Risk Communication Literature Review: Summary Report

Prepared by
Dr. Stephen Hill, PEng, Trent University
for the
Risk Management Directorate
Treasury Board of Canada Secretariat

Table of Contents

1 Introduction

1.1 Identifying the material to be reviewed

2 Definitions and concepts in risk communication

3 Differentiating risk communication from crisis and emergency communication

4 Key principles of risk communication

4.1 Establish a proactive and ongoing communication program between experts, risk managers, and the public

4.2 Actively plan risk communication

4.3 Ensure top-level support and clear management direction

4.4 Address organizational culture issues

4.5 Provide adequate resources

4.6 Build trust and credibility in government institutions

4.7 Share decision-making

5 Best practices for effective risk communication ("how-to")

5.1 Who is responsible to communicate

5.2 When to communicate

5.3 What to communicate

5.3.1 Developing message content

5.3.2 Dealing with confidential information

5.4 With whom to communicate

5.5 How to communicate

6 References Cited

7 Appendix A: Some factors affecting the perception of risk

1 Introduction

The purpose of this literature review is to compile and summarize information related to risk communication to support the development of risk communication guidelines as an annex to the Communications Policy of the Government of Canada.

This review specifically focused on information, themes and important aspects of the following elements of risk communication:

• Definitions and concepts,
• Principles (general guidelines),
• Practices, and
• "How to" (tools & techniques).

To accomplish this, a selection of key documents about risk communication was reviewed.

1.1 Identifying the material to be reviewed

Time restraints prevented an exhaustive review of the risk communication literature. Instead, a sample of key documents was chosen for review. These documents were selected based on:

• the quality and recentness of the content;
• their practicality and relevance to the public service; and
• the credibility, reputation, and affiliation of the author(s).

After identifying a large body of reports and guidelines on risk communication, the following were selected for more detailed review:

1. OECD Guidance Document on Risk Communication for Chemical Risk Management. July 2002. Available at www.olis.oecd.org/olis/2002doc.nsf/
   LinkTo/env-jm-mono(2002)18.
2. The Technical Basis for the NRC's Guidelines for External Risk Communication. January 2004. Available at www.nrc.gov/reading-rm/doc-collections/nuregs/contract/cr6840/.
3. Risk communication and government - Theory and Application for the Canadian Food Inspection Agency. 2001. By Jean Chartier Vice President Public and Regulatory Affairs and Sandra Gabler, Senior Science Policy Advisor Public and Regulatory Affairs. Canadian Food Inspection Agency Public and Regulatory Affairs Branch, Spring. Available at www.inspection.gc.ca/english/corpaffr/publications/riscomm/riscomme.shtml.
4. Draft Guidance Document on Risk Communication. October 1999. A companion document to the Health Protection Branch Decision Making Framework for Identifying, Assessing, and Managing Health Risks ( www.hc-sc.gc.ca/hpfb-dgpsa/hcrisk_cp_e.html).
5. Communicating Risk. UK Resilience. No date. Available at www.ukresilience.info/risk/index.htm (last updated August 30, 2004).
6. Principles of communicating risks. 1998. By Jean A. Mulligan, Elaine McCoy, and Angela Griffiths. Macleod Institute for Environmental Analysis. University of Calgary.

The results of this review are summarized under key themes identified by qualitatively contrasting and comparing the contents and ideas within each document.

2 Definitions and concepts in risk communication

Risk communication involves the two-way exchange of information between interested parties in order to make decisions about how best to manage risks. Risk communication can occur in many forms, from providing information to target audiences (primarily one way) to highly interactive stakeholder engagement and citizen dialogue (two-way).

To be effective, the management of risks by governments should be:

• integrated and built on credible assessment,
• socially and ethically acceptable to Canadians, and
• reflective of their values.

Effective risk management requires ongoing communication with the public and concerned stakeholders regarding both the characterization¹ and management of risk.

