Search Peter’s website:

U.S. Pandemic Vaccine Supply and Distribution: Addressing the Outrage

Who’d have thought, six months into the first flu pandemic in decades, that a major focus of public and media attention in the United States would be … not the pandemic, not the pandemic vaccine, not even the pandemic vaccine supply, but the vagaries of pandemic vaccine supply distribution?

Talk about outrage trumping hazard!

Wall Street firms Goldman Sachs and Citibank, for example, have been excoriated for the entirely benign behavior of ordering some vaccine and getting their orders duly filled (partly filled, actually) by the New York City Department of Health and Mental Hygiene.

Workplaces are one of the most efficient sites for vaccination programs, because (unlike clinics and doctors’ offices) people go there every day anyway. All Goldman Sachs and Citibank did was what lots of other big employers with their own medical staffs did, and what public health authorities had urged them to do – make plans to vaccinate their high-risk employees in-house rather than tell them to go see their own doctors. (Wall Street, too, employs pregnant women, people with asthma, and others in high-risk categories.)

But the public wants to give merchant bankers the flu, not protect them from it. “Vaccines should go to people who need them most, not people who happen to work on Wall Street,” announced Senator Chris Dodd of Connecticut. And Service Employees International Union official John VanDeventer wrote: “Wall Street banks have already taken so much from us. They’ve taken trillions of our tax dollars. They’ve taken away people’s homes who are struggling to pay the bills. But they should not be allowed to take away our health and well-being.”

The outrage was so mammoth it prompted Director Thomas Frieden of the U.S. Centers for Disease Control and Prevention (CDC) to send out an email reminding state and local health departments that the pandemic vaccine should be used only for people in prioritized categories. “Any vaccine distribution decisions that appear to direct vaccine to people outside the identified priority groups have the potential to undermine the credibility of the program,” the email pointedly noted. The operative word here is “appear.” Like other employers, Goldman Sachs and Citibank ordered only enough doses for their high-risk employees; like other employers (and hospitals, clinics, and doctors’ offices), Goldman Sachs and Citibank got less than they ordered.

Demands were floated for the banking companies to donate their doses to a community hospital or clinic, where they could go into arms of people not associated with economic collapse … not even as a secretary or janitor for such a stigmatized employer. (New York financial services giant Morgan Stanley received 1000 doses, but passed them on to hospitals in the city.)

Other recent vaccine distribution controversies:

• The White House denied reports that detainees at Guantanamo were to be vaccinated. “There is no vaccine in Guantanamo and there’s no vaccine on the way to Guantanamo,” White House spokesman Robert Gibbs told reporters. But Pentagon sources said some Guantanamo prisoners were in fact slated for vaccination as part of the military’s base protection program. A week later Guantanamo confirmed receipt of a vaccine shipment, but officials said terror detainees weren’t going to get any for now. Meanwhile, the Christian Science Monitor explored the broader issue of where prison inmates should stand in the vaccination pecking order.
• President Obama’s two daughters made news in early October for not getting vaccinated against the pandemic virus. Later in the month, as the vaccine shortage deepened, the girls were vaccinated after all. Bloggers treated both decisions as controversial, the first because it supposedly signaled that the vaccine might be dangerous and the second because millions of less well-connected children were going without. Strangely uncontroversial was the President’s decision not to get vaccinated himself; presumably he realized that the public would rather see the Leader of the Free World down sick with the flu for a week (or worse) than see him jump the line.
• All over the country, vaccination programs focused on the target groups the CDC had prioritized. But how aggressively to police the targeting was a contentious issue. If you’re vaccinating students, should you let the teachers get vaccinated as well? What about the spouse of a high-risk vaccinee who came along to the clinic and wants a shot too? Especially in venues where demand was light, many providers chose to get all the available vaccine into somebody’s arm ASAP, so as to qualify for another shipment. The result was endless media “investigative reports” about rumors that unprioritized people were getting vaccinated at Site X ahead of prioritized people at Site Y. In at least one state, Hawaii, state health officials went undercover to root out providers who might be vaccinating some low-priority people.
• It isn’t just the U.S. that has become obsessed with pandemic vaccine supply distribution. In Canada, two health care workers were fired for letting Calgary Flames hockey players and their families jump the line. Meanwhile, Canadian plans to vaccinate suspected Taliban members imprisoned in Afghanistan also came under attack. In Canada, as in the U.S., vaccine fairness meant that nobody (not even hockey players and presidents) should be promoted ahead of others who were at higher risk. But it also seemed to mean that some (like bankers and suspected terrorists) should be demoted regardless of their risk.
• In Germany, scandal erupted because the government had bought 25 million doses of one pandemic vaccine for the general public, and 200,000 doses of a different vaccine for government officials and soldiers. Both vaccines are approved in the EU and both are widely used – but the one for the general public contains a controversial adjuvant, provoking widespread charges of two classes of health care. (The controversy was reminiscent of a similar battle in the U.S. during the anthrax attacks of 2001, when exposed postal workers were prescribed doxycycline after members of Congress had been given ciprofloxacin.)

Complaints about inequity and line-jumping are the tip of the iceberg. Even more common are complaints about chaos and confusion – endless feature stories about people flooding their doctors’ phone systems and standing in line for hours in a frequently futile effort to get vaccinated. And coming soon, we would guess: complaints about maldistribution, about why there’s plenty of excess vaccine there but not nearly enough here.

