Once the first case of Ebola infection was diagnosed in the US, it was inevitable that it would attract great attention. What was not inevitable is the federal government’s epic failure in risk communication.
A search for on Amazon.com yields 1,288 hits. Many of these volumes are applicable to today’s Ebola crisis. Just one of them — a slim volume on the office book shelf that was published in 1990 based on research and experience dating many years before that — provides a representative guide.
Federal officials could have avoided many of the problems they now face if they had consulted this easy-to-read 150-page book (or any one of a hundred others) and acted upon its advice.
Here are three fundamental principles of risk communication drawn from Hance et al. (1990). Each principle is annotated with a a brief description of how federal officials violated it in the case of Ebola.
Recognize that effective risk communication depends on effective risk management. Openly talking about risk is necessary, not optional, and there is ample evidence that federal officials have not talked openly, thus undermining their credibility. The CDC said the risk of an Ebola outbreak in the US was too remote to warrant visible risk management action, then said “U.S. hospitals can safely manage patients with Ebola.” It is too early to know whether there will be an outbreak in the US, but events have shown that only a few US hospitals can effectively manage it.
Moreover, open communication is not sufficient. Federal officials also needed to take effective actions that reduced risk or its perception.In the case of Ebola, federal officials tried to assuage public concerns without taking concrete actions. When the first US case was diagnosed in a traveler from West Africa, federal officials continued to talk but take no visible actions to effectively manage the risk associated with such travel. They insisted that travel restrictions would not be effective, a conclusion they did not support with evidence and which was contrary to lay intuition. (A recent NBC News flash poll indicated 94% of respondents supported travel restrictions.) Now federal officials have to contend with Ebola and substantially diminished public confidence in their scientific and medical competence.
In contrast, the quarantine imposed by the governors of New York and (especially) New Jersey unambiguously reduced actual risk to their own residents, though perhaps not by much and at the cost of a high rate of false positives. But it also dealt effectively with public concern by minimizing false negatives, which federal officials’ responses have failed to do. It can be hypothesized that the more federal officials and public health elites criticize the state quarantines, the more support they will enjoy among the residents of states in which quarantines have been imposed. New Jersey Governor Christie’s decision to allow Kaci Hickox to go back home to Maine solved his risk management problem and made it Maine’s. Hickox is said to be refusing to comply with the travel restrictions that New Jersey imposed as a condition for her discharge, which makes Maine officials unhappy:
Maine officials have said that they would ask Hickox to quarantine herself at home until the passage of 21 days from her last possible contact with an Ebola patient, adding that they would make it involuntary if she resisted.
Hickox is, perhaps unwittingly, validating Gov. Christie’s decision not to trust her to self-quarantine.
Pay attention to outrage factors. Outrage arises a number of ways, such as when a risk could have been prevented, or when those perceived as being responsible for the risk are inattentive or dismissive of public concerns, or when experts behave arrogantly or condescendingly, or when officials and experts lie. When Hance et al. were writing in 1990, it was chemicals manufacturers that attracted this outrage. “Corporations,” they wrote, “can get into battles if they stubbornly assert their own ‘rightness,’ rather than dealing with community outrage.” For Ebola, replace “corporations” with “governments” and the sentence is apt. Federal officials and experts have displayed each of these behaviors.
Completing the irony, corporations are now attracting complimentary attention because the public appears to trust them more than governments to develop cures or vaccines.
Involve people in ongoing risk decisions. When people participate in decision-making, they are more likely to take ownership of the outcome. When they don’t participate, or more importantly, when experts and government officials exclude them from the decision-making process on the ground that they aren’t knowledgeable or smart enough, people conclude that the process of choosing among competing policy alternatives is fair or respectful. In the case of Ebola, federal officials have excluded the public from the decision-making process despite that fact that CDC’s own Framework for Preventing Infectious Diseases says that public participation is “vital” (p. 22). The public can be expected to distrust federal officials and blame them for every bad outcome, including bad outcomes that could not be avoided.
