How to Communicate with the Public About Chemical,Biological,Radiological,or Nuclear Terrorism: A Systematic Review of the Literature

Abstract

A deliberate attack involving chemical, biological, radiological, or nuclear (CBRN) material has the potential to cause substantial fear among the public. This presents problems for communicators, who will need to provide information quickly after an attack while ensuring that their messages are easily understood and likely to be attended to by members of the public. Identifying in advance what people would want to know, where they would get information from, and how messages should be presented might allow communicators to ensure that their messages have the best chance of having their desired effect. In this review, we identified all peer-reviewed studies that have assessed communication strategies or information needs using hypothetical CBRN scenarios or in actual CBRN incidents. We identified 33 relevant studies. Their results support existing psychological models of why people engage in health protective behaviors, with information about the severity of the incident, the likelihood of being exposed, the efficacy and costs or risks of recommended behaviors, and the ability of individuals to perform recommended behaviors being sought by the public. Trust plays a crucial role in ensuring that people attend to messages. Finally, while a large variety of spokespeople and sources were identified as being turned to in the event of an incident, the use of multiple information sources was also common, affirming the importance of communicating a consistent message through multiple channels. Further research is required to extend these predominantly US-based findings to other countries and to confirm the findings of research using hypothetical scenarios.

An attack with CBRN material has the potential to cause substantial fear among the public, and communicators will need to provide information quickly while ensuring that their messages are easily understood and likely to be heeded. Identifying in advance what people would want to know, where they would get information from, and how messages should be presented can allow communicators to ensure that their messages have the desired effect. The authors reviewed 33 relevant studies that have assessed communication strategies or information needs using hypothetical CBRN scenarios or in actual CBRN incidents.

Adeliberate attack involving a chemical, biological, radiological, or nuclear (CBRN) weapon might kill or injure many people and cause distress and uncertainty among many more. Its impact could be reduced if official agencies successfully informed the public about how to protect themselves and others before, during, and after the incident and about the true nature of any risk. Four features of CBRN terrorism make this communication difficult. First, CBRN arouses substantial fear among the public, which may impair people's abilities to comprehend information provided to them following an attack.¹ Second, members of the public have low baseline knowledge about CBRN agents, and any beliefs that they do hold about them are often incorrect.^2–5 Third, although familiarity with basic CBRN concepts may help people to prepare for an attack, the perceived low likelihood of an attack occurring may reduce the probability that people will pay attention to information provided during periods of seemingly low threat.^6,7 Fourth, following an attack, the speed with which information needs to be disseminated to and comprehended by the public may require a qualitatively different style of communication compared to other, more traditional, forms of nonemergency public health campaign, which may urge the public to take protective action but which can be planned well in advance, released over a lengthy period of time, and evaluated and adapted as they are implemented.

Because of these challenges, several studies have explored how to provide information to the public about CBRN. Although others have provided theoretical perspectives and commentaries on this literature,^1,8–12 aside from a brief review conducted by Wray and colleagues in 2005,¹³ no detailed systematic review has been published. In this article, we review all studies that have provided original data on how to communicate with the general public about CBRN terrorism.

Our objectives were to clarify: (1) what information people would like to receive before or during a CBRN incident; (2) the factors that determine whether people will seek out information concerning CBRN events; (3) the media sources they would turn to and the spokespeople they would trust; and (4) the factors that determine whether a specific spokesperson or source is used and whether information is seen as trustworthy. These issues have been extensively studied in this literature, on the basis that addressing the questions and uncertainties that members of the public will have, using communication strategies that ensure messages are widely accessed, and ensuring that messages are seen as credible should encourage members of the public to attend to, understand, and, if necessary, act on important recommendations.^8,11,13

Methods

Search Strategy

We searched Psychinfo, Medline, and Scopus using CBRN-related keywords and MeSH terms (eg, “terrori*” or “explosion*” or “cbrn”) in combination with communication-related keywords and MeSH terms (eg, “communication” or “educati*” or “preparedness”). Searches were conducted up to November 2011. We checked the reference sections of all included papers for additional citations. We screened the titles and abstracts of papers online and reviewed the full texts for any that appeared relevant.

Inclusion Criteria

We used several inclusion criteria. First, we included studies only if they were in English and published in the peer-reviewed literature.

Second, we included only studies that contained original data from a sample of the general public. Data collected from healthcare workers or police officers, for example, were excluded, as were data relating to potentially vulnerable groups, such as older adults or those with psychiatric diagnoses.

Third, we included only studies that related to actual or hypothetical incidents in which a CBRN agent was thought at the time to have been deliberately used. We excluded studies relating to emergency or disaster preparedness if they did not involve discussion of CBRN. We excluded studies that related to the use of CBRN agents during war.

Fourth, we included studies only if they (1) explicitly assessed the information that participants would like to receive before or during an incident, (2) contained data that related to where or how participants would like to receive this information, or (3) evaluated the effects of a communication strategy that was intended to alter perceptions, emotions, or behaviors before or during an incident. We included only studies that evaluated “mass communication,” excluding, for example, those evaluating the impact of personal protective equipment on communication between individuals.

Fifth, we excluded nonexperimental quantitative studies that included only nonrepresentative, “convenience” samples or samples that were not recruited from the general population.

Finally, we applied a critical appraisal tool¹⁴ to all qualitative studies that met our other inclusion criteria. This was used to highlight possible methodological deficits that might warrant exclusion.

Data Organization and Analysis

We divided studies based on whether they related to a hypothetical or an actual incident. For quantitative studies, we extracted details regarding sampling strategy, sample, methods, and main findings. For qualitative studies, we extracted details regarding study design, sample, the scenario presented to participants, and the main topics of any discussion. From each article, we listed the main themes that were identified relating to our objectives. We grouped together themes that appeared to relate to the same concept and have reported those concepts supported by at least 2 studies.

Results

Our searches retrieved 12,779 citations, most of which were obviously irrelevant. For almost all of these papers, it was evident that they did not meet our criteria, and we were able to quickly discount them. We did not record the reasons for discounting citations at this stage, for pragmatic reasons. However, the large majority were excluded because they contained no original data; used the term communication to refer to advances in telecommunication infrastructure or communication between organizations or healthcare workers rather than with the public; used the terms preparedness or readiness in relation to a healthcare system's level of readiness; or were from an entirely different field but happened to use one or more of our search terms in their abstracts (eg, in referring to an “explosion” in social media “communication” tools).