It would be hyperbolic to claim that people are more concerned about pandemic vaccine distribution than about the pandemic itself. But recent news coverage has sometimes given that impression. And some people who aren’t especially worried about the pandemic are nonetheless greatly offended that vaccine is available to others but not to them.

Our son says that most of his coworkers are in that category:

At work, the conversation periodically turns to H1N1. I mostly listen. People are surprised that there are not vaccines available, if not upset. When asked if they’ll seek the vaccine, most agree they will not. This is the reality TV syndrome: people being upset and dramatic about realities that don’t impact their lives in any significant way. “How dare you break your promise; if you had kept it, I would not have cashed in.”

There are two risk communication issues worth addressing here: outrage about how the vaccine supply is distributed (inequity, confusion, maldistribution) and outrage about the inadequacy of that supply. The two are closely related. The vaccine shortage is what makes vaccine distribution so difficult, controversial, and outrage-provoking. But they’re worth discussing separately.

Segmenting the Vaccine Market

First, though, it’s useful to segment the U.S. public into nine groups with regard to pandemic flu vaccination:

In the U.S., no pandemic risk communication task is more important than convincing Groups 1, 2, and 3 to get vaccinated as soon as vaccine is available in their community.

This is actually three quite different tasks. For Group 1 (not worried about the disease), it’s traditional precaution advocacy: motivating insufficiently concerned people to worry more about pandemic risk. For Group 2 (worried about the vaccine), it’s traditional outrage management: motivating excessively concerned people to worry less about vaccine risk. For Group 3 (mildly interested but easily deterred or distracted), it’s about overcoming barriers – making implementation easier, reminding people of their intention to get vaccinated, etc.

Convincing Groups 4 and 5 (lower-priority groups) to get vaccinated is on a back burner until there’s enough vaccine. But it will eventually be important to target them too, since about one-third of the people who end up with severe cases of this flu have “no underlying conditions.”

When we say that “no pandemic risk communication task is more important” than convincing people in prioritized groups to get vaccinated, we don’t mean to imply that a hard sell is the way to go about that task. We have already written about our reasons for preferring a soft sell, for advising public health officials to “offer” pandemic vaccine, not to “push” it. It would take a more severe pandemic than the one we face so far to make us change our minds about that.

Nonetheless, convincing people in high-priority groups to get vaccinated is a crucial pandemic risk communication task. And the furor over the inadequate vaccine supply and how that supply is being distributed – the outrage of Group 8 – is in some ways a distraction from that task.

The furor isn’t all bad. The human race is ornery; we tend to want what we can’t have. So the inadequacy of the vaccine supply is one important driver of vaccine demand. We will probably segue virtually overnight from an imbalance of demand over supply to an imbalance of supply over demand, especially if the pandemic ebbs at least temporarily just as we are flooded with vaccine.

People in Group 8 keep telling themselves and their friends how outrageous and scary it is that they can’t find any swine flu vaccine. This is probably better for the ultimate success of vaccination programs than if they were waiting patiently or, worse yet, deciding it’s more hassle than it’s worth, like the people in Group 9. At least some of October and November’s pent-up demand for vaccine will produce eager vaccinees in January and February.

A major pandemic risk communication goal, then, is to help sustain vaccine demand until people can get vaccinated – in other words, to keep as many people as possible in Group 8 from giving up on getting vaccinated and thus moving into Group 9. So addressing Group 8’s outrage isn’t just a distraction. It is a key to holding down the size of Group 9.

For health officials, responding to outrage about vaccine supply and vaccine distribution is an unfortunate – but unavoidable – distraction from building the case for vaccination itself. Outrage is like that: It distracts people from hazard. There’s little value in bemoaning this reality.

For the moment, one of the main jobs of pandemic risk communication is to manage the outrage about the inadequate vaccine supply and how it’s being distributed, in order to keep people in Group 8 – especially those in high-risk categories – from moving to Group 9 and deciding they don’t want the vaccine after all.

Vaccine Supply Outrage: How the Shortage Happened

As we write this in mid-November, there isn’t enough pandemic vaccine yet in the U.S. – not enough for everyone who wants it, not enough for everyone the experts consider at elevated risk, and certainly not enough for everyone in the country.

Most other developed countries in the northern hemisphere are in roughly the same boat; they’re getting some vaccine, but it’s dribbling in later and slower than they’d hoped. Developed countries in the southern hemisphere had to get through their winter pandemic flu season without any vaccine; now they’ve got some, but not much flu at the moment, and therefore not much demand for the vaccine.

The developing world, of course, never expected to have much pandemic vaccine – although many governments in the developing world are unwisely signaling to their publics that they’ll get an adequate supply “soon.” Developing countries are dependent mostly on vaccine contributions from the developed countries, contributions that are likely to be minimal as long as the shortage continues. In some ways (though obviously not in terms of reducing influenza mortality and morbidity) developing countries may be better off with no vaccine than with just a little. Officials in some of these countries worry about the destabilizing outrage likely to result when they announce who will receive the very small quantity of pandemic vaccine they’ll have to distribute.

The main reason the U.S. doesn’t have enough vaccine yet is that the pandemic vaccine has turned out to be more difficult to produce than officials had hoped (and strongly signaled).

