying so outright – that the country’s first few swine flu victims needed to be hospitalized because they were severely ill. This impression is particularly strong in countries with recent memorable severe disease outbreaks, such as China (SARS) and Malaysia (Nipah virus).

Consider for example this June 23, 2009 Xinhua story:

The Chinese mainland confirmed 49 new A/H1N1 flu cases from 6 p.m. Monday to 6 p.m. Tuesday, bringing the total to 490, with no reports of deaths….

Among the patients, 251 have been discharged from hospital while the other 239 were still being treated.

Or this story from Malaysia’s New Straits Times:

Seven more new Influenza A (H1N1) cases, including three siblings and the second local transmission, were reported yesterday, bringing the total number of confirmed cases in the country to 42….

Out of the 42 cases, 30 were still receiving treatment at [various hospitals around the country].

Here’s the worst case for exaggerated severity signals. Take a country whose people don’t experience an annual flu season and have little prior familiarity with what “influenza” means. Give that country a handful of identified swine flu cases to date, all of whom have been immediately hospitalized, regardless of the mildness of their symptoms. Give it a government whose swine flu strategy stresses containment, and whose swine flu messaging doesn’t stress mildness. Little wonder millions of people get the impression that the “dreaded swine flu” is extremely virulent – not just potentially, but right now.

Although signaling that swine flu is currently severe has been characteristic mostly of developing countries, some developed countries have also reported hospitalizations without clarifying that the people hospitalized weren’t necessarily very sick.

Singapore, for example, usually does excellent risk communication. But with 40 identified swine flu cases as of June 12, 2009, Singapore’s Ministry of Health stated: “So far, 17 patients have been discharged. The remaining 23 patients (18th – 40th cases) are still in the hospital and their conditions remain stable.”

In sum, this has been the swine flu risk communication pattern in a lot of countries dealing with their first handful of cases, especially (but not only) in the developing world: unduly over-reassuring signals about the efficacy of containment combined with unduly alarming signals about the current severity of those first few cases.

We realize this is a gross overgeneralization, especially since we are limited to the English-language press of mostly non-English-speaking countries. And in fairness, it is worth remembering that the health and societal impacts of swine flu will genuinely be less mild in developing countries than in the U.S. and other developed countries, simply because their healthcare systems are less robust and their populations have more underlying health problems.

Still, the pattern is real, and it is problematic. In the weeks ahead, we hope to see more developing countries base their reassurances on the accurate claim that swine flu is relatively mild so far, not on the misleading signal that containment will protect them from its ravages.

Containment and the WHO Phase 6 declaration

In its six-phase partition of the pathway to pandemic, WHO went from Phase 3 to Phase 4 on April 27, and two days later it moved to Phase 5. Many experts expected a quick ratchet up to Phase 6.

But in fact it took six weeks – until June 11 – for WHO to declare Phase 6, a full-fledged swine flu pandemic. Along the way, WHO also altered its definition of Phase 6 in a way that helped justify the delay.

Throughout the six weeks, WHO steadfastly maintained that the delay was grounded in science, not politics. Though many flu experts complained that the delay was scientifically unjustified, WHO officials said they were just waiting for incontrovertible evidence of sustained community transmission of the novel H1N1 virus outside the Americas, to meet the Phase 6 criterion of sustained transmission in at least two different WHO regions.

But political pressure definitely played a major role. A number of member states, including the United Kingdom and China, strongly urged WHO to delay declaring a pandemic, because they were not ready for a declaration that might unduly frighten their publics, leading to increased public pressure on national governments to implement epidemiologically futile and economically harmful containment measures such as border closures. The countries wanted more time to prepare their publics (and also to rethink any automatic Phase 6 triggers in their pandemic plans).

WHO expressed considerable sympathy for the member states’ entreaties. On May 22, for example, Interim Assistant Director-General Keiji Fukuda said that WHO was “trying to walk a very fine line between not raising panic and also not becoming complacent.” Leave aside that there is no fine line between panic and complacency, but rather a yawning chasm. Dr. Fukuda was responding empathically to governments’ fears of frightening their publics. Eventually, WHO also found ways to signal scientists and flu watchers – without directly saying so link is to a PDF file – that it realized the conditions for declaring a pandemic had in fact already been met.