Hance et al. also have a list of “ten ways to avoid communicating about risk.” In the case of Ebola, federal officials have committed each of these unforced errors to some degree.
Talk instead of acting. Talk is cheap; actions speak louder than words. New Jersey Governor Chris Christie has taken a public beating from federal officials, partisan opponents and the elite media (NYT, Atlantic ; contra see CSM) for imposing and enforcing a mandatory quarantine on travelers arriving at Newark’s Liberty International Airport who have been in direct contact with Ebola patients. However, recent polling prior to the November 4 election does not suggest that his constituents feel the same way. Rather, it is public confidence in the federal Centers for Disease Control that is reported to have declined “dramatically.”
Trivialize the risk. When federal officials, experts and elite but generalist media compare the risk of Ebola unfavorably to influenza, they trivialize the risk of Ebola. This insults people who care about Ebola. Telling people who care about Ebola to stop worrying and get a flu shot instead, they fuel public concern with anger and outrage. Telling people to “calm down” or “trust us” will fail as a risk communication strategy.
Hide behind lawyers. Centers for Disease Control officials and the White House object to the quarantine imposed by the governors of New York and New Jersey, alleging that it is not “science-based.” This opens a Pandora’s Box: on what other occasions have federal officials and the White House insisted that policy be based strictly on science? Some news accounts suggest that the CDC objected because it intruded on its own statutory authority, but more recent accounts indicate the CDC has backed off from this losing argument; it is legally dubious and provocative. The CDC has no statutory authority within state boundaries, and governors are loath to concede ever more authority to federal officials when it is the governors whom state residents will hold accountable.
To the ears of state officials, federal officials’ preemption claims sound like yet another intrusion in the exercise of state police powers. Effective risk communication requires respecting all stakeholders, but it’s especially damaging when federal officials antagonize stakeholders that are also sovereign governments with superior constitutional claims to authority.
If you make a mistake, deny it. Especially in the midst of crisis, it is difficult to admit error. Government officials are especially loath to admit error, but failing to do so won’t persuade people that prior statements or actions weren’t wrong. When CDC Director Tom Frieden implicitly blamed Texas Health Presbyterian Hospital nurse Nina Pham for failing to adhere to the CDC’s protocol, he antagonized all professional nurses even if his statement had been correct. And it turned out that his statement was incorrect. CDC’s original protocol was insufficient to prevent infection. Frieden subsequently revised the CDC protocol, asserting that the new protocol was “based on science” (what was the original protocol based on?). And he did so without actually admitting error, which did not protect the CDC from the loss of public confidence that CBS News polling reveals to have occurred.
Don’t speak plain English. Jargon impedes effective communication; it should be reserved for communicating within an expert community. At the same time, there is the opposite problem which also must be avoided: insulting the public by dumbing-down the message. A common way to dumb-down a risk message is to simply assert that the public is “safe,” and thus shouldn’t worry. There is little evidence that telling people not to worry has any beneficial effect; rather, it tends to antagonize people because they think they are being treated like children. In any case, telling even otherwise credulous not to worry will not work when they see doctors and nurses bedecked in what looks like extraordinary protective gear, and it doesn’t even work.
Lecture. Federal officials and public health experts have been quick to treat public concerns condescendingly (“I understand this can be deeply troubling“) dismiss public concerns as irrational or based on fear, and governors’ actions scientifically uninformed. This undermines the credibility of scientists. They have a role to play in informing policy decisions, but science cannot legitimately do more. Policy decisions are about balancing competing values, not plugging scientific data into an artificial intelligence program.
When events show that scientists are less knowledgeable than they claim to be, or that their purportedly scientific statements are actually based not on science but on different policy judgments, the public loses confidence that the experts really are in command of the facts or that they can be trusted to deliver facts objectively. Once trust is lost, or if officials and experts never bother to invest in its acquisition, the public will suspect that “the science” has been politicized and thus untrustworthy. Effective risk communication is impossible once the public distrusts the communicator.
