Explaining Risk Analysis

Protecting health and the environment

Michael Greenberg

• Publisher: Routledge
• Year: 2016
• ISBN:
  9781317298410
• Edition: 1
• Topic: Economics
• Subtopic: Sustainable Development
• Index: Economics

Risk analysis is not a narrowly defined set of applications. Rather, it is widely used to assess and manage a plethora of hazards that threaten dire implications. However, too few people actually understand what risk analysis can help us accomplish and, even among experts, knowledge is often limited to one or two applications.

Explaining Risk Analysis frames risk analysis as a holistic planning process aimed at making better risk-informed decisions and emphasizing the connections between the parts. This framework requires an understanding of basic terms, including explanations of why there is no universal agreement about what risk means, much less risk assessment, risk management and risk analysis. Drawing on a wide range of case studies, the book illustrates the ways in which risk analysis can help lead to better decisions in a variety of scenarios, including the destruction of chemical weapons, management of nuclear waste and the response to passenger rail threats. The book demonstrates how the risk analysis process and the data, models and processes used in risk analysis will clarify, rather than obfuscate, decision-makers' options.

This book will be of great interest to students and scholars of risk assessment, risk management, public health, environmental science, environmental economics and environmental psychology.

1
RISK ANALYSIS

A start

Introduction

I introduce risk analysis to students and public audiences in three steps over 10 to 15 minutes. First, I ask them to:

1. Close their eyes, and think of the first three human health and environmental hazards, risks, and events that concern, worry and scare them the most;

2. After a minute, I ask them to open their eyes and write these three items on a piece of paper or their computer; and

3. We discuss the items on their list.

With rare exceptions, nuclear war tops their list, followed by a nuclear power plant accident, a virus or bug that causes a pandemic, and sometimes hurricanes, earthquakes, tornadoes and other natural hazards that they have lived through or much more likely seen on television. All of these hazard events are low probability but potentially high consequence. Few in the audience pick alcohol or drug abuse, tobacco, engaging in dangerous physical behaviors, living in a home with flaking lead or asbestos particles and without safe stairs, and other common and potentially consequential hazards. The point of this initial exercise is to get them to recognize that low probability and high consequence events rise to the top of the concern list.

My second step is to use the first discussion to motivate them to call for analytical and replicable approaches to assess risk and thoughtful, principled analysis to inform decision-making processes. This second step requires a commonly accepted language of key terms and concepts. The Society for Risk Analysis (SRA) (www.sra.org) has published several glossaries. One available through the website includes short definitions of more than 300 terms. The second produced by the SRA Committee on Foundations of Risk Analysis divides the glossary into “basic concepts,” “related concepts, methods, and procedures,” and “risk management actions.” It acknowledges that it is unrealistic to assume that there will be a single definition of these terms and concepts (Committee on Foundations 2015). The second glossary provides multiple definitions of the word risk and other terms, and I am a strong advocate of using it because of the depth it provides.

I have collected more than 20 definitions of risk written by individuals with backgrounds in public health, engineering, economics and others. In this book, I use the following definition of risk developed by William Lowrance (1976), and arguably the most widely cited: Risk is a measure of the probability and severity of adverse events.

In regard to the definition of risk analysis, as part of the celebration of the thirtieth year of the Society of Risk Analysis in the year 2010, a group of the editors of the society’s journal, Risk Analysis, An International Journal summarized major accomplishments (Greenberg et al. 2012). Their effort was organized around the following six risk analysis questions:

1. What can go wrong? (hazard event);

2. What are the chances that something with serious consequences will go wrong? (likelihood);

3. What are the consequences if something does go wrong? (consequence);

4. How can consequences be prevented or reduced? (prevention);

5. How can recovery be enhanced, if the event occurs? (resilience); and

6. How can key local officials, expert staff, and the public organize and be informed to reduce risk and concern and increase trust and confidence? (organization).

The first three of these six questions are about risk assessment, the second set of three are risk management ones, and the six constitute risk analysis. There is nothing magical about two pairs of three questions each, that is, there are more nuanced versions of each of these six (Cox et al. 2008; Kaplan, Garrick 1981; Garrick 1984; Chankong, Haimes 2008; Greenberg et al. 2012). Whatever their imperfections may be, taken together the six questions constitute a systematic approach to try to understand risk and to use that knowledge to reduce vulnerability and potential harm.

Beginning with the first question, hazard events are naturally occurring and human actions that can occur and may trigger consequences. A slow leak of an underground gas line that no one sees, smells, or hears until a fire/explosion occurs is a hazard event, as is a major earthquake that everyone in the area immediately feels and hears. Likelihood (e.g., probability, chance) is a measure of certainty of an event. Sometimes the likelihood numbers are precise, such as 6 percent percent with 95 percent confidence limits between 3 percent and 12 percent, and other times the estimates are in categories with subjective breaks in the classification (almost none, low, medium and high). Consequences of hazard events include impacts on human health, ecological systems, social organizations, and economies. Consequences range from not noticeable to catastrophic world-ending.