Though we never thought swine flu panic was a realistic concern, we understood that public anxiety about novel H1N1 was higher in the rest of the world than it was in North America. In early June, we interpreted this as mostly a matter of timing: North America got the virus first, so North America became alarmed first, so North America got through its adjustment reaction first and settled into the New Normal first. The rest of the world would move along soon enough, we thought.

As Peter wrote in his June 4, 2009 “Swine Flu Pandemic Communication Update”:

Countries with only a few confirmed cases so far are following those few with nervous fascination. Countries that have just begun to confirm larger numbers of cases are following the rising tally with the same nervous fascination.

Unless something changes – like an increase in severity – there is every reason to expect that the rest of the world will probably follow North America’s example: People will get through their adjustment reactions. Then they will sink into complacency and accuse officials and the media of fear-mongering.

Though we weren’t very worried about a public overreaction to the Phase 6 declaration, we figured that the key to avoiding such an overreaction was making it crystal clear that the declaration didn’t signal an increase in the severity of the virus, only in the geographical breadth of its distribution. (If the virus does become more severe, greatly increased public alarm will not constitute an overreaction.) So we were among many who advocated the creation of a pandemic severity scale, so that WHO could separately characterize how widespread novel H1N1 had become (widespread enough to constitute a pandemic) and how severe the resulting illnesses were (mostly pretty mild so far).

We assumed that once countries had gone through their initial adjustment reactions, severity would be the linchpin of worldwide attitudes toward swine flu. That had been true in the United States. The U.S. public had been appropriately alarmed about novel H1N1 until it learned that the disease was usually mild; then it became unduly complacent.

We hoped a severity scale, independent of the pandemic phases, would help people distinguish the severity issue from the pervasiveness issue. We hoped such a scale might also help people distinguish current severity from potential future severity. Thus it could serve as a bulwark against both overreaction and complacency. As Peter wrote to CIDRAP News reporter Robert Roos:

People who are inclined to overreact to the Phase 6 declaration may find solace in the declaration that it’s only “Phase 6a” (or however it’s phrased), not the dreaded “6c.”  And people who are inclined to dismiss the whole question of pandemic preparedness may find reason to withhold judgment since there may be more severe sorts of pandemics yet to come.  Ideally, the term “pandemic” will ultimately trigger a response something like our response to the word “snowstorm”: “How bad this time?” [A hurricane might be a better comparison: “How bad does it look at this point? And how bad could it get?”]

But we now believe that the mild-versus-severe distinction is not at the heart of worldwide reactions to swine flu the way it was in the U.S. In many developing countries, there has been little effort to teach people that the pandemic H1N1 virus is mostly mild so far. In places where people have the impression that H1N1 is severe already, governments are repeatedly telling them not to panic – because “we don’t have any cases yet,” because “we have only imported cases,” because “we are isolating all the cases we find and quarantining their contacts,” because “everything is under control,” and for many other misleading reasons … reasons grounded not in the mildness of the pandemic but in their (doomed) determination to keep it from spreading.

In these countries, there was little reason for WHO to worry that people might overreact if they got the misimpression that a pandemic declaration meant novel H1N1 had become more virulent. And there was little need for WHO to battle complacency by teaching people that novel H1N1 could some day become more virulent. People thought it was highly virulent already.

In many countries, the big issue did not turn out to be the severity of the virus. It was – and still is – whether containment efforts will succeed or fail.

This puts the WHO Phase 6 declaration in a different light.

Declaring that swine flu is officially a pandemic doesn’t mean that it is or is not severe. So in places where severity is the key to public concern, clarifying that the H1N1 pandemic isn’t severe but could become severe is the key to creating a public that is neither overreacting nor complacent. (Not that we think this is easy; initial over-reactions, followed by prolonged and excessive complacency, is a frequent response to bad news about future possibilities.)

But declaring that swine flu is officially a pandemic really does mean that containment will fail.

One core meaning of the term “influenza pandemic” has always been that it’s unstoppable and will eventually get pretty much everywhere. The debate over how WHO should define the term has never focused on whether flu pandemics can or cannot be stopped. Flu pandemics cannot be stopped until they have run their course. The definitional debate was about how many different places needed to confirm sustained community transmission before WHO officially declared that the world had an unstoppable-by-definition flu pandemic on its hands.