Influenza vaccine manufacturing hasn’t changed much since the 1950s. Vaccine is grown in fertilized chicken eggs, a process that is slow at best, and very prone to glitches and delays. There are alternatives: The European Union, but not yet the U.S., has approved a Novartis vaccine that is cultured in vats of animal cells instead of individual chicken eggs. In normal times, manufacturers have at least seven months to produce the two seasonal flu vaccines, one for the northern hemisphere and one for the southern hemisphere. In normal times, they have a pretty good idea what next season’s demand will be, based on the previous season’s vaccine uptake. And even though manufacturing capacity has been ramped up enormously in recent years – spurred mostly by bird flu concerns – there still isn’t enough manufacturing capacity to meet the world’s needs. There’s just about enough to meet the actual annual demand in the developed world – as long as nothing happens to increase the demand or impede the supply.

Now take this stretched-to-the-limit vaccine manufacturing capacity and add a pandemic. Not only have manufacturers tried to adhere to a pandemic flu vaccine manufacturing schedule that’s shorter than their seasonal flu vaccine manufacturing schedule. They’ve also tried to keep the seasonal schedule from falling apart as a result. And, of course, they’ve tried to do both without cutting any safety corners.

As if that weren’t enough, the U.S. decided to order vaccine without adjuvant. Adjuvants are chemicals added to vaccines to increase their effectiveness, and thus decrease the amount of vaccine needed per dose. Most of the world (that is, most of the developed world) ordered pandemic vaccine with adjuvant. The World Health Organization (WHO) recommended adjuvanted pandemic influenza vaccine, so that more people could be vaccinated sooner.

But adjuvants are controversial. And no adjuvanted flu vaccine has been approved by the U.S. Food and Drug Administration, so an adjuvanted pandemic vaccine would have to be given “experimental” or “emergency” approval – handing vaccination opponents a fat new talking point and contributing to even more vaccine mistrust than already exists. Rather than risk increased vaccination skepticism, the U.S. ordered unadjuvanted vaccine – which meant it needed more vaccine per dose. (The U.S. also ordered bulk adjuvant, to be used only if the pandemic turned more virulent.)

To improve the odds of getting as much vaccine as quickly as possible, the U.S. placed orders with five different manufacturers. But all the vaccine manufacturers were working from the same vaccine seed stock, grown from the virus that infected some of last spring’s earliest swine flu victims. Except for one company, MedImmune, they all found the vaccine tough to grow. So pandemic vaccine production fell far behind the projected, estimated, wished-for, but always tentative “schedule” that had been communicated to the public.

It was clear early on that the pandemic vaccine virus was a slow-grower. There have been news stories about the problem for months; here’s one from early August by the Canadian Press’s Helen Branswell, the dean of pandemic flu reporters.

As far as we can tell, nobody is to blame for the fact that production fell behind schedule. As far as we can tell, in fact, it was always likelier than not that production would fall behind schedule. The “schedule” was an aspirational goal. Everything would have had to fall into place for the schedule to be met – and that rarely happens even with seasonal flu vaccine production.

A lot of things did fall into place. For awhile it looked like every vaccinee would need two doses; we caught a break when it turned out that a single dose would probably be immunogenic enough for everybody except young children. But a few things didn’t fall into place – most notably the troubles growing the vaccine virus. This was a big disappointment, but it shouldn’t have been a big surprise.

But it was a big surprise to the general public, possibly to most local and state health departments, and definitely to frontline health care providers – despite their previous experience with delayed seasonal flu vaccine. Nearly everyone felt that officials had promised that by the end of October the U.S. would be swimming in vaccine. So by the end of October we were swimming in outrage instead.

Vaccine Supply Outrage: How Officials Overpromised

The big error, then, was overpromising. Not literally. U.S. public health officials rarely if ever “promised” enough vaccine by the end of October. But they certainly created that impression … or at least allowed that impression to circulate unrebutted.

This violates three fundamental principles of crisis communication: Don’t over-reassure ; don’t be overconfident (especially about reassuring information); and correct misimpressions .

Here’s the sound bite we should have been hearing repeatedly and consistently from early May 2009 onward:

With luck, we hope to see the first pandemic vaccine doses rolling off the line sometime in October. But vaccine manufacturing is finicky and delays are likely. The sad truth is, it could easily be early 2010, or even much later, before we have enough vaccine for everyone in the U.S. who wants it. For now and an unknown number of months to come, we will have to face this pandemic without a vaccine.

Yes, we know: Reporters prefer clear, unnuanced estimates to ones that are riddled with qualifiers. Unless sources are insistent and skillful, the qualifiers don’t necessarily make it into the story. It’s not enough to say “We hope to have this much” instead of “We’re confident we will have this much.” You have to say, explicitly, every time: “We’re aiming for this much, but we have real doubts about our chances of success. We could easily end up with less, or even none.”

Now we are going to document some of the overconfident, over-reassuring official messaging we have seen about pandemic vaccine availability. We may give more examples than you think necessary. We’re belaboring the point because most officials who made this error do not feel that they did so. They feel they never “promised” vaccine, and that makes it harder for them to do good outrage management about what came across as overpromising to the public.

We hope that if officials can accept that they overpromised – or at least that that’s how they were heard – they will be better able to be honest and empathic in addressing the fully justified public outrage they are confronting today.

The expectation that there would be plenty of vaccine within six months after the appearance of a pandemic strain actually predates the discovery of novel H1N1 in late April 2009. Back when bird flu, not swine flu, was the main focus of pandemic concern, officials kept “warning” that there wouldn’t be much vaccine for three to six months. This over-optimistic prediction (a promise masquerading as a warning) was built into official pandemic planning.

The single most over-optimistic prediction, masquerading as a warning, was published by the U.S. Department of Health and Human Services in early 2006, as part of a pre-pandemic “message map” :

Vaccine might be in short supply in the early stages of an influenza pandemic.

