Hazard

PART I
FUNDAMENTAL CONCEPTS
IN RISK MANAGEMENT

FUNDAMENTAL CONCEPTS
IN RISK MANAGEMENT
1. HAZARD, RISK, AND RISK MANAGEMENT





A thing is safe if its attendant risks are deemed
to be acceptable.
This insightful phrase, coined by William Lowrance in 1976,
constitutes the bedrock of modern risk management theory and practice
when dealing with hazardous chemicals and other environmental
contaminants in our contemporary society. But what exactly does it
mean?
Clearly we all desire a reasonable degree of safety with respect to our
personal health and the health of our loved ones. But how willing are
we to accept the fact that no activity in life is absolutely safe under all
circumstances? Most of us will acknowledge that life presents us with
many risks of uncertain probability and severity, and that a risk-free life
would invariably be one that is impoverished, unexciting, and
ultimately futile. So we often accept certain risks, some of them quite
cheerfully, when they are associated with personal benefits to our
lives—especially those that contribute to the betterment of our
employment prospects, material prosperity, and recreational enjoyment.
Many of us consider voluntary exposure to hazardous situations as
acceptable whenever the expected benefits outweigh the perceived
risks, assuming of course, that we retain a free choice in deciding
whether or not to subject ourselves to these hazards.
On the other hand, we will barely tolerate or actively oppose risks
beyond our personal control, regardless of the anticipated degree of
risk. Such involuntary risks include those created by the wide array of
man-made chemical pollutants that seem to pervade our natural
environment. We encounter these involuntary risks reluctantly, with a
sense that we accrue no detectable benefit for ourselves, while
worrying that our health might be seriously impaired, even to the point
of premature death. Ideally, we seek to avoid these risks absolutely. So
this type of risk we want minimized, or preferably eliminated



altogether. In fact, many of us tend to be skeptical about statements
made by those in government and industry who assure us that
something is absolutely safe, and that ‘nothing can possibly go wrong.’
But how feasible are strategies dedicated to the total elimination of
toxic hazards in the environment? Some short-lived substances of
minimum commercial value can indeed be successfully banned from
production, import, and use. But to the extent that many hundreds of
chemicals play an essential role in modern industrial societies, their
continued use will require the implementation of environmental
controls that are strict enough to reasonably ensure the health of
humans and other living things, while not imposing an unnecessary
burden of production limitations and costs of control for the affected
industries.
Thus, a realistic definition of environmental safety will often entail
incurring a very small risk to personal health—one that is commonly
deemed acceptable or tolerable because it produces a relatively
insignificant probability of disease incidence when compared to other
more prevalent disease states or compared to other more important
disease causes. The risk management process is the means by which
governments and other standard-setting organizations seek to define a
rational level of acceptable or tolerable risk for an environmental
hazard—by considering the severity and probability of harmful health
effects, the amount of environmental exposure experienced by human
populations, the sources and means of control for the contaminant, and
the expected costs and benefits of various risk reduction strategies.
Like all decision-making systems, risk management has its particular
strengths and weaknesses. Its greatest strength is thought to be the
scientific rigour with which the risk management process applies
logical and consistent methods to arrive at sound decisions for
technically complex problems. Its greatest potential weakness lies in
the mindless lock-step application of formalized rules for risk decisionmaking, without making adequate allowance for human concerns
beyond the scope of established scientific criteria. The seemingly
incomprehensible technical jargon employed in the production of risk
management reports is also a common problem. The former trait leads
to technocratic decisions that often do not jibe well with social reality,
while the latter encourages obscurity rather than clarity in
communicating how decisions were achieved.
In certain instances, the decision tools of a risk management system
may even be misused, with possible serious consequences for
population health or for the economic sustainability of a segment of
society. It has been said that democracy is the least effective form of
government for managing a country, except for all the others. Like
democracy, risk management derives its greatest advantage, not from
technocratic rationality, but by its ability to organize and examine







scientific and socioeconomic information in a public forum open to free
communication and debate by all concerned parties. In this sense, risk
management should be viewed as means of promoting public
accountability by communicating information about risk among
stakeholders, with the objective of producing decisions that include
both sound science and prudent decision-making.