And of course if flu pandemics are unstoppable, then containment is a doomed strategy – a useful, perhaps even crucial, but nonetheless temporary expedient to buy time to prepare for the inevitable. As Director-General Chan put it in the text of the June 11 declaration link is to a PDF file: “You can only do containment for some weeks and if you are seeing continuing spread of infection in the community, it is important that they move to mitigation measures.”

Dr. Chan was even more explicit in her pre-declaration briefings with key reporters.  Here’s how the Associated Press reported the story:

“The world is moving into the early days of its first influenza pandemic in the 21st century,” Chan told reporters. “The virus is now unstoppable.”

“However, we do not expect to see a sudden and dramatic jump in the number of severe and fatal infections,” she added.

So swine flu is unstoppable but so far it’s fairly mild. (It still causes significant disease – that’s another core meaning of the term “influenza pandemic.”) This is reassuring news, on the whole, if you’re in the U.S., where swine flu has been clearly unstoppable for a month already. But it could be very alarming news in the scores of countries that still say they’re going all out to stop it, especially if these countries are not hearing the “fairly mild so far” part of the message.

Somewhat belatedly, then, we understand better what many WHO member state governments were afraid of. To a large extent they had created their own problem by the signals they had sent – but it was still a real problem. They had reason to worry that a WHO pandemic declaration might increase anxiety in populations that had been over-alarmed about the current severity of most swine flu cases and had been over-reassured about their governments’ ability to keep it out.

In the days and weeks since the June 11 pandemic declaration, we haven’t seen a sudden upsurge in worldwide swine flu anxiety, much less swine flu panic. But many countries’ people and politicians do continue to be upset about “the dreaded swine flu” – not upset about what it might turn into, but upset about what they think it is already.

Maybe they’re a little more upset than they were before the declaration.

And in the days and weeks since the June 11 declaration, pressure on governments to make containment work – an impossible demand – has visibly increased. In many countries, citizens and opposition party officials have become upset about their governments’ failure to do better entry screening and contact tracing, for example, or about “belated” school closures.

The pressure to implement technically unsound containment measures surfaced even in the Philippines, where the government has been clear about the relative mildness of most cases, and where the shift from containment to mitigation is well underway. On June 23, officials “closed down” the lower house of Congress for five days and sent 3,000 workers home after a staff aide died of swine flu – the first swine flu death in the country. The purpose of the shutdown, officials said, was “to allow ‘decontamination’ and sanitation of all buildings” and “the necessary contact tracing of possible victims and their families.” When at least low-level community transmission is already occurring, as it is in Manila, sanitizing buildings and tracing contacts are not a good use of resources.

Most governments that were reassuring their people by over-promising containment rather than by stressing relative mildness before the June 11 declaration are still doing so. Their people, as far as we can tell, are still buying it.

Except for tourists caught in quarantine nets, there has been little public protest against excessive and futile containment measures. The protests, where they occur, are on behalf of more containment – for example, protests that more schools should be closed, and faster.  People are not saying to their leaders: “How dare you promise to protect us from the dreaded swine flu when Margaret Chan says it’s unstoppable?” 

Containment over-reassurance is still working, in other words, despite the WHO pandemic declaration.

The short-term cost of containment over-reassurance is bad public health policy. People hold their governments to their promise of successful containment, and therefore exert pressure for containment measures that may be expensive and harmful, and are certainly futile.

The long-term cost is what happens when it becomes clear that containment has failed. Containment will fail, everywhere. Over-reassuring leaders will not have prepared their people for the current reality of a mostly mild but soon-to-be-widespread disease, or for the prospect of a long-lasting pandemic that may become more severe.

And they will not have prepared their people for the immediate changes in policy that will occur once they stop trying to stop the spread of pandemic H1N1 and commit instead to coping with it and mitigating its effects.

Suddenly they will have to explain that even though they have been hospitalizing mildly symptomatic people for weeks, now such people should stay home; that even though they have been giving Tamiflu to everybody with symptoms (and many of their contacts as well), now the Tamiflu is being saved for people who are very sick or have underlying medical conditions.