Might be in short supply?

In scores of avian and pandemic influenza communication trainings, consultations, and workshops, we have cited this as an egregious example of disingenuous refusal to acknowledge the most basic fact about pandemic influenza vaccine: Based on current technology, you cannot start to produce a vaccine until after you know what influenza strain is causing the pandemic.

Most other overpromises have at least suggested some time lag between the start of the pandemic and the start of vaccine manufacturing.

Delaware’s Department of Health and Social Services, for example, couldn’t have been clearer in its “Delaware Pandemic Influenza Plan” :

When a Pandemic Influenza A virus is identified, it will likely take between three to six months to produce and deliver sufficient vaccine to inoculate the entire U.S. population.

The “Texas Department of Health Pandemic Influenza Plan” makes the same explicit assumption:

It will take three to six months after the novel virus is identified before the vaccine is available for distribution, unless a DNA vaccine is developed and deemed safe and necessary.

Not everyone was so optimistic. Houston’s health department took a more cautionary view: 

The time from a candidate vaccine strain to the production of the first vaccine dosage could be three to six months or more…. Once vaccine is available, it may take several months [more] to produce an adequate supply of vaccine for the entire U.S. population.

Over-optimism wasn’t confined to the U.S. The Singapore government usually does excellent risk communication, and rarely over-reassures its population. But an October 2008 news release was an exception:

Singapore has placed a deposit to ensure it is among the first to get any pandemic vaccine – no matter what strain of flu virus causes it.

But manufacturers will need three to six months to make the vaccines after a pandemic is declared.

We initially fell for the “three to six months” prediction too. In a July 2005 column entitled “Superb Flu Pandemic Risk Communication: A Role Model from Australia,” we criticized an Australian news release about the country’s pandemic vaccination planning for failing to acknowledge that “most experts believe it will take at least three to six months to produce such a vaccine once a pandemic strain of influenza is identified.” Actually, there were plenty of experts who thought three to six months might be enough time to have a vaccine, but not enough time to have a lot of it. But the most widely quoted experts were implying ample vaccine (for developed countries with money to buy it) in three to six months.

So when swine flu appeared in late April of 2009, “three to six months” was already the meme. Here’s an example – one of hundreds – from John Lauerman’s April 30 Bloomberg story:

Vaccine makers have contacted the World Health Organization about obtaining samples of the virus needed to make a vaccine. Making flu vaccine can take three to six months.

Another sign of over-optimism: hundreds – perhaps thousands – of local health department drills to practice mass pandemic vaccination. Each drill was invariably covered by local media, with health officials burbling on camera about how prepared they had become to inoculate huge numbers of people in a matter of hours. For years we made a hobby of scanning these stories to see two things:

1. Did the stories mention that the health department in question had no vaccine with which to inoculate those huge numbers of people … and no real idea when sufficient vaccine might be forthcoming? This relevant problem – nothing to put into the syringes – was rarely mentioned.
2. Did the health departments also drill a scenario in which huge numbers of frightened people descend on the mass vaccination site, but the vaccinators have only a small amount of vaccine? Nope. Never. There was always plenty of vaccine in these drills, and so large numbers of people were quickly vaccinated and went home happy.

Given the tendency of local health officials to accent the positive, the tendency of journalists to strip off qualifiers, and the already well-entrenched “three to six months” meme, the CDC and other national health agencies tried – and failed – to encourage a more realistic assessment of the vaccine production schedule.

Here for example is Anne Schuchat, speaking at the CDC’s press briefing on June 11, the day WHO announced that a flu pandemic had begun.

The second question was about whether we have a vaccine. That’s a really good question. You know, we have mechanisms in place to develop a vaccine and to test it and to study it, and we’re going to need to look at the results. We’re going to need to study the results. We cannot assume that everything is going to go perfectly. There may be some bumps in the roads, there may be some really big bumps in the road. I think we need to be prepared for the possibility that a good vaccine may not be created or that we may not have it in sufficient time before we have a lot of disease….

I think we have to be ready for the idea that we may not get a vaccine as soon as we’d like it, or we may not get a vaccine that works as well as we would like it, or we might not even get a vaccine. We’re really taking all the steps we can to make sure that we have one if we need it. But I think there is uncertainty in that just like there is uncertainty in the nature of the influenza virus.

But on the same day at the same press briefing, CDC Director Thomas Frieden sounded much more optimistic:

We can expect to see continued efforts to develop a vaccine and we hope and anticipate that that may be in place by the fall.

Here are several relevant comments on vaccine availability expectations, all from Anne Schuchat, all from the July 24 CDC press briefing:

At this point the secretary [Kathleen Sebelius] has announced that we are planning for a voluntary vaccination program in the fall, assuming availability of appropriate vaccines and that the virus hasn’t changed so substantially that a vaccine wouldn’t work….

What I can say, though, is that vaccination is one of the best ways to prevent influenza and its complications. That’s why we vaccinate intensively for seasonal influenza and why we’re working on having a vaccine available in the fall for this new virus….

Production is also going forward right now. The U.S. has procured vaccines from five companies, and those companies are all taking the steps to make large amounts of vaccine available…. At this point, the U.S. government has procured large amounts of vaccine but we haven’t yet made recommendations on what populations ought to be offered vaccine…. [Dr. Schuchat probably didn’t realize that most people would hear the word “procured” as suggesting a done deal, not just a bunch of government procurement contracts with uncertain delivery schedules.]