Prevention consists of risk management options to stop consequential events from happening. The main focus should be to reduce vulnerability, which is a state in which a person, an institution or a physical system is susceptible to damage that can lead to human health, economic, environmental, and social consequences. With regard to gas leaks, for example, monitoring and physically inspecting gas lines and replacing vulnerable ones are prevention policies; and not using equipment that causes sparks until you know where the gas line is a prudent step. Education is a key part of prevention. At least once or twice a week, I hear a radio announcement directing listeners to call 911 and evacuate if they see, hear or smell any signs of a gas leak. They may be wrong, perhaps there is no gas leak, but assuming that there is a gas leak, evacuating prevents deaths and injuries.

Resilience is the capacity of a system to continue operating under stress and restore normal functions or as close to normal as possible. Rerouting gas around a leak can help limit the damage to a small area, and fixing a leak can restore function. Sometimes, unfortunately, return to the previous normal is infeasible and a new normal emerges, which may mean displacement for a long period of time and inconvenience at a minimum. Many readers know at least one person who has lived through some of the massive hurricanes, earthquakes, and tornadoes during the twenty-first century and is still displaced from their previous residence (Greenberg 2014).

A successful organizational response requires high quality, dedicated personnel and budgetary resources. Cooperation among individuals and organizations is essential, including formal agreements among units and practice. Given trends that suggest pressure to reduce current levels of personnel and budgets, cooperation will become more critical (see Chapter 11).

As I indicated in the preface, agencies and companies have their own versions of the six risk analysis questions, and many of these are illustrated in this book. The major criticism I have heard about the six questions is from some government and private organization representatives. Their view is that risk assessment and risk analysis are part of risk management, not that risk assessment and risk management are part of risk analysis. For example, Reinschmidt et al. (2005), in a project for the U.S. Department of Energy (DOE), asserted that identification and analysis of project risks are part of risk management. (See also the International Organization for Standardization (ISO) in their ISO 31000:2009 ISO Guide 73:2009.)

Are these differences important? They should not be because a good risk analysis will include identification of possible events, likelihood, consequences, as well as prevention, resilience, and organizational capacity. Yet, these differences in the role of the elements of risk analysis could be important if managers direct that only hazards required by law to be studied are to be evaluated, and thereby hazards that scientists and the public believe might be a threat are ignored. It is important that risk assessment scientists maintain their capacity to explore an unlimited range of hazards. Also, differences in what is considered risk analysis practice can be important if those in charge mandate that analysts should automatically default to checklists and categorical risk categories, which may obscure important information and curtail opportunities for deep thinking about causes and effects.

Overall, I use the six risk analysis questions as the organizing framework for the book. I have included examples that were driven by risk assessor findings and others driven by risk manager necessities. The later group include challenges from public groups. While cases are primarily from the United States, I have introduced an international perspective in every instance, especially in regard to risk management perspectives, which can be quite different among and within countries.

Key probes, cause-and-effect and uncertainty in risk analysis

Understanding cause-and-effect and reducing uncertainty are the two most important academic challenges faced by risk analysts, and they are part of every chapter in this book and every risk analysis you will ever read. In order to sensitize you to their importance, I have provided several brief notes about them before you get deeply into the book. First, I provide a checklist (Table 1.1) that I use to manage my entry into every risk analysis case, and then a set of cause-and-effect questions (Table 1.2) that I have been applying to my work for many years (see Cox 2013 for a much longer discussion of cause-and-effect in risk analysis). Then, I briefly discuss how analysts deal with information gaps. Even though all the checklist and the cause-and-effect questions, as well as the suggestions about uncertainty are not discussed in every chapter of this book, they should be addressed in practice, where there are likely to be consequences for ignoring them. The checklist reminds me not to forget any key issues, and frankly I have (e.g., cascading effects), which is why I prepared the simple checklist, and each of these short checklist questions has additional follow-up probes as needed. Once I have some answers to the checklist questions, then I face, sometimes struggle with, the reality of the causes and effects implied in my initial answers to the checklist questions. Third, managing the risk analysis requires making decisions based on rules, guidelines and practice about managing uncertainty.

Answering these checklist, cause-and-effect and uncertainty issues is a challenge, in some cases a nightmare-causing one, and yet an opportunity to add clarity to the policy process. Tasked with providing support for decision-makers who are not necessarily very interested in science, cause-and-effect relationships, and have little patience for details, risk analysts try to deal with real issues in real time. They use rules or develop guidelines that help them choose the best data sets. They explore and choose mathematical models that fit the available data, typically beyond the range of the data, and they add conservative confidence limits in the name of coping with deep uncertainty and adding a margin of protection.

When the databases are too few and/or poor quality, outside experts are invited to participate. But even expert knowledge elicitation can produce arguable results about what are the primary and secondary causes, and about how much certainty we can have in relationships, and even what is most important to know. ...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Dedication
• Table of Contents
• List of figures
• List of tables
• Preface
• Acknowledgments
• 1 Risk Analysis
• Part I Basics
• Part II Cases
• Part III Supplements
• Index