Unprepared for these shifts, people will be understandably confused, mistrustful, angry, and alarmed. Instead of apologizing for blindsiding them, their governments will patronizingly assure them – yet again – that there is no reason to panic.

Whether pandemic H1N1 remains mostly mild or turns more virulent, it will continue to spread, until it has run its course and (probably) becomes a new seasonal flu strain. By that time, one-third to one-half of the world’s population will have had it (or been vaccinated against it).  Long before then, people will realize that it is, indeed, unstoppable.  And they will have that much less confidence in their governments, for having claimed or signaled otherwise.

Maybe they won’t be all that surprised.  It’s not as if their governments haven’t misled them before.  But over-reassurance about swine flu containment will add one more brick to the wall of public health mistrust.

In the wake of WHO’s June 11 pandemic declaration, some countries (including New Zealand and Singapore) issued statements candidly explaining that their containment measures were temporary. Most didn’t. The June 12 post-declaration statement issued by the United Kingdom, for example, included an excellent explanation of the purpose of containment, but made no mention that containment in the U.K. was likely to fail (or had already failed).

Australia’s gradual shift from containment to mitigation coincided with the WHO pandemic declaration. WHO Western Pacific spokesman Peter Cordingley told a Bloomberg News reporter on June 13:

Melbourne has pulled back to a defensive position, it’s realized that basically you can’t stop this virus…. This virus is being seeded in most countries by people traveling from other countries. It’s inevitable…. Trying to beat back the virus is just going to be a waste of resources and manpower…. It’s now all about mitigation and looking after those at risk, like those with underlying health conditions.

Much more typical, sadly, was this June 22 article from the small island nation of Fiji, dependent almost entirely on tourism.

Containment’s critical for Fiji: Dr Samuela

It will take more than a month to determine whether stakeholders are able to contain the spread of the dreaded Influenza A(H1N1) virus, says acting deputy secretary public health service, Dr Josaia Samuela….

In much of the world, the containment strategy – and the signal that swine flu is severe but stoppable – is the heart of swine flu risk communication and continues to reign supreme. For now.

Containment in the U.S.

On April 24, 2009, then-acting CDC Director Richard Besser announced the apparent – and soon to be proved – connection between the severe respiratory disease outbreaks in Mexico and eight confirmed U.S. cases of novel influenza. That day, at the start of what was to become the H1N1 pandemic of 2009, Dr. Besser said that “containment is not very likely” because “we’re seeing cases in Texas and we’re seeing cases in San Diego without any connection between them which makes us think that there has been transmission from person to person through several cycles.” 

From the outset, then, the U.S. government believed – and said – that containment was a lost cause in the U.S., not worth attempting.

For a while, the U.S. did try to track cases and their contacts. Especially in the first week or so when the initial impression from Mexico was of an extremely virulent influenza, the main goal of tracking was to find new cases quickly so they could be treated as early as possible. Later, when it became clearer that the virus was fairly mild, tracking helped identify where there was already sustained community transmission and where there wasn’t yet. But stopping the spread of novel, soon-to-be-pandemic H1N1 was never the goal in the U.S.

Various U.S. cities and states nonetheless adopted local containment measures, especially in the early days of the epidemic, when it wasn’t clear yet that it was relatively mild. Later, after there was obviously sustained community transmission of relatively mild disease, some U.S. communities still resorted to containment measures considered unlikely to help reduce flu transmission: school closings, extreme attention to sanitizing surfaces, provision of Tamiflu (particularly in schools) to symptom-free contacts of flu patients, etc.

In his first announcement, Dr. Besser had warned:

Our recommendations, advice, approaches will likely change as we learn more about the virus and we learn more about its transmission.

I want to acknowledge that we’re likely to see local approaches to controlling the spread of this virus, and that’s important; that can be beneficial; that can teach us things that we want to use in other parts of the country…. Because things are changing, because flu viruses are unpredictable and because there will be local adaptation, it’s likely that any given moment there … may be confusing or conflicting information available. 

The issue of whether and when to close schools soon became emblematic of “local approaches” and “confusing and conflicting information.”