At this point the planning and investments that the U.S. government made suggests to us that we are likely to have plenty of vaccine for the groups that are targeted. Of course it is always risky to say that because influenza vaccine manufacturing is not always as predictable as you would like. And sometimes we have surprises. But at this point we’re expecting there to be a reasonably large numbers of doses available and the middle of October is the point that we’re looking at in terms of our planning, that we hope to be able to launch a vaccine program, assuming several factors in terms of safe and effective vaccine is available and no big change in the antigen properties, we’re planning the middle of October timeline. The exact number of doses that we’ll have, whether everything will be ready to go, those are things that we really have to be prepared for some surprises around.

Other U.S. government officials were more willing to guess at the numbers. For example, on July 23, the day before the Schuchat briefing above, Reuters ran this story:

GAITHERSBURG, Maryland (Reuters) – The U.S. government has bought 195 million doses of H1N1 swine flu vaccine for a possible autumn vaccination campaign, a U.S. federal official said Thursday. The U.S. Health and Human Services Department has also contracted for 120 million doses of adjuvant, a compound to stretch the number of doses of vaccine needed, the department’s Dr. Robin Robinson told a meeting of Food and Drug Administration advisers….

Earlier, an FDA official said vaccine makers were only getting about 30 percent as much vaccine from eggs as they usually get with seasonal influenza vaccine. Robinson said HHS had accounted for this in planning for a possible influenza pandemic….

He said HHS had planned to have 160 million doses available right away, and then 80 million a month afterward.

The numbers being thrown around a month later were a lot lower. At the CDC’s August 21 press briefing, Jay Butler said: “We’re expecting somewhere between 45 million and 52 million doses of vaccine to be available by mid-October. This will be followed by weekly availability of vaccine up to about 195 million doses by the end of the year. Keep in mind, these numbers are driven by a number of variables in the manufacturing process … and so those numbers can be subject to change.”

A month after that, at the September 18 CDC press briefing, Jay Butler was apparently trying to lower expectations still further, but without saying so explicitly:

So this has been able to be developed very rapidly, and we anticipate that it will be available during October. We actually anticipate being able to start receiving orders for the vaccine in – by early October. And actually, vaccine going out and being distributed by providers by the first week of October.… Initially, we anticipate that about 3.4 million doses of vaccine will be available….

Not everyone in the U.S. government thought lowering expectations was necessary. On September 15, Secretary of Health Kathleen Sebelius told a Congressional committee :

The federal government is also preparing for a voluntary national vaccination campaign for the 2009 H1N1 virus starting in October. With unprecedented speed, we have completed key steps in the vaccine development process – we have characterized the virus, identified a candidate strain, expedited manufacturing, and performed clinical trials. The speed of this vaccine development was possible due to the investments made through ASPR/BARDA over the past six years in advanced research and development and infrastructure building. One-hundred ninety-five (195) million doses of H1N1 vaccine have been purchased from five manufacturers by the U.S. government….

The large scale 2009 H1N1 vaccine program will begin mid-October with small amounts of vaccine becoming available the first full week in October.

Does that sound to you like Secretary Sebelius was confident in mid-September that HHS could launch a smooth 195-million-dose rollout in mid-October? It does to us.

On October 23, the Center for Infectious Disease Research and Policy summarized the course of events this way:

The flow of vaccine to US providers has been far below earlier projections. At a news briefing today, Dr. Tom Frieden, director of the Centers for Disease Control and Prevention (CDC), said a total of 16.1 million doses have become available for distribution so far. Yesterday Dr. Anne Schuchat of the CDC predicted that the number will reach 28 million doses by the end of this month.

Back in July, HHS had projected that 120 million doses of vaccine would be ready in October. But in mid-August that was trimmed to a predicted 45 million doses by mid-October, after manufacturers found that the egg-based vaccine production process was not yielding nearly as much virus as expected.

It doesn’t really matter whether U.S. public health officials actually “promised” there would be ample vaccine by the end of October, or implied it, or merely allowed others to surmise it without working to correct the record. Nor does it matter whether officials were trying to reassure the public or were genuinely overconfident themselves.

What matters is that most Americans came to expect there would be ample vaccine by the end of October. And there wasn’t. When the audience ends up with a false expectation, by definition the communicator did something wrong.

The U.S. government has had two core messages about pandemic vaccination:

1. People should get the best information, and then decide for themselves whether to get vaccinated. We hope you will take our advice to get the vaccine as soon as you can.
2. We expect to have very large amounts of vaccine by fall.

The first message was right on target. The public has been urged but not pressured to get vaccinated as soon as possible, especially those in high-risk groups.

The second message was overconfident and over-optimistic.

Vaccine Supply Outrage: Mishandling the Apology

Inevitably, some who were misled ended up criticizing officials for under-delivering (which isn’t really their fault) rather than for overpromising (which is). But most critics – even politicians – actually seemed to get the distinction. Consider the lede of this October 28 story in the Wall Street Journal:

Two top senators overseeing the national response to the swine flu are sharply questioning the government’s handling of the vaccination program, in one of the first indications that it could become a political issue for the Obama administration.

In a letter sent Tuesday to Kathleen Sebelius, secretary of Health and Human Services, Sens. Joseph Lieberman (I., Conn.) and Susan Collins (R., Maine) praised the government’s initial steps on the H1N1 swine flu, but said they now have “strong concerns” about the fallout from faulty estimates for how much vaccine would be available.

“Unfortunately, these missteps in estimating available doses of H1N1 vaccine have effects beyond just growing public frustration; they have the potential to critically undermine our vaccine distribution efforts, which depend on accurate estimates of vaccine availability,” the senators wrote.