Most people will readily accept the need for public discussion about how to manage a risk. Fewer people, however, will recognize that risk characterization requires at least some value judgments by experts. The U.S. National Research Council (NRC) argues that public dialogue is essential for improving risk assessment and guarding against taking an overly narrow view of the issues (NRC (U.S.). Committee on Risk Characterization. 1996).

Notions of risk communication have evolved over the past two decades. Risk communication was originally seen as an information dissemination process (sometimes called the factual information model or empty bucket model), where the misinformed and uneducated public and concerned stakeholders were informed about the expert risk assessments and risk management decisions. This view of one-way communication from experts to the public is now seen as outdated.

Now, according to most academics, the empty bucket model should be replaced by a democratic model, focused on two-way communication rather than one-way, mutual understanding rather than exertion of power, and equal and fair participation rather than persuasion. Certainly, this view represents the dominant view of risk communication as reflected in the Canadian Standards Association Risk Management Standard (CAN/CSA-Q850-97 1997), which sees risk communication as a bidirectional information exchange that happens at all stages of the risk management process with extensive dialogue and stakeholder participation between the risk manager and concerned stakeholders.

The chronological stages of our understanding of risk communication have been described as (Fischhoff 1995; Powell and Leiss 1997):

1. focusing on the science, getting the numbers right, and simply providing 'objective' risk information (up to mid 1980's);
2. explaining, educating, and persuading the public about technical risk assessments (mid 1980's to mid 1990's); and
3. building relationships and trust through open two-way dialogue, partnership, and shared decision-making (mid 1990's forward).

While the democratic model of risk communication represents a normative ideal, it is likely naïve to believe that persuasion and the exertion of power that accompany vested and strategic interests will soon cease to be an important component of risk management and communication. Indeed, the emerging challenge of risk communication by government may well rest in creating the conditions for the ideals of the democratic model to come to fruition.

Table 1. A selection of risk communication definitions applied in certain circumstances

3 Differentiating risk communication from crisis and emergency communication

Risk and crisis communication differ with respect to the urgency for information sharing. Risk communication focuses on the possibility of events occurring in the future, where decisions about managing and preventing downside risks are the focus. Crisis communication, on the other hand, deals with an event that has already occurred where there is an immediate need for information and mitigation actions.

The main focus of crisis or emergency communication is to provide information that supports the immediate objectives of crisis management and mitigation, namely preventing major loss of life, property, and natural environments. Crisis communication plans are a distinct activity from risk communication (although they are clearly related) and should be prepared and rehearsed as a routine part of emergency preparedness.

4 Key principles of risk communication²

The review of the documents selected indicates the following principles are important to guide risk communication activities:

• proactive and ongoing;
• actively planned;
• supported by senior management;
• reflective of organizational culture;
• provided with adequate resources;
• build trust and credibility by demonstrating integrity, competence, empathy, and openness; and
• share decision-making with the public.

Each of these key principles is briefly elaborated upon below.

4.1 Establish a proactive and ongoing communication program between experts, risk managers, and the public

A proactive and special effort is required to ensure that risk communication between experts, risk managers, and the public is routine and sufficient. The OECD write (pg. 17): "In every case the best risk communication programme is proactive communication. The success of such a programme depends in part on (a) initiating it early in the cycle of concern, (b) devoting sufficient resources and attention to the task to ensure that key messages are communicated effectively to interested parties, and (c) making risk communication an ongoing and continuous activity in order to establish an atmosphere of mutual trust and respect.

The need for ongoing risk communication is well documented. Failures in risk communication - where a significant public controversy over a particular risk overwhelms efforts to manage the risk (i.e., the social amplification of risk) - are often the result of an information vacuum, where risk managers make no special effort to routinely involve the public in their assessments and management decisions. Instead, the information vacuum is filled by other sources. The partial or irregular translation of scientific information, combined with channel and media distortions, public perception (e.g., the famous "fright" factors, see Appendix A) and incorrect information creates risk communication failures.