Early on, when the outbreaks in Mexico were justifiably alarming, the CDC had recommended “consideration of school and child care program closure.”

Nearly a month later, on May 22, the CDC decided that “the large number of confirmed or probable cases … makes individual school and child care program closure less effective as a control measure.” Therefore, it said:

School dismissal is not advised for a suspected or confirmed case of novel influenza A (H1N1) and, in general, is not advised unless there is a magnitude of faculty or student absenteeism that interferes with the school’s ability to function.

But the CDC appropriately left communities some wiggle room, noting that school closure decisions should be based on “local considerations, including public concern and the impact of school or child care program absenteeism and staffing shortages.”

Throughout the spring, there was confusion and anger over what parents perceived as inconsistent, sometimes misleading, and even heartless local decisions about school closures. There were accusations of both overreacting and underreacting – but the latter were naturally more heart-rending and newsworthy, especially when family members of flu victims blamed the death of their loved ones on the schools staying open.

Other measures – such as giving Tamiflu to healthy classmates of flu victims – led to similar controversies. “Why aren’t you giving my kid Tamiflu? They’re doing it in that other school. You did it last time.”

Once there is ongoing community transmission of a pandemic flu virus known to be fairly mild, the chief purpose of these measures – school closures, contact prophylaxis, the obsession with hand hygiene and sanitization of surfaces – isn’t actually containment. It is reassurance. The measures are faux containment. But local officials rarely say so. Instead, they signal that they are determined to keep people, especially children, from catching swine flu.

This violates a very basic principle of risk communication. In terms of the hazard-versus-outrage distinction, it is applying a hazard solution to an outrage problem.

In his June 4, 2009 “Swine Flu Pandemic Communication Update,” Peter wrote about “precautions as reassurance” – the tendency to try to reassure the public with otherwise inadvisable measures against swine flu. He wrote:

Assuming some people are genuinely overanxious and in need of reassurance – a big assumption – is it helpful to respond to their anxiety with precautions that are technically marginal or even potentially harmful? The right answer for potentially harmful precautions is virtually always no. The answer for marginal precautions, I think, depends on whether such precautions will actually be experienced as reassuring. Often they are experienced instead as evidence that the risk is huge, and later (when these marginal precautions are abandoned) as evidence that the authorities are irresolute, inconsistent, and uncaring…. Marginal precautions are probably likelier to be experienced as reassuring if they are explained as responses to public anxiety, rather than simply being deployed as if the authorities thought they were technically sound.

The CDC’s May 22 update had acknowledged that communities might want to close schools in response to “public concern.” When that’s the purpose of a school closure decision, it is good risk communication practice to say so, empathically and respectfully. Do not signal that you’re trying to “contain” the outbreak. School officials who decided not to close schools usually leaned on the CDC’s conclusion that school closures are not likely to contain the outbreak. But officials who closed schools typically signaled that it was, indeed, for containment, not “just” because of parental concern.

It doesn’t have to be that way. In the early days of the 2003 SARS crisis, for example, there was tremendous pressure on Singapore’s government to close the schools. But the Ministry of Health said there was no medical reason to do so. Prime Minister Goh Chok Tong publicly described a cabinet meeting at which the various Ministers argued about whether or not the schools should be closed. The final decision, he said, was to close them, even though some thought it was over-reacting.

Then the Ministries of Education and Health issued a joint statement describing the school closings as a “precautionary step.” Despite the lack of medical grounds, the statement said, “principals and general practitioners have reported that parents continue to be concerned about the risk to their children in schools.” As we commented at the time:

In one sentence, Minister Teo Chee Hean assured four groups of stakeholders that they were being heard and taken seriously: principals, general practitioners, parents – and the general public.

Local faux containment is one piece of the U.S. swine flu containment risk communication story. The other piece, arguably the bigger piece, is the failure of the U.S. to support other countries’ initial containment efforts – and its failure to explain, or even to acknowledge, its failure to support those efforts.

Remember, the U.S. is the mother of all swine flu-exporting countries. But aside from urging people repeatedly to stay home if they’re sick, the U.S. government has made no effort whatsoever to prevent people with swine flu from boarding planes bound for foreign lands. There isn’t a lot of evidence from previous outbreaks that airport “exit screening” helps much. But there is some anecdotal evidence that awareness of exit screening does deter sick people from trying to get on a plane.