The remedy for having been overconfidently over-reassuring is to apologize – early and often.

Maybe that’s what White House Senior Advisor David Axelrod was trying to do on November 2. Here’s the relevant portion of Axelrod’s interview with NPR’s Scott Simon:

Scott Simon: On Friday, the president talked about his frustration that H1N1 vaccine hasn’t gotten out to more Americans. In August, the Centers for Disease Control said that 120 million doses would be available. They later scaled that back to 45 million. We’re speaking today, on the last day of October, 25 million doses reportedly are ready. Did the government overpromise?

David Axelrod: Well, I think the manufacturers overpromised, and what was reported was the representations that were made to us. The fact is that this is a problem that’s abating every day. And yes, we thought we would have 40,000 now, we have 26 —

Simon: — Forty million —

Axelrod: — I’m sorry, 40 million. Now we have 26 million. We believe that that is improving on a daily basis, and we’re going to have an ample supply in very short order. So yes, we probably did overpromise, and we overpromised on the basis of what was represented to us.

Maybe that sounds like an apology to you. To us, it sounds like scapegoating the pharmaceutical industry. Not to mention that Axelrod is overpromising once again: to have an ample supply “in very short order.”

Here’s a similar non-apology apology, this one from HHS Secretary Sebelius in a November 11 ABC News Report: “We were relying on the manufacturers to give us their numbers, and as soon as we got numbers, we put them out to the public. It does appear now that those numbers were overly rosy.”

Managing outrage about the inadequate pandemic vaccine supply mostly means that CDC and HHS officials should repeatedly:

• acknowledge that they gave the public and providers a misimpression about vaccine availability;
• apologize for having done so; and
• validate and empathize with the public’s and providers’ entirely justified frustration and anger about the complex, confusing, ever-changing, and unpredictable vaccine distribution system.

The key words above are: acknowledge, apologize, validate, empathize … and repeatedly.

This is not easy for officials to do – especially “repeatedly.” Many of them don’t feel that they overpromised about the timing of vaccine availability, even though that is the strong impression received by the public. They can point to numerous times when they talked, sometimes at length, about the unpredictable vicissitudes of flu vaccine production.

In addition, some of them are probably angry at themselves for uncritically passing on optimistic pharmaceutical company predictions. In hindsight they may even feel a bit foolish about doing so. They were monitoring production problems closely; they had decades of institutional experience with seasonal vaccine production problems; they thus had reason to know the predictions were optimistic. In many cases they have been projecting their anger at themselves onto the pharmaceutical companies, and doing so in full view of the media – thereby risking additional public mistrust of Big Pharma, inevitably including mistrust of Big Pharma’s assertions about vaccine safety.

And some officials are undoubtedly angry at the media for harping on the delays. And at the public – especially Group 8 – for not taking the delays in stride. And at politicians for making political hay out of the problem. (If the Obama administration can’t manage a vaccine rollout, some have said, how can it possibly manage health care reform?) And at all of us for failing to recognize what an incredible achievement it is to have any vaccine a short six months after the discovery of a new influenza strain.

In the three paragraphs above, we are trying to show our own empathy with senior officials. We are using the risk communication technique of “telling people stories about themselves.”

This is a technique officials need to use far more than they do now when talking with angry stakeholders. It’s as important as acknowledging that they overpromised, and as important as apologizing (without scapegoating Big Pharma) for having done so. It’s as important as making sure they don’t continue to overpromise, but instead offering intentionally conservative estimates of each week’s additional vaccine supply so we can be happily rather than unhappily surprised for a change.

We will all calm down a lot quicker – politicians, media, providers, the general public, and Group 8 – if officials unearth and retell the most vivid stories they can find about hellish lines, Kafkaesque runarounds, and the rest. Extra credit if they abashedly recount their own earlier over-optimistic predictions as well.

There is, of course, a risk communication seesaw in play here. With rare exceptions, nobody has suffered unduly because officials were too optimistic. The vaccine shortage itself is another matter. People have died, and will die, because there wasn’t enough vaccine soon enough. But all that went wrong because the CDC and HHS overpromised is that people endured more hassle than they should have had to endure.

The CDC and HHS shouldn’t be the ones to say that their overpromising led to nothing more serious than hassle. (And of course officials should never, ever gloss over the fact that vaccine delay has resulted in more deaths.) But that’s something the rest of us can say, to ourselves if not to the CDC and HHS, if officials sound sufficiently impressed with our frustration and anger. This is the seesaw: If officials say it was merely a hassle, we’re going to feel like they just don’t understand how awful it was. If they tell us stories about how awful it was, we’re likelier to remember that it was merely a hassle. And then, ideally, more of our outrage will focus where it belongs, and where it can do the most good: on the need to make the modernization of vaccine production a priority. (Of course we’ll also still remember that the CDC and HHS overpromised.)

Above all, CDC and HHS are wrong to deny that they gave the public its inaccurate impression that there would be plenty of vaccine by mid-October. Those two agencies have some of the most sophisticated and rapid systems in the world for monitoring media content and public opinion. For months they have been tracking what the media are saying and the public is thinking about pandemic vaccine. They knew or should have known that we were expecting more than they could be confident of delivering.

The outrage of Group 8 – and the rest of us – about the surprising unavailability of pandemic flu vaccine is justified. And managing justified outrage appropriately is important, before officials can return to the paramount task of convincing Groups 1, 2, and 3 (and even Groups 4 and 5, and perhaps even Group 9) to get vaccinated as soon as there’s enough vaccine.