The need to be proactive with risk communication is also well understood. At any time, any stakeholder in a risk issue can seize strategic control of the public's attention and the issue through their risk communication efforts. Stakeholders with vested interests may not serve the public interest the same way that the government should. By failing to be proactive, an agency or department may find that they lose strategic control of the risk issue.

Figure 1. A risk information vacuum, possibly filled with misinformation and/or controversy
(from Powell & Leiss, 1997)

4.2 Actively plan risk communication

While many risk communication activities will be unplanned, there is unquestionably merit in planning ahead and being prepared.

To the extent possible, the activities of risk communication should be organized along a plan-do-monitor-learn management system. For instance, the TBS Integrated Risk Management Framework (2001, pg. 33) lists the following activities for risk communication (all to be carried out within the direction of the Government Communications Policy):

• issue identification and assessment;
• analysis of the public environment (including stakeholder interests and concerns);
• development of consultation and communications strategies;
• message development;
• working with the media; and
• monitoring and evaluating the public dialogue.

Developing a risk communication plan will ideally involve both the manager(s) responsible for a risk and a communication specialist who can provide direction regarding developing a good communication plan.

4.3 Ensure top-level support and clear management direction

The support of upper management is crucial to effective risk communication (OECD, 2002, pg. 25). The Office of the Auditor General, in Chapter 1, Integrated Risk Management (2003), identified senior management support as essential to ensuring that effective risk management becomes embedded within the organizational culture rather than remaining a paper exercise. The same holds true for risk communication.

4.4 Address organizational culture issues

Organizational culture often encourages practices and activities that run counter to effective risk communication. For instance, the U.S. Nuclear Regulatory Commission (2004) found that employees preferred to hold back information until all analysis was complete. The desire was to avoid going public until all answers could be provided with certainty. Addressing organizational culture presents a major challenge to risk communication and relates directly with the need for top-level support and direction.

4.5 Provide adequate resources

A widely-held view is that governments and industry fail to devote sufficient resources or attention to good risk communication, leading to many high profile failures of risk communication and management. "The present imbalance in resources devoted to risk assessment, on the one hand, and risk communication, on the other, must be overcome if society is to realize the value of its investments in scientific risk assessment. A good shorthand rule would be to invest one dollar of risk communication effort for every dollar devoted to risk assessment" (Leiss 1999).

4.6 Build trust and credibility in government institutions

The public's trust in government institutions has been steadily eroding over the past three decades. Regrettably, the 2003 Government Communications survey found that almost as many Canadians (31%) have a low trust of the Government of Canada as have a high trust (35%) (Communication Canada 2003). It can be argued that citizens are less confident in the ability of social institutions, such as government and science, to act in their best interests. Further, citizens are more disconnected with the political process and less trusting of government institutions. Effective risk communication is a necessary - but likely insufficient - part of the ongoing effort to repair this trust and credibility.

The first step in building credibility and trust, however, is to critically evaluate whether the public's trust in your agency or department is actually warranted. According to the OECD (2002, pg. 22), the first principle of risk communication is "to start with a critical review of one's own performance" (page 6).

The OECD outlines its components of public trust (pg. 22):

• "perceived competence: degree of technical expertise in meeting institutional mandate;
• "objectivity: lack of biases in information and performance as perceived by others;
• "fairness: acknowledgement and adequate representation of all relevant points of view;
• "consistency: predictability of arguments and behavior based on past experience and previous communication efforts;
• "sincerity: honesty and openness;
• "faith: perception of 'good will' in performance and communication."

Similarly, a 2001 Canadian Centre for Management Development Action-Research Roundtable identified four components of trust and credibility in risk management (Bélanger, Baillard et al. 2001):

• integrity,
• competence,
• empathy, and
• openness.

The CCMD report noted that "[i]ntegrity, as it is commonly used, can be seen to include terms such as honesty, predictability, consistency, character, credibility, legitimacy, and dependability" (page 10).

The CCMD's second component and the OECD's top component - competence - goes directly to the ability of government to manage risk in the public's interest.