We reviewed the full text for 320 papers, of which 29 met the inclusion criteria. These reported data from 33 studies dating from 2002 to 2011. Eleven qualitative and 12 quantitative studies provided information about hypothetical CBRN scenarios, while 8 qualitative and 4 quantitative studies contained data about actual CBRN incidents (2 quantitative studies reported data relating to both categories). The hypothetical scenarios ranged from vignettes portraying severe terrorist attacks (eg, an improvised nuclear device, VX gas) to discussions of general emergency preparedness that also touched on preparedness for CBRN incidents. Studies of actual incidents focused on the aftermath of the 2001 US anthrax attacks, except for 1 study conducted during the polonium-210 incident that occurred in London in 2006-07.

The quality assessment of qualitative studies highlighted 2 studies relating to the anthrax incident that we felt were of borderline quality.^15,16 In each case, information about the methods was limited and it appeared that no recordings had been made of the focus groups or interviews. However, exclusion of these studies did not make any dramatic difference to our results, and we felt the studies had notable strengths in having been conducted relatively soon after the incident, while events were still fresh in the minds of participants. We therefore elected to retain them in the review.

Evidence from Hypothetical Scenarios

What do people want to know?

Eight qualitative studies (Table 1)^17–24 provided data on the information people would want to receive during a CBRN attack. Table 2 summarizes the main themes. Participants most frequently requested information on what protective actions were being recommended and how they should be performed. Although the specific questions that people had about protective actions varied depending on the scenario, the fundamental need to understand what actions were recommended was usually presented as a priority.^17,18,21

Table 1.

Methodological Characteristics of Qualitative Studies Using Hypothetical Scenarios

Reference	Sample	Scenario Discussed: Main Focus of Discussion
Becker, 2004¹⁸	12 focus groups (n=approximately 120); broad mix of sociodemographic characteristics	Improvised nuclear device: emotional reactions, immediate concerns, behavioral intentions, and information needs and where they would seek information
Glik et al, 2004¹⁷	10 focus groups (n=93); broad mix of sociodemographic characteristics	Botulism: knowledge, beliefs, emotional responses, intended actions, and information seeking
Henderson et al, 2004¹⁹	13 focus groups (n=90); broad mix of sociodemographic characteristics	VX gas: responses, concerns, and information needs
Wray and Jupka, 2004²⁰	12 focus groups (n=129); broad mix of sociodemographic characteristics	Plague: knowledge and attitudes, information needs, and information-seeking strategies
Aldoory and van Dyke, 2006²⁸	6 focus groups (n=62) with university students (5 groups) or professionals	Contamination of food crops: preferences for news about food safety and food terrorism
Carter-Pokras et al, 2007²⁶	6 focus groups with Latino community members (n=45) or health promoters (n=6)	General emergency preparedness: how to define or identify an emergency, sources of information that will be relied on in an emergency
Gibson et al, 2007²²	Focus groups (n=75); broad mix of sociodemographic variables	General emergency preparedness: concerns about terrorism, how the threat of terrorism has affected them, ways of coping or preparing, views on information sources
Marshall et al, 2007²¹	12 focus groups (n=112); broad mix of sociodemographic variables; participants all from Rhode Island	General emergency preparedness: past emergencies, sources and barriers for information, who is trusted, likely compliance, preferred communication means
Meredith et al, 2007³¹	8 focus groups (n=75) with African American participants from Los Angeles	Smallpox: responses to a variety of new information presented during the scenario
Glik et al, 2008²⁴	9 focus groups (n=73) and 27 cognitive interviews; broad mix of sociodemographic characteristics	Botulism: responses to information materials provided during the study and recommendations for improvement
Rinchiuso-Hasselmann et al, 2010²³	8 focus groups (n=74); participants from English-, Spanish-, and Chinese-speaking communities in New York City	Anthrax attack requiring mass distribution of prophylaxis to the public: perceptions of public health emergencies and emergency distribution strategies, willingness to collect medication for friends and neighbors, information needs

Table 2.

Information Needs Identified in Qualitative Studies of Hypothetical Scenarios

Main Themes	Subthemes, with References to Supporting Studies
Recommended protective actions	• How to prepare^17,22 • How to protect self and others from exposure^17,21,34 • Where to go²¹/What shelter is suitable?¹⁹/How to take shelter if in the car¹⁸/Should people stay at home?²⁰ • How long is sheltering required?¹⁹ • How is the illness prevented?¹⁷/How is transmission prevented?²⁰
Medical treatments and prophylaxis	• What is the treatment?^17,24 • Where can treatment be obtained?^17,20 • How available are the required medications?²⁰/Is sufficient medication available?²³ • How will mass prophylaxis distribution sites be used?²³
Efficacy or rationale	• Evidence that recommendations are “tried and tested”¹⁸ • Description of the rationale for recommendations^21,22 • How effective is treatment for different levels of exposure?¹⁹
How do people become exposed?	• What is the agent?^17,18/What types of the agent exist?²⁰ • How does someone become infected?¹⁷/Is food and water safe?¹⁸/Is the food chain affected?¹⁹/What are the mechanisms of exposure?¹⁹/How does the agent “work”?^18,24/What should one do with pets?¹⁸/ Is exposure to pets or livestock problematic?^19,20/How does it spread?²⁰/What is the concept of “contagion”?²⁴
How to tell if treatment is required	• How can people tell if they have been exposed?^18,20/What are the symptoms?^17,20/How are the symptoms different from those of other illnesses?²⁴/When is medical attention needed?¹⁸ • What are the long-term effects?¹⁸
Severity of the incident	• How much damage has been caused?¹⁸/How much of the agent has been released?¹⁸/ What is the magnitude of the attack?¹⁹
Likelihood of being affected	• Where is it safe?^17,18/Where is the threat?¹⁷/What area is affected?¹⁸/How big an area is affected?¹⁸/Who is affected?¹⁷/How far away is safe?¹⁸ • Which direction is the wind blowing in?¹⁸/Where is the threat heading?¹⁹/Could it happen again?¹⁸
Justice issues	• What happened?¹⁷/Why did it happen?¹⁸ • Who is responsible?^17,18,20/What is being done to find the perpetrators?²⁰ • Is it still in progress?¹⁹
Roles of emergency responders	• How do emergency warning systems work?¹⁷ • What are the roles of different official agencies?¹⁷ • How is the state preparing for an incident?²¹
Where to get more information	• Where to get accurate, credible information²¹/Contact telephone numbers²⁰
Meaning of jargon	• “Shelter in place”¹⁸/“plume”¹⁸/color codes for US Homeland Security Advisory System¹⁸/“aerosol attack”²⁰/“toxin”²⁴/“antitoxin”²⁴/“airborne but not contagious”²⁴
Response costs and risks	• How safe is any medication?²³ • Could exposure occur while attending a mass prophylaxis center?²³ • How secure are points of dispersal?²³

Themes concerning recommended medical treatments or prophylaxis were also identified, with participants wanting to receive basic factual information on what countermeasures or treatments existed, when it would be appropriate to obtain them, and how to obtain them.