Vaccine Distribution Outrage

As we noted earlier, there are at least three vaccine distribution issues:

• Unfairness and line-jumping
• Chaos and confusion
• Maldistribution

Officials’ response to outrage about line-jumping has been pretty much to ally with the outrage. Some politicians have more than allied with the outrage; they have provoked it and led it. For the most part, public health officials have expressed tepid agreement that, yes, it’s important to make sure scarce vaccine doses go to members of prioritized groups, with no favoritism for, say, Wall Street bankers.

Fair enough. But here’s some of what they haven’t stressed:

1. Some Wall Street bankers are members of prioritized groups. People shouldn’t be allowed to jump the line because of their employers, but neither should they be relegated to the end of the line because of their employers.
2. There’s no real consensus on how to prioritize among the prioritized groups. Is vaccinating a healthy child more or less important than vaccinating an asthmatic adult? What if the asthmatic adult is in prison?
3. A speedy vaccination program may be just as important as a fair vaccination program. You can’t afford to be paralyzed by searching for the Most Deserving Vaccinee in town before moving on to the second most deserving. Health departments look for vaccination venues where lots of prioritized people congregate naturally: hospitals, schools, and workplaces. The venues with slimmer pickings, like individual doctors’ offices, sometimes lag behind.
4. It’s not easy to decide what to do when unprioritized people show up and want a vaccine dose. Should you turn them away? Even if you have lots of doses, if you won’t get more till you use these up, and if the line is short today?

Of course these statements will backfire if they sound defensive – or, worse yet, if they sound like a counterattack. To avoid sounding that way, risk communication offers a number of strategies. Among them:

Dilemma sharing. People imagine that it’s obvious what fair distribution of vaccine would entail, and they imagine fairness is the key standard – until officials start explaining some of the dilemmas they’re actually facing in deciding what’s fair, and in balancing fairness against efficiency. I’d love to see officials seeking public advice in real time on how best to resolve these dilemmas (which are values questions, not medical questions) – as they did before the pandemic and again last summer before any vaccine was available. At the very least officials should talk about how tough these dilemmas are to resolve.

Empathic identification. It’s easy – and important – to be empathic about people’s anger at genuine line-jumping. But for a lot of prospective vaccinees, the furor over fairness is really about unacknowledged selfishness and unacknowledged disdain: I want to be at the head of the line, and I don’t want certain groups (Wall Street bankers and Guantanamo detainees, for example) to be in the line at all. The tough part of empathy is to find a middle ground between being oblivious about such unacknowledged feelings and being intrusive about them. “A lot of people might feel….” “We in the health department struggle not to give in to feelings like….”

The risk communication seesaw. Whenever people are ambivalent, communication happens on a seesaw: We tend to go to the side of our ambivalence that others are neglecting. So if a health department puts too much stress on efficiency, for example, the ambivalent public will insist all the more strenuously on the primacy of fairness. A more seesaw-savvy way to make the point: “Even though it’s important to get vaccine into people’s bodies as quickly as possible, we may have to slow down a little to make sure we’re making the highest-priority use of each dose.” This formulation could actually lead some to worry that you might paralyze the program with too one-sided a focus on fairness.

In general, we think, outrage about chaos and confusion has been handled better than outrage about unfairness.

Even before it was clear that the vaccine production schedule would be seriously delayed, it was clear that there would be a little vaccine available before there was a lot. There were at least three options for dealing with this inevitability.

• Officials could have let the vaccine pile up unused until there was enough for everybody – or at least until there was enough to avoid the sensation of severe shortage. Canada pursued this option for awhile, until it became clear that the wait wouldn’t be short. From the outset, the U.S. decided that it would be unethical to put outrage management ahead of public health in this way. The goal was to get every dose into somebody’s arm or nose as quickly as it became available.
• Officials could have flooded one state with ample vaccine before moving on to the next. That would be efficient and organized, with no local confusion over where to find vaccine, no battles over why this hospital has some but that clinic doesn’t, no long lines of frustrated people. But it would mean huge national battles over why Alabama has more vaccine than it needs while Arkansas has none at all. As far as we know, the U.S. never seriously considered this option.
• Or officials could do what they decided to do: dribble the vaccine out as quickly as it came in – with states getting a little more every day, in proportion to their populations. This necessarily meant that for awhile every state would have some vaccine but no state would have enough vaccine. And it meant maximum chaos and confusion.

The chaos and confusion could have been minimized if state and local authorities had realized that the vaccine might be in short supply for longer than they were anticipating. This was the most serious harm done by overconfident, over-optimistic promulgation of the “three to six months” meme. Most health departments did good planning for how to get a flood of vaccine into people’s bodies quickly. They didn’t appreciate that they needed to plan also for how to manage a long, slow dribble of vaccine.

If they’d known to plan for the dribble scenario, health officials could have developed better systems for telling people who’s got some vaccine available and who doesn’t. Of course they’d have had to decide whether they really wanted people to know. On the one hand, good information would save the public from wasting time on false leads. On the other hand, good information would enable to public to harass providers all the more. (If you’re a clinic with 200 vaccine doses, do you really want a million vaccine-seekers to know it?) With this in mind, the state of Texas decided not to reveal where it was shipping its vaccine supply.

If they’d known to plan for the dribble scenario, health officials could also have given local providers better advice on how to manage the chaos. Some doctors, for example, complained that their telephone lines were so deluged with vaccine-seekers that “real” patients couldn’t get through – a problem they could have readily solved with modern telecommunications technology. “Press one if you are calling about pandemic flu vaccine.”