To be sure, empathy, openness and honesty are necessary to build and maintain credibility and trust - but they are not sufficient. Being a competent risk manager is critical. Widespread and repeated incompetence is the surest way to lose the public's trust. Indeed, there is a view that an organization's ability to properly manage critical moments in its history (e.g., some form of crises) is central to shaping public opinion of that organization's competence and integrity - one quickly thinks of Johnson & Johnson's famous handling of their Tylenol contamination. There also is the unfortunate dilemma posed by "tainting by affiliation" - a government's credibility can be tainted by the (perceived) incompetence of other institutions or jurisdictions. As such, the credibility of government institutions is - in many ways - a shared resource.

Although trust may be difficult and complex, at the very least, it is quite likely that the activities and efforts of integrated risk management and communication should be useful in building trust in government.

4.7 Share decision-making

The democratic model of risk communication requires two-way communication and points to the need for some degree of shared participation in decisions about risk. The degree of public engagement remains a point of debate; openness in communication is not the same as empowerment. While the tension between technocratic and democratic decision-making - dating back to Plato's notion of guardianship - will likely never be fully resolved, the prevailing view of this debate is that broad public engagement should lead to more transparent, robust, and acceptable policy decisions in the long run, even though it may seem difficult or destabilizing in the short term.

Nonetheless, advice to "engage the public and involve them in decision-making" is not particularly helpful to a manager. Every member of the public cannot be involved in every last decision: When should the public be included in decisions? How? Who should be involved? To what extent should they be included?

While there are no easy answers to these questions, a general guideline might be to involve the public and stakeholders as much as practically possible; more involvement is better.

Techniques for public engagement involvement are well known and include, among others: citizen juries, citizen panels, deliberative polling, search conferences, scenario workshops, and town halls.

Samples of 'principles for good risk communication' taken from documents reviewed

5 Best practices for effective risk communication ("how-to")

Risk communication requires a series of choices about when, what, with whom, by whom, and how to communicate. There is no universally right way to answer these questions but rather wide-ranging advice that can assist in answering each question in context.

5.1 Who is responsible to communicate

The agency or department with the responsibility to manage a particular risk in the public interest is responsible for communication about that risk. In most cases the manager(s) who deal with the risk will need to work in conjunction with a communication specialist to develop and implement a risk communication program.

However, the agency must be aware that, at any time, outside stakeholders (e.g., industry or non-governmental organizations with a clear vested interest) may take the initiative to communicate about any risk. By failing to be proactive, an agency or department may find that they lose strategic control of the risk issue.

Nonetheless, seizing the strategic advantage is difficult if the fundamental skills of risk communication and issue management are lacking. Leiss (2001) argues that, despite growing competency in risk assessment and risk management, most institutions lack analogous skills in managing risk issues (that is, in managing issues when controversy on how to best manage the risk(s) is intense) - in particular, risk communication. Resolving this problem will require a concerted effort to build risk communication skills and competencies within the public service.

5.2 When to communicate

Risk communication needs to be both proactive and ongoing. Knowing when to communicate about which risk requires an understanding of the type of risk situation. For instance, is the risk one that is well understood and defined (e.g., smoking), requiring continual communication and education to change behaviour? Or is the risk highly complex and uncertain (e.g., genetically modified foods) that raises questions regarding the ability of the institutions to deal with the risk? Or perhaps the risk is one that is likely to be controversial (e.g., terrorism) and, by its nature, will require extensive public dialogue and involvement when designing risk management programs. ³

Several schemes to classify risk situations have been offered in the literature (e.g., see Krimsky and Golding 1992). The OECD describes one typology of risk situations faced by communicators (pg. 17):⁴