While factual information was requested on these topics, supporting information was also required. In particular, a need for information about the efficacy or rationale of recommendations was raised. Understanding why recommended actions would be effective often necessitated a degree of background knowledge that participants did not have. In addition to basic information such as what the exposure was, this background information centered on how one might become exposed and how to tell if medical attention is necessary. With regard to the first issue, participants raised questions about how one could become affected, whether food and water would be safe, whether the agent was contagious, how the agent “worked,” and whether pets or livestock might pose a hazard. With regard to the second issue, questions focused on how to tell if one had been exposed, what the symptoms of exposure are, and how to distinguish them from the symptoms of other illnesses. In the absence of this information, participants were sometimes dubious about the need for particular protective actions²⁴ and whether they would follow recommendations.^18,22

Participants also expressed a need to understand the severity of the attack and, more particularly, the likelihood that they or their family would be affected. Justice issues were identified in some studies, with participants keen to understand why an attack had happened, who was responsible, and what was being done to find them.

Smaller themes, apparent in fewer studies, included a desire for information about the role of emergency responders and where to get further information. Requests were also identified about the meaning of jargon, reflecting a concern that the information provided about a CBRN attack might be too technical to understand.^18,20,21 The specific pieces of jargon asked about depended on the scenario but included terms such as “shelter in place,” “plume,” and “toxin.”

Finally, one study²³ that focused on whether people would collect prophylactic medicine following an attack identified several requests for information about the possible costs or risks of obtaining medicine. These included concerns about the safety of the medication, the possibility of being exposed to the agent while at a mass distribution center, and a need for reassurance about security at the center.

In terms of quantitative data, Pollard reviewed 3 large cross-sectional surveys conducted in the US that asked participants what information they would like to receive in the event of a bioterrorist attack.²⁵ Participants wanted to know: the name of the agent and how it could be transmitted (91% to 97% of participants); how to minimize the risk of infection (89% to 94%); treatment recommendations (86% to 92%); the number of people who had been affected and their location (70% to 81%); the nature of any travel restrictions (69% to 79%); and the names of the suspects (40% to 44%).

What influences whether people will seek information?

Six qualitative and 4 quantitative studies provided data on the motivations for and barriers to seeking information about CBRN. The qualitative studies identified the following as motivators for information seeking prior to an attack: knowing how to protect self and family,^22,26 being able to help others,²⁶ and feeling more secure or reassured.²⁶ Barriers included being too busy to act on any information²⁶ and believing that preattack warnings reflected the government “crying wolf.”²² The quantitative studies supported these themes. In a small US survey, seeking out information about bioterrorism was significantly more likely among respondents who felt bioterrorism was a serious social problem, who felt they had fewer constraints in engaging in preparatory behaviors, and who felt it was more of a personal issue to them.²⁷ In a Canadian survey,⁷ seeking out information about terrorism was significantly predicted by perceptions about the seriousness of an attack, the likely impact of an attack on the respondent's life, the ability of the respondent to cope with an attack, and the probability of an attack occurring. While higher perceptions of probability, ability to cope, and personal impact were associated with greater information seeking in this study, higher perceptions of severity were associated with lower inclination to seek information.

Three qualitative studies briefly touched on factors that would motivate people to seek information after an attack had taken place. These described the same fundamental motivators: uncertainty about what had happened and a desire to know what to do.^17,18,22 One study addressed the issue in more detail by presenting participants with a scenario involving the covert contamination of food crops.²⁸ Three types of factor were found to determine whether people would attend to information about the attack. “Problem recognition” was largely determined by the rank of the official spokesman communicating about the incident, the number of deaths that occurred, and whether the individual ate the type of food that had been contaminated. “Perceived level of involvement” was determined by the attack's geographical proximity, the participant's perceived similarity to the victims, their perceived susceptibility to the risk, and perceptions as to whether the spokesperson shared the risk with them. “Barriers” included lack of access to media sources, lack of time, dietary or other restrictions that prevented them from complying with official recommendations, fatalism, low self-efficacy, information overload, and prioritization of other day-to-day worries. In a subsequent experiment, the same team presented 4 versions of a similar scenario to 4 groups of participants.²⁹ The versions differed in terms of the similarity of the victims to the participants and whether the spokesperson shared their risk. While perceived similarity predicted whether the participants would seek out more information after hearing about the incident, there was no effect of the shared risk manipulation.

Two quantitative studies assessed the likely impact of pre-event education about CBRN terrorism. In the first, 116 university students read either “potent” descriptions of bioterrorism or a message that contextualized the risk as being relatively minor.³⁰ Subsequent levels of anxiety among students in the potent information group showed a significant increase. The second study evaluated the impact of an education campaign that saw the distribution of a CBRN-related booklet to all 440,000 residential addresses in Rhode Island.²¹ Although telephone surveys before (n=251) and 1 month after (n=250) the intervention identified some improvements in preparedness behavior, only 24% of respondents in the “after” survey recalled receiving and reading the booklet, and only 10% reported engaging in any of the preparedness behaviors it recommended.

Where do people want to receive information from?