We don’t mean to make light of the chaos. Millions of people struggling unsuccessfully to get vaccinated represent a huge waste not just of energy and patience, but also of good will and public health credibility. And in some places health care itself may have been disrupted. On October 29, for example, the Rochester (MN) Post-Bulletin reported that the Mayo Clinic received 200,000 phone calls in a single morning after announcing that it had some vaccine available. A lot of them were undoubtedly people getting busy signals and calling back again and again – but even so, the chaos can’t have been conducive to the Mayo Clinic getting its important work done.

Still, from a public health perspective the distribution system is working pretty well. Most vaccine doses seem to be ending up quickly in appropriate vaccinees. The distribution pattern has certainly been chaotic and therefore frustrating – to providers and patients alike – but it hasn’t actually been inefficient with regard to getting all the available vaccine into eligible bodies without delay.

And there is even a side-benefit to the chaos. People have to work harder than they should to find out where there might be some vaccine available. It takes real effort – tracking down rumors, waiting on hold and in line, putting up with the hassle and the frustration. Some prospective vaccinees give up, and some who give up rationalize that “I didn’t really want that vaccine anyway.” But those who persevere and ultimately prevail in this chaotic competition tend to be the ones who most want to get vaccinated. Instead of making them give up, the competition itself makes them want to get vaccinated all the more, and makes them all the more relieved and triumphant when they finally succeed.

These are the conditions least conducive to imagining harmful side-effects. The chaotic rollout of the pandemic vaccination program, in short, may have diminished the number of local controversies about vaccine safety.

Nonetheless, the chaos has been costly to the public, in terms of both lost time and, more important, lost patience.

One of the best risk communication decisions CDC Director Thomas Frieden made was to warn people about the impending chaos. This is in accord with the crisis communication principle of anticipatory guidance . It is also in stark contrast to the CDC’s failure to warn people adequately about the impending shortage.

The CDC’s preferred metaphor for the chaos was a “bumpy road.” Thus these excerpts from three press briefings in a row:

Frieden on September 25: The choice was between having vaccine stack up in warehouses until we have enough to distribute substantial quantities everywhere or getting it out as soon as it was available to us for distribution.  And we think the only right answer, the only ethical answer is to get it out as rapidly as possible particularly since we have flu spreading widely.  What that means is that in the coming weeks, there is going to be some roughness.  It’s going to be a little bumpy because in different states there will be different levels of preparedness and readiness and planning.  There will undoubtedly be places where people want to get vaccinated and can’t in early to mid October particularly. 

The CDC’s Anne Schuchat on October 1: We’re all in this together.  We do expect some bumps in the road over the course as we begin this process together.  We’re working really closely with the states and large cities to make things go as smoothly and as effectively as possible. 

Frieden again on October 6: It’s being delivered and each Friday, we’ll provide information on the amount of vaccine available to each state and the amount of vaccine each state has ordered.  That’s a little complicated, because what we have decided to do is make vaccine available as soon as it comes off the production line.  That means, it’s coming available in lots and states learn each day, of additional vaccine available to them.  It’s a little bit of a messy process and we expect it to be somewhat bumpy in the first few weeks. 

The anticipatory guidance has stood the CDC in good stead. Of course there have still been lots of news stories about the chaotic vaccine rollout in various cities. But a high proportion of the stories, especially the national roundups in major media, have included Frieden’s prediction that the early vaccination process would be bumpy. Whereas the vaccine shortage seemed to contradict the CDC’s expectations, the vaccine distribution problems confirmed those expectations.

Public health officials could do much more to empathize with the frustration so many vaccine-seekers are experiencing – and the frustration providers are experiencing as they try to cope with all those vaccine-seekers without enough vaccine.

Most of the acknowledgments-of-chaos we have seen sound pro forma, almost minimizing, something along the lines of “there have inevitably been some glitches as people have coped with the vaccine scarcity.” The seesaw principle applies here too. If officials were more dramatic about how “awful” it has been for some vaccine-seekers and some health providers, the public would have an easier time putting the experience into context: “It wasn’t that awful, just awfully inconvenient.” That’s what a good risk communicator wants people to end up telling themselves – not what the communicator should be telling the people.

And far too many health officials have been thanking the public and the providers for their patience. This is a lot less empathic, and a lot less effective, than acknowledging the justice of their impatience. It also sounds kind of saccharine, and probably generates a “You’re not welcome!” kind of reaction. “Thanks for my patience? What patience? I’m furious at you!” That’s how the risk communication seesaw often works.

The third aspect of vaccine distribution outrage hasn’t been highly visible yet: maldistribution. So far, there haven’t been a lot of places where vaccine was visibly going begging. And when that did happen, the media tended not to make too much of it. The prevailing theme of vaccination news coverage has been shortage. So stories about frenzied parents desperately trying to find some vaccine have outcompeted stories about clinics with more vaccine than they could get rid of … unless, of course, those clinics were caught vaccinating some lower-priority walk-ins.

But as more vaccine gets delivered, and as the shortage story gets old, the maldistribution story should start to surface more: “Why can’t they get the vaccine where it’s needed?” A little anticipatory guidance might be useful now. Empathic, apologetic acknowledgment of the problem – even vivid dramatization of the problem – will be needed in the weeks ahead.

Copyright © 2009 by Peter M. Sandman and Jody Lanard

Contact information page:    Peter M. Sandman