• "Routine Risk Situations These risks are well known to scientists; risk managers are aware of the potential consequences and few uncertainties remain... Communication for this type of risk requires mainly the assurance that the risk is indeed a routine case and that all management organisations are well equipped to perform the necessary tasks for consumer protection...
• "Risks with high uncertainty These risks are less known and may lead to consequences that are not fully understood. Some health impacts and the full scope of environmental impacts may still be under debate. In these situations, risk managers need to address the fears of the unknown. The main goal here is to address the competence of risk management organisations to monitor impacts, to reverse decisions if negative impacts become visible, and to proceed using a precautionary approach in order to avoid irreversible damages...
• "Risks with high potential for controversy These risks may be uncertain or not, but they trigger highly controversial or emotional responses. Often public outrage is associated with these risks. The controversies are often caused by different views about the legitimacy of the product or its release. A good example may be the exposure to electromagnetic fields from mobile phone base stations. Many people feel involuntarily exposed to this risk, fear long-term health impacts and regard this risk as a violation of fairness since they might not use mobile phones themselves. Risk perception research has identified the main risk characteristics that trigger or amplify public concern and anxiety. Risk communication in highly controversial settings requires the discussion of public values, lifestyles, and world-views. Stakeholder involvement is an inevitable element of an effective communication programme, if highly controversial risks are at stake."

Routine and ongoing risk communication is required for those risks that are well understood. On the other hand, controversial risks, where public outrage can escalate, demand proactive communication. Being proactive is important because public attitudes about a risk, once formed, can be difficult to alter.

Proactive communication requires the scanning of risk issues to identify those that may "take off" in the public's mind, so that an early communication program can be put in place.

Risk situations might also be categorized according to the "fright" factors that heighten risk perception and cause outrage (see Appendix A). For instance, memorable or horrific events will stay with the public and increase the desire for information and action on a particular risk. The media play an important role in publicizing and amplifying risks (particularly because of the media's affinity for conflict, blame and "dread" news). The Walkerton water case had some of the essential features for risk amplification via the media: involuntary exposure to the hazard, harm to children, questions of blame and accountability, recognizable villains and dupes in the local water management, heroes in the form of the medical officer, and so on.

While risk communication that involved only factual evidence would not likely require extensive dialogue and empowerment, a risk issue that involved conflicting world views certainly would. Be aware that a discussion about factual evidence may obscure deeper misgivings about institutional performance and/or conflicting world views.

5.3 What to communicate

The content of a risk communication program will derive from the risk at hand. It is essential to consider your audience's needs in designing the specific content.

First, anticipate the interests and knowledge level of your audience and design your communication program to match their needs (OECD, 2002, pg. 11). Indeed, simply learning the type of information that a particular group seeks requires dialogue and communication. One technique for understanding the needs of your audience is the "mental model" approach developed by Ann Bostrom and others at Carnegie Mellon (Bostrom, Atman et al. 1994; Morgan, Fischhoff et al. 2002). Indeed, simply learning the type of information that a particular group seeks requires dialogue and communication. One technique for understanding the needs of your audience is the "mental model" approach developed by Ann Bostrom and others at Carnegie Mellon (Bostrom, Atman et al. 1994; Morgan, Fischhoff et al. 2002). Mental models reflect the public (or a particular stakeholder) view of a risk. The mental model is then used to design effective content for a risk communication program.

Second, structure your information so that simple messages are at the beginning of a communication effort and more elaborate and detailed information follow (OECD, 2002, pg. 12). It is useful to layer information so that a reader can uncover as many details about a risk as they see fit; the Internet provides a wonderful medium for this layered approach. The website Greenfacts.org provides a good example of this in action. As well, providing references and links to other organizations that deal with the same risk issue can demonstrate openness and build credibility. That is, the receiver can seek more detailed information or another opinion.

Third, be cautious with making comparisons between different sets of risk. Risk managers often compare two or more risks in an effort to convince an audience that a particular risk is acceptable by virtue of its relationship to another risk. But there are inherent dangers in this approach: comparing risks that are seen quite differently by the public (for instance, because the "fright factors" such as horror, control, etc differ between the two risks) can damage a risk communicator's credibility (Bennett, pg. 10).

Regardless of the risk communication content, improvements can be found with some form of pre-test of the material.