Eleven qualitative studies^{17–24,26,28,31} assessed from where participants wished to receive information before or during an incident. These identified a wide range of preferred spokespeople and sources. Spokespeople encompassed most national and local-level organizations that might be involved in responding to an incident. Sources for information focused on national and local media, the internet, contact with friends and neighbors, and any emergency communication mechanisms that might be introduced. Participants often reported that they would use multiple sources in order to check for consistency and to obtain additional details.^{17,19–21,31}

Nine cross-sectional surveys assessed from where people wished to receive information (see online Supplementary material at www.liebertonline.com/bsp). Seven were large, methodologically robust surveys of the general populations of Canada³² or the US.^5,25 Participants generally rated people associated with the healthcare system as preferred and trusted spokespeople, with the mass media being the source most likely to be used to access information. These surveys also confirmed that people would generally refer to multiple sources and spokespeople. The remaining 2 studies were smaller and less methodologically robust.^27,33 Overall, however, they confirmed that people are likely to use multiple information sources.

What determines where people wish to receive information from?

Eleven qualitative studies contained information about the factors that determine why someone would seek information from a given source.^{17–19,21–24,26,28,31,34} Several themes were identified (Table 3), of which trust easily predominated.^17,19,23,24 Trust was composed of several more specific subthemes. These included: preexisting familiarity with a source; the perceived competence, experience, and expertise of the source; the source's perceived honesty; the consistency of messages between different sources and between the statements and actions of individual agencies; and faith that an individual or organization could be trusted, borne out of experience or a lack of any alternative. The importance of honesty, in particular, was also supported by one cross-sectional survey.²

Table 3.

Themes Identified in Qualitative Studies of Hypothetical Scenarios

Theme	Subthemes
Trust	• Trust is an important determinant of whether a source is used.^17,19,23,24 • Familiarity with source is a determinant of trust.²¹ • Competence of sources determines if people wish to hear from them.²²/Sources are more trusted if they seem to know what they are talking about,³¹ have credentials or experience,³¹ and are well-informed.²⁶ • Honesty is a determinant: People try to judge if information is complete and accurate,^31,34 has been held back to prevent panic or for political gain,^18–20,31 or is the product of sensationalist reporting.^17–20 • Consistency is important: People check if information between sources is comparable^20,31 and if messages are consistent with official actions.³¹ • Faith is important: People may have immediate trust in a source, partly through lack of any choice but to trust their messages.^22,23,31
Format	• Identification of source as official: easily differentiated from “infomercials”²⁴/has clear logos visible^24,28,31/has the source of the information clearly identified²⁰ • Written materials are more credible than other formats.²⁴/Written materials are less ephemeral and easier to review.²⁴/Written fact sheets are useful before an incident^21,23,26 (supplemented by other formats²⁶). • Radio and television formats present information too quickly to absorb.²⁴/ Television formats are better at demonstrating “action steps.”²⁴
Language	• Provision of information in the person's first language is useful.^19,26/Information in different languages may be used by people to check for nuances.¹⁹
Benefits of local sources	• Local sources are turned to for information that is more specific to the local area.¹⁹/Local sources are perceived as providing information faster,¹⁷ having a better or more detailed understanding of local situation,¹⁸ and being more credible and dedicated.³⁴
Timing	• Television, radio, and internet are used during the period immediately after an attack, then newspapers and government or university websites for more detail as it appears.^17,20

Aside from trust, several pragmatic factors were also identified as determinants of where people would turn, including the format and branding of information and the availability of information in an individual's first language.^19,26 Although a clear desire was expressed by participants across the literature for information from national organizations and media, several studies identified an additional desire for information from local sources,^{17–20,22,26,34} which were felt by some to have a better insight into the local situation as well as being more credible and dedicated when it came to providing information to protect local people.

Evidence from Actual Incidents

Eight qualitative studies assessed how people perceived communication during the US anthrax attacks^{15,16,35–39} or the London polonium-210 incident⁴⁰ (Table 4). All interviewed people who were at risk of having been exposed. Four quantitative studies employed a cross-sectional design to assess information needs during these incidents.^16,25,40 One was conducted among postal workers at risk of exposure to anthrax.¹⁶ The other 3 questioned unexposed members of the public.

Table 4.

Methodological Characteristics of Qualitative Studies Based on Actual Incidents

Reference	Sample	Main Focus of Discussion
Jefferds et al, 2002¹⁶	Individual interviews (n=22) and 5 focus groups (n=38); postal workers from a facility contaminated with anthrax	Evaluation of interventions that were used to support adherence to antimicrobial prophylaxis
Stein et al, 2004³⁷	Semistructured interviews (n=54); postal workers from a facility contaminated with anthrax and Senate employees from a contaminated building	How participants learned about their potential exposure, how they assessed their risk, what actions they took to protect their health, perceptions about the pros and cons of following public health recommendations
Blanchard et al, 2005³⁸	4 focus groups (n=36), postal workers from a facility contaminated with anthrax; 1 focus group (n=7), Senate employees from a contaminated building	Issues relating to communication and trust about the public health response
Miro and Kaufman, 2005¹⁵	A first set of 7 focus groups (approximately n=100); postal workers from 2 facilities contaminated with anthrax A second set of focus groups (n unreported); postal workers from 1 facility contaminated with anthrax	Gaps in knowledge about anthrax and preventive antibiotic therapy, understanding about antibiotic side effects, factors affecting adherence to antibiotic regimen, trustworthy sources of information, how to increase adherence to antibiotic regimen
North et al, 2005³⁵	5 focus groups (n=28); political staff from Capitol Hill offices, potentially exposed to anthrax	Issues, thoughts, and feelings about the anthrax incident
Clarke and Chess, 2006³⁶	In-depth interviews (n=15); members of faculty and students from a university campus who were present during a white powder incident	Information that people recalled receiving, how they received it, and how they responded to it
Rubin et al, 2007⁴⁰	In-depth interviews (n=86); members of the public potentially exposed to radioactive polonium-210 following the poisoning of Alexander Litvinenko	Reasons for contacting health agency, perceptions of the incident, what information was received and how it was perceived, effects of the incident on their lives, whether they did or would have accepted testing for exposure, what their understanding of the test and its results were
Quinn et al, 2008^39,45	6 focus groups, individual interviews, and key informant interviews (total n=65); postal workers from 3 facilities contaminated with anthrax	Experiences during the crisis, risk perceptions, how and where information was obtained, trustworthiness of sources, preferences for public health measures, appropriate spokespeople, and the anthrax vaccine

What information did people want to know?