Many countries have tried to implement “entry screening,” aiming to identify and isolate symptomatic travelers as they enter the country. Some have also quarantined travelers whose seatmates were symptomatic, and even tracked down and quarantined travelers whose seatmates later became symptomatic. The evidence on the efficacy of entry screening is mixed. It didn’t seem to help much during the SARS epidemic of 2003. But SARS had a far longer incubation period than influenza, so most infected incoming travelers had no symptoms yet. Some countries now think that entry screening is a worthwhile containment measure against the flu.

Of course entry screening, like other containment measures, is pointless once a country has widespread, sustained community transmission. But that doesn’t explain why the U.S. nonchalantly exported swine flu to country after country – countries that were trying to delay local transmission as long as they could.

Both entry and exit screening are part of U.S. pandemic planning. Supplement 9 of the Pandemic Influenza Plan of the U.S. Department of Health and Human Services, for example, is entitled “Managing Travel-Related Risk of Disease Transmission.” It includes this intriguing paragraph:

If an influenza pandemic begins outside the United States, public health authorities might screen inbound travelers from affected areas to decrease disease importation into the United States.  If a pandemic begins in or spreads to the United States, health authorities might screen outbound passengers to decrease exportation of disease.  Early in a pandemic, state and local health departments might also implement domestic travel-related measures to slow disease spread within the United States. 

When swine flu struck, the U.S. made some initial efforts to screen people crossing the border from Mexico. It made no effort whatever to screen domestic passengers or outgoing international passengers.

The United States is quick to tell other countries how we are helping the world do battle against pandemic H1N1 – with lab support, teams of epidemiologists, etc. But we do not talk about why we refused other countries’ demands for exit screening. The U.S. has never yet acknowledged or explained this failure, never acknowledged how much anger it has provoked in other countries, and never apologized.

Containment versus pervasiveness

Here’s a statistic we have mentioned several times in this column, but now we want to put it front-and-center: Like other influenza pandemics, the swine flu pandemic WHO declared on June 11 is expected to infect one-third to one-half of the world population over the next couple of years.

That’s just an educated guess by flu experts. Every flu pandemic is different, and this is only the fourth one in the past century. Many experts think people who were exposed to earlier H1N1 viruses in the years before 1957 may have partial immunity. And depending on the success of work to develop and mass-manufacture a vaccine against this pandemic H1N1 virus, some of the world’s people may get vaccinated in time to prevent infection. On the other hand, the world is more interconnected now; there are very few “isolated” places left. On balance, one-third to one-half is still the best guess.

With 6.7 billion people in the world, that means we should be planning for roughly two to three billion cases of swine flu.

The WHO tally stands at 59,814 cases as of June 26. Of course nobody is really trying to keep count; most people with flu symptoms don’t get tested to see if it’s swine flu. The U.S. tally stands at 27,717 as of June 26 – but on June 25 a CDC surveillance official estimated that as many as a million Americans may have already been sickened with novel H1N1. Say it’s two million worldwide. That leaves 1,998,000,000 to 2,998,000,000 cases yet to come.

For the U.S. population of 300 million, we should be planning for 100-150 million swine flu cases, of which we have experienced roughly one million so far.

Few governments anywhere in the world are working hard to give their people an accurate impression of how pervasive swine flu is expected to be.

When we talk to friends and acquaintances, we sometimes probe for this understanding. Almost invariably, it’s not there. Most of the people we talk to get it that swine flu isn’t over; they think it has receded from the northern hemisphere for the summer but is expected to return in the fall.

But since almost every swine flu news story they’ve ever seen or heard featured some small tally of confirmed cases (local, statewide, national, international), they grossly underestimate the actual number of cases so far. And since no news story they’ve ever seen or heard has explained that pandemics spread like geometric progressions (maybe even fractally), they picture it returning in the fall pretty much the way they think it was this spring, with cases by the handful, not by the busload. Even those who understand that pandemic H1N1 might turn more virulent haven’t quite absorbed that, virulence aside, H1N1 is expected to get much, much, much more prevalent.