Don’t let your guard down. Federal officials and scientists have projected an image of hyper-rationality, sometimes an utter absence of emotion. If others disagree with their policy prescriptions, however, they are accused acting irrationally (obviously bad) and being driven by emotion (worse). Hance et al. say the absence of emotion, or more importantly, the refusal to take emotions seriously, destroys effective risk communication. Ridiculing so-called irrational or emotional perspectives wins no friends and earns many enemies.
Wait to talk until someone else does, then respond defensively. Blaming others for the risk management problem — whether it is nurses or doctors, reporters or pundits, or elected officials of the other party — does nothing to build the public confidence needed to support decision-making — especially science-based decision-making, which is not well defined but presumably better than decision-making based on politics. And professions of unqualified support for science-based decision-making fall flat when made by persons who have a reputation for preferring decisions based on politics.
A variant of this “don’t” is to wait until others make difficult policy decisions, then criticize them. Federal officials have refused to enact publicly credible travel restrictions without clearly articulating a scientific reason. They say travel restrictions “won’t work,” but when pressed their argument is unrelated — it’s that travel restrictions might discourage health care workers from serving in West Africa. While state quarantines might have that effect, it is irrelevant to the risk management problem governors face. Governors can be deterred from making policy decisions that federal officials don’t like, or consider inconsistent, by taking actions that substantially reduce or eliminate the risks that governors otherwise have the constitutional authority to manage.
Scrimp on resources for communication. Federal agencies do not scrimp on resources, particularly in their communications shops. But they do scrimp on technical competence among the personnel who work in these shops. Hardly any of them have expertise in any technical field, much less infectious disease generally or Ebola in particular. So it is generally understood that agency communications staff actually know little or nothing. Risk communicators who have to rely on talking points prepared by others are ineffective.
Wing it. Improvisation often makes good theater, but it makes poor risk communication and risk management, and it communicates technical incompetence. In the case of Ebola, the CDC says those who have had direct contact with an Ebola patient only need to be monitored, not quarantined as New York and New Jersey are doing. They say voluntary self-quarantine is sufficient. But the public may not be satisfied or assured by these voluntary actions, and if they fail (as happened in New York, dealing with failure becomes a state government nightmare. The CDC’s advice is not accompanied by CDC responsibility for managing failure.
Meanwhile, federal officials are indeed “winging it” by implementing contradictory policies. The White House objects to New Jersey’s 21-day quarantine,on health care workers who had direct contact with Ebola victims, but the Defense Department is quarantining all soldiers returning from West Africa despite their lack of contact. This is perceived as incompetence or hypocrisy.
President Obama’s decision to appoint a single person to coordinate policy (and presumably messaging) is sensible even if woefully late. However, his decision to appoint an outsider communicates no confidence in the federal agencies and officials who are supposed to be expert. Appointing a well-known Democratic Party operative as his Ebola “czar” communicates another problematic message — that Ebola is a partisan political problem, not a public health problem. That message antagonizes all manner of stakeholders, including governors (who especially chafe at being called partisan by other partisans), scientists (who claim to hate partisan politics), senior agency managers (who consider partisan politics beneath them), and legislators (whose better angels may care about public health but need all the help they can get resisting the temptation to be partisan themselves).
It cannot be predicted how much damage this particular appointment will inflict on federal Ebola risk communication; it is for now uncertain whether the new czar will have a risk communication role. But recall the first principle of risk communication listed above: effective communication requires effective risk management. A federal Ebola “czar” perceived to be a partisan political operative will not be respected by governors, scientists, senior agency managers, or legislators. None of them have any obligation to cooperate, and many of them now have an incentive to obstruct or go their own way. Effective federal risk management, on which effective federal risk communication depends, appears likely to remain elusive.
Billie Jo Hance, Caron Chess, and Peter M. Sandman, Industry Risk Communication Manual: Improving Dialogue with Communities, Lewis Publishers, 1990.