Table 5 lists the main themes identified concerning the type of information that people wished to know. A particular need was to understand the likelihood of having been exposed. This included a desire for information about where and when contamination had occurred, details about the purpose of tests for exposure and when the results of these tests would be available, and an explanation of the results. How to tell whether treatment is required was also important, with participants asking about symptoms to watch out for, how to differentiate the symptoms of exposure from those of other illnesses, what the presence or absence of specific symptoms might indicate, and whether treatment could be halted if no symptoms developed. Two additional themes related to the efficacy/rationale and costs or risks of recommended actions. For example, in the anthrax incident, protective actions for postal workers included wearing gloves or washing hands, but some asked why this would work if anthrax spores could be inhaled. In the costs and risks theme, details were requested about the possible side effects of antibiotics and about how well the antibiotics had been tested.

Table 5.

Themes Identified in Qualitative Studies of Actual Incidents

Theme	Subthemes
Likelihood of being affected	• Likelihood of being exposed^36,40/Results of environmental tests for hazardous substance^16,35/What area has been contaminated?^35,37,45/What is the risk?¹⁶/What are the results of the outbreak investigation?¹⁶ • The exact purpose of any tests for exposure that are carried and how to interpret their results^35,37/When will medical test results be available to the individual?^35,38/What are the medical test results?³⁷/What do the medical test results mean?⁴⁰
Efficacy or rationale	• Information about anthrax and antimicrobial prophylaxis¹⁶/Efficacy of antibiotics³⁵ • The rationale or need for protective measures such as wearing gloves or washing hands questioned^16,37,45 • Rationale for any differences in treatment protocols compared to previous incidents or uses³⁹
How to tell if treatment is required	• What symptoms to look out for⁴⁵/How to differentiate symptoms from other causes³⁷/Can medication be stopped if no symptoms develop?³⁷/What are the implications of having certain symptoms?⁴⁰/What is the long-term viability of spores already held in lung (ie, the possible latency period for illness)?¹⁶ • Who is most likely to be susceptible (eg, by age, current medication, etc)?³⁷
Response costs and risks	• Side effects and drug interactions of medical countermeasures^16,37–39 • How well tested is the intervention?³⁹/Why is treatment “experimental”?³⁹
Information specific to individual circumstances	• Ability to ask personal questions is valued.¹⁶/Individualized information on the likelihood of exposure is sought.^37,40

One additional cross-sectional survey (n=1,000) contacted a representative sample of the London population during the polonium-210 incident.⁴⁰ This asked participants whether they agreed or disagreed with statements made by the UK Health Protection Agency. The associations between believing these statements and perceiving one's own health to be at risk were assessed. The only statement to show a substantial association with concern related to the limited spread of the contamination—namely, “if you have not been in one of the areas known to be contaminated, then there is no risk to your health.” Participants who did not believe this were more likely to feel that their own health was at risk.

Where did people want to receive information from?

Participants reported using multiple sources of information. The main health agencies responding to the incident were the most commonly mentioned, together with employers and occupational health physicians for people who were exposed while at work. Numerous other sources and spokespeople were also mentioned, ranging from the mass media to friends and family. The only quantitative study to assess this issue among postal workers confirmed that most (82%) wanted to receive information from multiple sources and in multiple formats.¹⁶

In terms of the general public, 2 large, nationally representative US surveys conducted at the time of the anthrax attacks asked members of the general public where they had sought information from about anthrax or other severe illnesses.²⁵ Of the respondents to each survey, 34% and 45% reported that they had not sought any information. Among the remainder, local television and radio (60% and 48%), cable or network news channels (40% and 26%), and internet health sites (22% and 13%) predominated.

What determines subjective impressions about communication?

While studies of hypothetical incidents provided information about factors that would attract someone to a particular information source or spokesperson, people who were potentially exposed during an actual incident usually had little choice but to receive most of their information from the health agencies that were involved or from their employer. As such, these studies focused on how people evaluated that information. Table 6 describes the main themes that determined these subjective impressions.

Table 6.

Themes Identified in Qualitative Studies as Determining the Subjective Opinion of Potentially Exposed People about Communications

Theme	Subthemes
Insufficient information	• Increases concern^35,38,40,45
Trust	• Trust as an important factor^{35,37–39,45} • Faith: Accepting benevolent intent,³⁷ despite poor communication³⁵ or limitations of communicators knowledge⁴⁵/previous negative interactions with the communicator can degrade faith^39,45 • Motives: Collecting “data” from affected people for studies may give mixed messages about priorities, depending on how studies are explained.^{35,37–39,45}/Perception that economic impact is being emphasized causes concern.^38,39/Perceptions that public health response is designed for responding agencies to “cover [their] ass.”³⁷ • Familiarity: Access to a single, familiar spokesperson provides reassurance.^35,38/Familiar spokesperson or organization needs to be trusted.⁴⁵ • Advocacy: Having an identifiable individual working on your behalf to get information.^35,37 • Competence: Authorities seeming oblivious to limitations of their knowledge increases perceptions of incompetence or deceit.³⁵/Poor organization of those responsible for communication degrades trust.^38,45/Perceptions that those communicating have insufficient knowledge or expertise reduces trust.^15,37,38,45 • Honesty: Sensationalist media reporting causes anger.³⁵/Openness about uncertainties or internal deliberations welcome.^35,45/Perceived lack of openness or deliberate deceit by government causes concern.^38,40,45 • Unsubstantiated reassurance: Reassurances that are later challenged by worsening events or that do not tally with official actions,⁴⁵ or that are not supported by evidence,⁴⁰ cause concern and loss of trust. • Consistency: Contradictory information increases worry^35,38 and degrades trust^38,45 if contradictory over time,^35,45 between sources,^35,39,45 or if apparent contradictions exist between messages given out and actions taken.^35,45 Inconsistency in messages between official agencies^39,45 or between workers from the same agency seen as particularly problematic.⁴⁵
Language	• Lack of interpreters for people with hearing impairments causes concern.³⁸
Information from an appropriate source	• Obtaining information from the media instead of employer or public health agency increases anger.³⁵/Information should come from official agency.¹⁵
Timely, regularly updated information	• Timely information seen as important.^38,45/Regularly updated information is required.⁴⁰/Several opportunities to discuss information needed as new issues arise.¹⁶

At the most basic level, complaints about insufficient information were common, with participants noting that a lack of timely, regular updates could leave them feeling uncertain. Linked to this was a desire to obtain information from an appropriate source. In the absence of regular updates from official agencies, participants sometimes received their updates from the mass media. Having information presented in a person's first language was less frequently mentioned, although 1 study did note dissatisfaction among hearing-impaired postal workers at the lack of interpreters to help explain the situation.³⁸