!
"

Risk and hazard are two distinct, but interrelated, concepts. A hazard
represents a chemical, physical, or biological substance that has the
potential to produce harm to health if it is present in the environment
and comes into contact with people. The hazardous properties of an
environmental agent are defined according to the nature and severity of
its harmful consequences. Fortunately, many hazards can be either
contained or avoided, so not every potential environmental hazard
poses an actual health risk. A risk, in turn, is defined as the likelihood
of adverse health effects arising from exposure to a hazard in a human
population, which is conceptually expressed as the product of two
factors—the probability of exposure and the severity of the
consequences.








The degree of risk can be determined in one of two ways. Some risks
can be measured directly by observing past and present disease
incidence patterns in the human population. Alternatively, risks can be
calculated indirectly, by estimating the theoretical level of human
exposure and the potential severity of health effects as predicted by
experimental studies. The health risks from low-level exposure to
environmental hazards are commonly determined by the indirect
method, because there is not enough consistent and reliable evidence of
measurable health effects in human populations exposed to low levels
of hazardous environmental agents. It must be recognized that the risks
from low-level environmental exposure are theoretical calculations, not
observed facts. So one must exercise considerable caution when
employing such risk predictions for controlling hazards in the
environment. This is why risk assessment and risk management
decisions are both difficult and important.
The risk of illness or death arises in everyday life from many potential
sources—physical risks from accidents and violence, health risks from
infectious microbes and environmental pollutants, or medical risks
from the adverse effects of pharmaceutical drugs and medical devices.
Most people want to exert some reasonable degree of control over the
important risks in their lives. They use their personal experiences or the
insights of knowledgeable professionals to estimate the nature and







Hazard is commonly defined as “the
potential to cause harm”. A hazard can
be defined as “a property or situation
that in particular circumstances could
lead to harm.”
…
Risk is a more difficult concept to
define. The term risk is used in
everyday language to mean “chance of
disaster”. When used in the process of
risk assessment it has specific
definitions, the most commonly
accepted being “The combination of the
probability, or frequency, of occurrence
of a defined hazard and the magnitude
of the consequences of the occurrence.”
Royal Society (U.K.), 1992

degree of risks from various sources; they evaluate the level of risk that
they are prepared to accept or tolerate; and they try to control risk by
balancing acceptable risk against the perceived benefits for a particular
course of action.
These three activities—risk estimation, risk evaluation, and risk
control—collectively comprise the core of the decision process that we
call risk management. In this sense, people continually undertake risk
management in many aspects of their everyday lives. To the extent that
we usually consult with other people who might be affected by our
decision, before making important risk choices, the process of risk
communication is also an essential component of risk management.

$%

&'

!


"
!(
)
!
*%&


In the context of environmental health, the risk management process
can be organized into several distinct activities. The three core
activities that constitute the essential decision-making steps in the risk
management process are each involved in examining different aspects
of the risk problem:
risk estimation
the use of science-based risk information and analytical methods to
characterize the nature and extent of environmental health risks in
the human population;
risk evaluation
consideration of the economic, social, political, and legal factors
that influence a decision to adopt a particular course of action to
reduce health risks—in some risk frameworks, the quantitative
economic analysis of the benefits and costs of risk reduction is
combined with results of the risk estimation process, so that a ‘risk
assessment’ may subsume part or all of risk evaluation;
risk control
the selection of options and the commencing of actions intended to
reduce risk to an acceptable or tolerable level; this activity is often
referred to as “risk management,” but the term “risk control” is
more specific and better reflects the objectives of the activities it
denotes.
These core processes are conventionally arranged in an ordered
sequence of steps, so that risk estimation is typically thought to come
first, risk evaluation second, and risk control third. This stepwise
arrangement is intended to ensure that the source of information in risk
decision-making flows primarily from well-validated scientific studies,
before moving onto the more value-laden consideration of