And that’s in the United States, where swine flu was confirmed first, where it has been confirmed most often, and where the government has been candid about the fact of community transmission from Day One.

So we still see many headlines like this one, from a June 20 article in a western New York State newspaper: “More Swine Flu Cases Possible in Tompkins County.”

And we still see passages like this one, from a June 18 article in a Minnesota newspaper:

Do not panic.

Do prepare and be thinking about how your family and business might deal with a seriously threatening local H1N1 flu outbreak.

It’s not here yet, thank goodness. It might not come.

That wonderful second paragraph, caught between those two awful paragraphs!

One of the most important things for people to know about the swine flu pandemic is that it’s just starting. It may or may not get a lot worse in terms of virulence, but it almost certainly will get a lot worse in terms of pervasiveness.

So how bad is it to go through a pandemic that infects one-third to one-half of the population over a couple of years? After all, the ordinary seasonal flu is estimated to infect 5 to 20 percent of the U.S. population every year, most of them in the winter – and we get through that without much fuss.

Nobody has the answer to that key question. Nobody even knows whether the pandemic toll will come on top of the seasonal flu toll, or whether pandemic H1N1 will crowd out one or both of the two seasonal influenza A viruses that currently circulate every winter. Nobody has any real idea what the population burden of influenza will be for the next couple of years, but it is likely to be much greater than average.

Will it get bad enough to overwhelm healthcare facilities? Nobody knows. But hospitals in cities like Rochester (New York) and Winnipeg (Manitoba) have already reported being overwhelmed by the comparatively few severe swine flu cases they have faced so far.

Will it get bad enough to shut down factories, schools, and essential services due to absenteeism? Again, nobody knows. But some local ambulance companies have already reported limited service because of swine flu absenteeism. And today’s just-in-time supply chains are fragile. If too many coal miners or railroad employees call in sick with the swine flu at the same time, coal-fired power plants could run out of fuel and major cities could go dark. If the factories in developing countries that manufacture surgical masks and generic drugs can’t operate or can’t ship, U.S. hospitals and pharmacies could find themselves with empty shelves. These are just “tip of the iceberg” examples of possible supply chain problems in our globalized just-in-time economy.

It’s hard to find this kind of information in the mainstream media. Official sources aren’t stressing it, and most journalists aren’t looking for it. (Readers who think the media invariably sensationalize risk might pause to consider this.)

There are exceptions. On June 22, a New Zealand Ministry of Health official speculated that up to half the country’s population could be infected over the next few months. The same day, an official from the Philippines Department of Health speculated that 90 million Filipinos  – 25 percent of the population – might get swine flu over an unknown time frame.

These are two rare examples of officials quoted in the mainstream media of their countries, trying to tell people what to expect during the pandemic of 2009, based on knowledge from the pandemics of 1918, 1957, and 1968.  Few other officials have done likewise, and few other mainstream media have picked up these sorts of numbers.

How many deaths do these sorts of numbers imply? Let’s speculate that the current virus will turn out roughly as virulent as the seasonal flu – no worse, but no better. Mortality from the seasonal flu is thought to be around 0.1 percent, one death for every thousand cases.

So if pandemic H1N1 ends up infecting one-third of the world’s 6.7 billion people and killing 0.1 percent of those it infects, we can expect 2.2 million swine flu deaths in all, worldwide. WHO has recorded 263 of those deaths so far.

Based on the same assumptions, the U.S. can expect 100,000 of its 300 million people to die from pandemic H1N1. That’s about three times as many as die from the seasonal flu every year – and it will be mostly young people, not mostly the elderly. The CDC has recorded 127 of those 100,000 deaths so far.

That’s what we probably face if swine flu remains “mild.” Swine flu risk communicators should be helping people visualize these things, vividly, so we can all start preparing – emotionally, cognitively, and logistically.

Why have we included a section on the likely pervasiveness of a swine flu pandemic in a column on containment as signal? Because what the containment strategy signals most fundamentally is that swine flu will not be pervasive, that governments will manage to keep it from ever becoming pervasive. That misleading signal greatly impedes the world’s progress in coming to grips with our first influenza pandemic of the 21st century.