As with the literature on hypothetical incidents, trust and its components were common themes, with most studies describing the absence or deterioration of trust as a key problem. Similar subthemes to those found in the hypothetical scenario literature were sufficient to explain the findings relating to trust. First, preexisting faith that responding agencies would act in the best interests of those affected was mentioned as a reason why messages might be believed and acted on. The perceived motives underlying the public health response could also affect the credibility of communications. For instance, an apparent focus by officials on the scientific opportunities presented by an incident or on its economic or political impact could damage trust. Familiarity with a source was also important, particularly if the source was trusted prior to the incident. Having an identifiable individual working as an advocate or intermediary between affected people and responding organizations was seen as helpful, by providing a stable, trustworthy way of obtaining information. The perceived competence of communicators was raised in several studies, with organizations that seemed inexperienced, poorly organized, or unaware of the limitations of their knowledge being less trusted. The honesty of sources was also frequently questioned. Perhaps linked to the subtheme of honesty, unsubstantiated reassurance was also presented as a problem, particularly where events subsequently proved initial reassurances to be unfounded. Finally, a lack of consistency was worrying, with consistency in information sought over time, between sources, and between spokespeople from the same agency.

Discussion

While communicating with the public about a future CBRN emergency will pose challenges, our review provides some guidance on what to say and how to say it.

What to Say

On the most fundamental level, our findings support the need for communicators to consider the existing understandings and “mental models” of the public and to tailor their messages accordingly to correct serious misunderstandings and to resonate with current conceptions.⁴¹ Doing this may help members of the public to better understand the rationale and importance of those protective actions being recommended.

Our findings are also in line with general psychological models from the field of behavior change, such as Protection Motivation Theory.⁴² This model suggests that people are motivated to take action to protect their health if they believe that a given threat is likely to affect them and will have severe consequences, if the protective action appears effective and does not have excessive costs or risks associated with it, and if they believe that they are capable of performing the action (“self-efficacy”).

Before an incident takes place, for example, the literature suggests that not everyone will wish to engage with information about CBRN attacks^21,25 because they do not believe CBRN scenarios are likely to happen or likely to affect them personally even if they do happen.^7,27,28,40 These perceptions may be difficult to challenge, not least because they may be reinforced by a belief that governments often “cry wolf” about health risks.^22,30,43 Nonetheless, it may be possible to identify naturally occurring moments when people's sense of safety becomes less secure and they become receptive to information about emergency preparedness.⁴⁴ When such instances occur, people might be encouraged to access CBRN-related information if factors that reduce a person's self-efficacy for using the information are tackled: for example, perceptions that people are too busy to act on any emergency preparedness advice^26–28 or that personal actions will not make any difference following an attack.^7,28 Motivating people to access information might also be achieved by emphasizing how effective this will be in improving outcomes that people value, such as learning how to protect themselves and their family,^22,26 helping others,²⁶ or feeling more secure.²⁶

Once an attack has occurred and when the risk of being exposed to a CBRN agent is ongoing, people may be less skeptical about how serious a problem CBRN is but will still have a pressing need to understand how likely they are to be affected. Clear advice on this and on what actions to take to prevent exposure should be a priority.^17,18,21,25 Information about the efficacy of recommended actions is also needed if uptake is to be maximized. Evidence from the qualitative work in this area suggests that this can be achieved by explaining how these actions work and how we know that they are effective.^16,24,37,45

Among people who have potentially already been exposed, information relating to the likelihood of exposure is crucial as a determinant of levels of anxiety and uptake of medical countermeasures. Information in this situation should focus on the findings of relevant medical, environmental, and police investigations.^{16,35–37,40,45} For those who believe that they have been exposed, issues surrounding medical countermeasures or treatments become the priority. Ensuring that people understand when to seek treatment can be difficult, particularly as many of the early symptoms associated with CBRN exposures can be similar to those of other common illnesses or anxiety.^{18,20,24,37,40,45} To prevent a surge of low-risk patients from overwhelming medical facilities, communicators should focus on explaining the objective signs that exposure may have occurred (eg, location or fever) rather than subjective symptoms.⁴⁶ Where mass prophylaxis is required, questions may be raised as to the efficacy, safety, and need for medical countermeasures.^{16,23,35,37–39} Experience during the US anthrax attacks^37,39 and the 2009-10 H1N1 pandemic⁴⁷ suggests that ensuring that people begin and then complete their course of medication may be difficult. Describing how medications have been tested, when and why they are required, and how well they work should assist people in making informed choices about them.

How to Say It

Our review reaffirms that trust and its components are central determinants of whether a message will be attended to or have its desired effect.^48,49

During a major incident, the public will turn to many information sources and spokespeople to gather information and check the validity of messages. Ensuring that consistent messages are given by representatives from multiple organizations is the best way to achieve the widest reach and strongest impact. Efforts should be made to ensure that this includes representatives and community leaders who are trusted by different sections of society and that information is provided in multiple languages. Consideration should be given to engagement with local information sources that people may turn to for advice about the situation in their vicinity.

For small incidents or incidents involving long-term contact between an official agency and an identifiable group of people, steps should be taken to ensure that trust is maintained and that communications remain credible. It should be possible to capitalize on existing levels of trust if a familiar, respected organization takes the lead in communicating. The use of independent advocates willing to act as points of liaison between affected groups and responding agencies may also help to maintain trust and provide an assurance that the views and concerns of affected people are being taken into consideration. Demonstrating that responders are motivated solely by the well-being of those affected, that they are competent, and that they are being open and honest will also help to maintain trust and may prevent responders from falling into the trap of offering nonspecific, counterproductive “reassurances.” Finally, while inconsistency between official messages and media reporting may be unavoidable, demonstrating that responders can be trusted may reduce the negative impact of such inconsistencies, particularly as many people already suspect that media reporting will be inaccurate during a crisis.^17–20

Methodological Limitations

While a substantial body of work was identified in our review, several caveats should be borne in mind.