5.3.1 Developing message content

5.3.2 Dealing with confidential information

There are instances when information cannot be released to protect privacy, commercial interests, and national security. "If all the information at your disposal cannot be released at the time when a request for information is received:

• "release as much information as you can, as soon as you can;
• "explain why you cannot release all the information;
• "make and keep a schedule for releasing the remaining information;
• "tell those enquiring what you are doing; and
• "describe actions that may be taken by the public until more information is available" (Health Canada, Draft Guidance Document on Risk Communication, 1999, pg. 16)

5.4 With whom to communicate

In broad terms, the target audiences for risk communication include (OECD, 2002, pg 18):

• the general public;
• interested public and key stakeholders; and
• the media.

Approaches will vary depending on the target audience of a risk communication activity. Regarding the general public, most guidelines provide advice to keep message content as simple as possible without removing essential information in an effort to address less scientifically literate Canadians. However, the documents reviewed for this report do not specifically mention the difficulties in communicating about risks to less literate Canadians and those who simply are unable to read.

5.5 How to communicate

While the content may differ, the techniques of risk communication are much the same as other communication programs and insight from professional communication specialists is required.

6 References Cited

Bélanger, J., V. Baillard, et al. (2001). Building Trust: A Foundation of risk management. Action Research Roundtable on Risk Management. I. Shugart. Ottawa, ON, Canadian Centre for Management Development: 21.

Bennett, P. and K. Calman (1999). Risk communication and public health. Oxford ; New York, Oxford University Press.

Bostrom, A., C. J. Atman, et al. (1994). "Evaluating risk communications: Completing and correcting mental models of hazardous processes: II." Risk Analysis Vol 14(5): 789-798.

CAN/CSA-Q850-97 (1997). Risk management: Guideline for decision-makers. Etobicoke, Ont., Canadian Standards Association, Standards Council of Canada.

Communication Canada (2003). Listening to Canadians. Catalogue Number: PF4-7/2003-1-2. Ottawa, On, Government of Canada.

Fischhoff, B. (1995). "Risk perception and communication unplugged: Twenty years of process. Addressing agencies' risk communication needs: Next steps (1994, Annapolis, Maryland)." Risk Analysis Vol 15(2): 137-145.

Krimsky, S. and D. Golding, Eds. (1992). Social Theories of Risk. Westport, Connecticut, Praeger.

Leiss, W. (1999). The Importance of risk communication in the risk management of chemicals. Newsletter of the International Council on Metals and the Environment [ICME]. 7.

Leiss, W. (2001). In the chamber of risks: Understanding risk controversies. Montreal, McGill-Queen's University Press.

Morgan, M. G., B. Fischhoff, et al. (2002). Risk communication: a mental models approach. Cambridge ; New York, Cambridge University Press.

National Research Council (U.S.). Committee on Risk Characterization. (1996). Understanding risk: informing decisions in a democratic society. Washington, D.C., National Academy Press.

Powell, D. A. and W. Leiss (1997). Mad cows and mother's milk: the perils of poor risk communication. Montreal, McGill-Queen's University Press.

7 Appendix A: Some factors affecting the perception of risk

1 In their book Understanding Risk: Informing Decisions in a Democratic Society, the National Research Council (U.S.). Committee on Risk Characterization defines risk characterization as a synthesis and summary of information about a hazard that addresses the needs and interests of decision makers and of interested and affected parties. Risk characterization is a prelude to decision-making and depends on an iterative, analytic-deliberative process.

2 For this report, a principle will be defined as a basic truth or assumption about risk communication practices. A principle represents a widely held belief - i.e., those themes that were commonly mentioned in the documents reviewed - about the appropriate course of action for risk communication. The "best practices" section of this report outlines aspects related to implementing these principles. Nonetheless, there is undoubtedly overlap and grey areas between a principle and best practice.

3 Of course, the aftermath of a terrorist action represents crisis communications.

4 The three types of risks are derived from Ortwin Renn's three levels of risk communication activities: providing factual evidence and probabilities (routine risks); debating institutional performance, expertise, and experience (complex risks); and discussion of conflicting conceptions about worldviews and value systems (controversial risks).