Commentators who scoff at precautions because there have been only 263 worldwide deaths and 127 U.S. deaths so far haven’t done the appropriate mathematical extrapolation. But governments that allow their people to imagine that getting swine flu is tantamount to a death sentence, that successful containment is therefore an urgent necessity, and that successful containment is an achievable goal are even further off the mark. This is the perception of people in many countries around world – largely due to misleading signals, in the absence of countermanding words, from their leaders.

In sum

It is hard for us to escape the conclusion that many governments are choosing between two dominant strategies of talking about swine flu. Either they stress that it’s mild, or they stress that they’re going to stop it. And when the latter governments fail to stop it – when ongoing community transmission becomes obvious – they belatedly switch to stressing that it’s mild.

The U.S. (and Canada and Mexico) didn’t really have a choice. Since pandemic H1N1 got its start in North America, and was fairly widespread by the time it was identified, U.S. health authorities mostly did not try to keep it out or pretend they were trying to keep it out.

When it looked like it was severe, they candidly told us so. On April 24, when the news broke that the alarming Mexican disease outbreaks were almost certainly connected to the eight cases of novel swine flu in the U.S., Acting U.S. CDC Director Richard Besser acknowledged the grounds for alarm:

First I want to recognize that people are concerned about this situation.  We hear from the public and from others about their concern, and we are worried, as well. Our concern has grown since yesterday in light of what we’ve learned since then.

This statement was dramatically different from hundreds of later statements by officials around the world, telling people not to panic, not to be alarmed, not to worry, and sometimes not even to be concerned.

As soon as the initial scary Mexican data turned out wrong, U.S. officials focused on teaching people that swine flu was turning out mostly mild so far. That didn’t keep some local communities from pursuing a policy of faux containment in order to reassure their citizens. Still, the dominant U.S. swine flu public health story wasn’t about containment – and so the dominant U.S. swine flu risk communication story isn’t about containment as signal.

Elsewhere in the world, initial containment efforts made public health sense, and so containment as signal was an available risk communication option – unwise, but available. Many governments chose it. As a result, many of the world’s people now think swine flu is very severe, and are counting on their governments to contain it. And many of the world’s governments are reluctant to move on to mitigation after containment has run its course, because they have set up the failure of containment, which was inevitable from the outset, to look like it’s their fault.

Swine flu really is pretty mild so far – that is to say, it appears to be a serious disease much like the seasonal flu, except that it mostly spares people over 50 and children under 2. Reassurance that is grounded in swine flu’s overall mildness is justified. Of course swine flu may or may not stay mild. So it’s important for people to understand that “mild” doesn’t signal “permanently mild” – and to be prepared, logistically and emotionally, for the possibility of an increase in virulence at any time. And because pandemics are unstoppable, even a mild pandemic can mean billions of illnesses and millions of deaths. So it’s important for people to understand that “mild” doesn’t signal “trivial” – and to know what precautions they ought to be taking to reduce their chances of getting (and spreading) the disease.

Still, mildness is a good reason for reassurance, and talking about mildness is one good focus for swine flu risk communication.

Containment is not a good reason for reassurance, and the signals sent by containment constitute misleading swine flu risk communication.

Containment is thought to be a good short-term public health strategy, to delay the local spread of the disease in order to buy time for preparedness and public education. But unless governments work hard to explain what the containment strategy does and does not mean, containment sends the signal that swine flu can be stopped (or why would we be trying?) and that swine flu must be stopped (because it’s so severe). And once people believe, mistakenly, that swine flu is severe but stoppable, they understandably put pressure on their governments to persevere in their containment efforts.

Those efforts are worse than ineffective; they are harmful, especially because they waste and misdirect scarce public health resources, and because they do not prepare the public for what is coming. And when containment fails, as it ultimately must, the outcomes include a public that is more mistrustful, more alarmed, and less prepared than it would otherwise have been, if its leaders had countermanded the containment signal with candid words.

Copyright © 2009 by Peter M. Sandman and Jody Lanard

For more on crisis communication:    link to Crisis Communication index
For more on infectious diseases risk communication:    link to Pandemic and Other Infectious Diseases index
      Comment or Ask      Read the comments
Contact information page:    Peter M. Sandman

Website design and management provided by SnowTao Editing Services.