First, in many of the qualitative studies we drew on, the original authors warned against generalizing from their results. However, the consistency with which findings were identified across the literature suggests that they do have applicability beyond the contexts and participants of the original studies. Nonetheless, the emphasis in this literature on studies from the US does raise some concerns as to whether the results can be applied to other cultures. While it seems probable that the general themes will translate across cultures, other aspects—for example, the types of spokespeople or sources that would be trusted—may not. Additional research may be warranted to check the extent to which findings from the US transfer to different countries. Similarly, changes over time may also have rendered some of the findings out of date. The possible impact of the recent expansion of online social media, for example, remains to be seen.

Second, one issue that we did not explore in detail was differences in information needs among different sections of society, particularly people from different ethnic groups. In practice, however, these differences appear to be limited.²⁰ Exceptions include the need for information in one's first language,^18,19,21 the need for additional information among populations that might be considered particularly vulnerable,²¹ and differences in who would be the most trusted spokesperson.^21,23

Third, while many of the findings from studies on hypothetical scenarios were also supported by studies conducted during the US anthrax or UK polonium-210 incidents, one area where data from actual incidents was notably absent was information needs in the immediate hours or days following an incident. Findings here reflect studies using hypothetical scenarios. These have important limitations. In particular, it is uncertain whether the intentions and information needs described by participants in the relative comfort of a focus group discussion concerning a purely hypothetical incident would necessarily translate into actual behavior during the chaos of the early stages of a real incident. Ensuring that studies are able to capture information needs in real time during the next major incident will require that research be planned in advance. This requires commitment from researchers, ethics boards, and funding bodies but would be helpful in both advancing knowledge in this area and providing input for any ongoing public health response.

Fourth, we cannot be certain that our searches identified all relevant literature in this area. Had we searched additional databases or used broader search terms, it is possible that we would have identified additional papers. The absence of any specific MeSH terms or keywords for this field makes searching this literature particularly difficult.

Finally, our review is necessarily limited by the nature of the studies reported in the literature. These are primarily descriptive. Studies that empirically evaluate the impact of communications about CBRN are rare. While understanding what people would like to know about CBRN and how they would prefer to receive this information should help communicators to ensure that their messages are attended to, evaluating whether these messages have the desired effect on behaviors or emotions requires more attention.

Conclusion

While communicating with the public about a CBRN threat will present challenges, the research carried out in this field to date suggests several factors that communicators can consider when designing their messages. By ensuring that messages cover areas that members of the public wish to hear about, that the messages are communicated by trusted spokespeople and via widely accessed sources, and that trust in the lead organizations is promoted and maintained, it may be possible to encourage more people to adhere to recommended protective behaviors. These findings are strikingly similar to those found in the wider social science literature concerning public responses to emergency warnings in general.⁵⁰ While the public's general lack of knowledge about CBRN threats inevitably presents challenges for those responsible for designing messages, it is reassuring to know that the established principles of risk and crisis communication remain applicable.

Footnotes

Acknowledgments

We are grateful to the UK's Defence Science and Technology Laboratory, who funded this review, and to the UK Health Protection Agency's Emergency Response Development Group Psychosocial and Behavioural Issues subgroup for their feedback. All views expressed are those of the authors and not necessarily those of their funders or employers.

References

Keselman

, Slaughter

, Patel

. Toward a framework for understanding lay public's comprehension of disaster and bioterrorism information. J Biomed Informatics, 2005; 38:331–344.

Fischhoff

, Gonzalez

, Small

, Lerner

. Evaluating the success of terror risk communications. Biosecur Bioterror, 2003; 1:255–258.

Blendon

, Desroches

, Benson

, Herrmann

, Taylor-Clark

, Weldon

. The public and the smallpox threat. New Engl J Med, 2003; 348:426–432.

Marshall

, Begier

, Griffith

, Adams

, Hadler

. A population survey of smallpox knowledge, perceptions, and healthcare-seeking behavior surrounding the Iraq invasion—Connecticut 2002-03. Biosecur Bioterror, 2005; 3:246–255.

SteelFisher

, Blendon

, Ross

et al. Public response to an anthrax attack: reactions to mass prophylaxis in a scenario involving inhalation anthrax from an unidentified source. Biosecur Bioterror, 2011; 9:239–250.

Rubin

, Brewin

, Greenberg

, Simpson

, Wessely

. Psychological and behavioural reactions to the bombings in London on 7 July 2005: cross sectional survey of a representative sample of Londoners. BMJ, 2005; 331:606–616.

Lee

JEC

, Lemyre

. A social-cognitive perspective of terrorism risk perception and individual response in Canada. Risk Anal, 2009; 29:1265–1280.

Covello

, Peters

, Wojtecki

, Hyde

. Risk communication, the West Nile virus epidemic, and bioterrorism: responding to the communication challenges posed by the intentional or unintentional release of a pathogen in an urban setting. J Urban Health, 2001; 78:382–391.

Wray

, Kreuter

, Jacobsen

, Clements

, Evans

. Theoretical perspectives on public communication preparedness for terrorist attacks. Fam Community Health, 2004; 27:232–241.

10.

Vanderford

. Breaking new ground in WMD risk communication: the pre-event message development project. Biosecur Bioterror, 2004; 2:193–194.

11.

Glik

. Risk communication for public health emergencies. Annu Rev Public Health, 2007; 28:33–54.

12.

Rogers

, Amlôt

, Rubin

, Wessely

, Krieger

. Mediating the social and psychological impacts of terrorist attacks: the role of risk perception and risk communication. Int Rev Psychiatry, 2007; 19:279–288.

13.

Wray

, Becker

, Henderson

et al. Communicating with the public about emerging health threats: lessons from the Pre-Event Message Development Project. Am J Public Health, 2008; 98:2214–2222.

14.

Walsh

, Downe

. Appraising the quality of qualitative research. Midwifery, 2006; 22:108–119.

15.

Miro

, Kaufman

. Anthrax in New Jersey: a health education experience in bioterrorism response and preparedness. Health Promot Pract, 2005; 6:430–436.

16.

Jefferds

, Laserson

, Fry

et al. Adherence to antimicrobial inhalational anthrax prophylaxis among postal workers, Washington D.C., 2001. Emerg Infect Dis, 2002; 8:1138–1144.

17.

Glik

, Harrison

, Davoudi

, Riopelle

. Public perceptions and risk communications for botulism. Biosecur Bioterror, 2004; 2:216–223.

18.

Becker

. Emergency communication and information issues in terrorist events involving radioactive materials. Biosecur Bioterror, 2004; 2:195–207.

19.

Henderson

, Henderson

, Raskob

, Boatright

. Chemical (VX) terrorist threat: public knowledge, attitudes, and responses. Biosecur Bioterror, 2004; 2:224–228.

20.

Wray

, Jupka

. What does the public want to know in the event of a terrorist attack using plague? Biosecur Bioterror, 2004; 2:208–215.

21.

Marshall

, Petrone

, Takach

et al. Make a kit, make a plan, stay informed: using social marketing to change the population's emergency preparedness behavior. Soc Mar Q, 2007; 13:47–64.

22.

Gibson

, Lemyre

, Clement

, Markon

MPL

, Lee

JEC

. Terrorism threats and preparedness in Canada: the perspective of the Canadian public. Biosecur Bioterror, 2007; 5:134–144.

23.

Rinchiuso-Hasselmann

, Starr

, McKay

, Medina

, Raphael

. Public compliance with mass prophylaxis guidance. Biosecur Bioterror, 2010; 8:255–263.

24.

Glik

, Drury

, Cavanaugh

, Shoaf

. What not to say: risk communication for botulism. Biosecur Bioterror, 2008; 6:93–107.

25.

Pollard

. Public perceptions of information sources concerning bioterrorism before and after anthrax attacks: an analysis of national survey data. J Health Commun, 2003; 8,Suppl 1:93–103.

26.

Carter-Pokras

, Zambrana

, Mora

, Aaby

. Emergency preparedness: knowledge and perceptions of Latin American immigrants. J Health Care Poor Underserved, 2007; 18:465–481.

27.

Lee

, Rodriguez

. Four publics of anti-bioterrorism information campaigns: a test of the situational theory. Public Relat Rev, 2008; 34:60–62.

28.

Aldoory

, Van Dyke

. The roles of perceived “shared” involvement and information overload in understanding how audiences make meaning of news about bioterrorism. Journal Mass Commun Q, 2006; 83:346–361.

29.

Aldoory

, Kim

, Tindall

. The influence of perceived shared risk in crisis communication: elaborating the situational theory of publics. Public Relat Rev, 2010; 36:134–140.

30.

Lightstone

, Swencionis

, Cohen

. The effect of bioterrorism messages on anxiety levels. Int Q Community Health Educ, 2005; 24:111–122.

31.

Meredith

, Eisenman

, Rhodes

, Ryan

, Long

. Trust influences response to public health messages during a bioterrorist event. J Health Commun, 2007; 12:217–232.

32.

Lemyre

, Turner

, Lee

JEC

, Krewski

. Public perception of terrorism threats and related information sources in Canada: implications for the management of terrorism risks. J Risk Res, 2006; 9:755–774.

33.

DiGiovanni

Jr , Reynolds

, Harwell

, Stonecipher

, Burkle

Jr . Community reaction to bioterrorism: prospective study of simulated outbreak. Emerg Infect Dis, 2003; 9:708–712.

34.

Wray

, Rivers

, Jupka

, Clements

. Public perceptions about trust in emergency risk communication: qualitative research findings. Int J Mass Emerg Disasters, 2006; 24:45.

35.

North

, Pollio

, Pfefferbaum

et al. Concerns of Capitol Hill staff workers after bioterrorism: focus group discussions of authorities' response. J Nerv Ment Dis, 2005; 193:523–527.

36.

Clarke

, Chess

. False alarms, real challenges: one university's communication response to the 2001 anthrax crisis. Biosecur Bioterror, 2006; 4:74–83.

37.

Stein

, Tanielian

, Ryan

, Rhodes

, Young

, Blanchard

. A bitter pill to swallow: nonadherence with prophylactic antibiotics during the anthrax attacks and the role of private physicians. Biosecur Bioterror, 2004; 2:175–185.

38.

Blanchard

, Haywood

, Stein

, Tanielian

, Stoto

, Lurie

. In their own words: lessons learned from those exposed to anthrax. Am J Public Health, 2005; 95:489–495.

39.

Quinn

, Thomas

, Kumar

. The anthrax vaccine and research: reactions from postal workers and public health professionals. Biosecur Bioterror, 2008; 6:321–333.

40.

Rubin

, Page

, Morgan

et al. Public information needs after the poisoning of Alexander Litvinenko with polonium-210 in London: cross sectional telephone survey and qualitative analysis. BMJ, 2007; 335:1143–1146.

41.

Morgan

, Fischhoff

, Bostrom

, Atman

. Risk Communication: A Mental Models Approach. New York: Cambridge University Press, 2001.

42.

Rogers

. A protection motivation theory of fear appeals and attitude change. J Psychol, 1975; 91:93–114.

43.

Rubin

, Potts

HWW

, Michie

. The impact of communications about swine flu (influenza A H1N1v) on public responses to the outbreak: results from 36 national telephone surveys in the UK. Health Technol Assess, 2010; 14:183–266.

44.

Vineburgh

, Ursano

, Hamaoka

, Fullerton

. Public health communication for disaster planning and response. Int J Public Policy, 2008; 3:292–301.

45.

Quinn

, Thomas

, McAllister

. Postal workers' perspectives on communication during the anthrax attack. Biosecur Bioterror, 2005; 3:207–215.

46.

Rubin

, Dickmann

. How to reduce the impact of “low-risk patients” following a bioterrorist incident: lessons from SARS, anthrax, and pneumonic plague. Biosecur Bioterror, 2010; 8:37–43.

47.

Quinn

, Kumar

, Freimuth

, Kidwell

, Musa

. Public willingness to take a vaccine or drug under emergency use authorization during the 2009 H1N1 pandemic. Biosecur Bioterror, 2009; 7:275–290.

48.

Rubin

, Amlôt

, Wessely

, Greenberg

. Anxiety, distress and anger among British nationals following the Fukushima nuclear accident. Br J Psychiatry, 2012; 201:400–407.

49.

Frewer

, Scholderer

, Bredahl

. Communicating about the risks and benefits of genetically modified foods: the mediating role of trust. Risk Anal, 2003; 23:1117–1133.

50.

Mileti

, Sorensen . Communication of Emergency Public Warnings. A Social Science Perspective and State-of-the-Art Assessment. Oak Ridge, TN: Oak Ridge National Laboratory, 1990.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.04 MB