#

In addition to the three major steps,
there may also be one or more
preliminary stages comprising the
initiation of the review process, and
the preliminary analysis of available
scientific and technical information;
as well as a terminal stage of
implementation and monitoring of the
actions aimed at risk reduction.

socioeconomic factors and technical control options. In most countries,
with the notable exception of the United States, each of the steps is
intimately connected to the others by a continuous two-way flow of
communication occurring in an iterative manner.
An effective risk management process also requires ongoing
consultation among all concerned parties, or stakeholders, to resolve
questions of policy, science, and societal concerns. A stakeholder is
defined as a person or group that could be affected or potentially
affected by a specific health risk, any groups or organization that would
be affected by efforts to manage the source of the risk, and the risk
managers themselves. The risk communication process provides an
essential coordinating function, by supporting the activities of
information exchange and mutual consultation among various
stakeholders—governmental, non-governmental, and private-sector
organizations—throughout the various phases of the risk management
process. Risk communication activities represent the organizing
principle which links stakeholders together to enable a coherent
decision-making process.

"
!(
)
!
,)
-%"

Many organizations with responsibility for managing environmental
health risks have adopted some form of risk framework to organize the
activities needed to support the development of sound environmental
strategies, and to improve the understanding of the stakeholders. The
risk management process adheres to this systematic decision-making
framework, whose activities are arranged as an ordered sequence of
steps. While a risk framework describes the steps that decision-makers
should follow to arrive at a final environmental control strategy, the
framework is not intended merely to ensure procedural correctness or
to serve as a workflow diagram. A well-designed risk framework
specifies all the component activities of the risk management process,
their interconnections, the flow of information, and the critical decision
points within the process.
Risk frameworks have been devised by several organizations in
Canada, the United States and elsewhere. All are intended to provide a
structured approach to health risk assessment, evaluation, and
management. While various risk frameworks generally agree about
broad principles, some differences do exist in risk terminology, the
level of detail, the importance of factors such as risk communication,
and the involvement of stakeholders in the overall process.
Although at present there is no official Canadian framework for risk
management, a consensus guideline for risk management has been
recently developed by the Canadian Standards Association, entitled
CAN/CSA-Q850-97 Risk Management: Guideline for Decision-Makers.





+

A risk management framework is a
formal method for assessing and
managing health risks. The
framework that this transition team is
working on will give all of Health
Protection Branch (HPB) a common
basis for addressing risk assessment
and management.
HPB transition plan -- The revised
framework will provide a basis for
Health Canada's strategic approach
to public health risk management,
and provide the foundation for
development of Departmental
policies on risk management. To the
extent possible, the framework should
be consistent with other national and
international frameworks which are
relevance to programs within the
Health Protection Branch. The
framework will provide a structured,
consistent, and comprehensive
approach to risk assessment and
management within the Branch, but
at the same time, allow for flexibility
in implementation. It will also help to
identify the roles, responsibilities,
and interactions of partners and
other stakeholders in the risk
assessment and management process.
INFORMATION October 1998
Health Protection Branch
Branch-Facts

An analogous risk framework, Risk Determination: A Model for Risk
Assessment and Risk Management, has been developed by the Health
Protection Branch (HPB) of Health Canada.
In the United States, the earliest and most elaborate risk framework for
environmental health was developed by the U. S. National Research
Council (NRC) in 1983. The 1983 NRC report Risk Assessment in the
Federal Government: Managing the Process defined the major stages
of the risk assessment and risk management process, and described, in
great detail, the structure and logical development of the risk
assessment process. The risk assessment process is expected to deal
exclusively with numerically quantifiable scientific data and statistical
methods used to analyze that data, i.e., Risk Identification and Risk
Estimation issues. Thus it is commonly asserted that risk assessment
methods rest firmly on the foundations of ‘sound science’, in
accordance with the principles of scientific objectivity, absence of bias
or prejudice, and apolitical approaches to appraising evidence. Such
worthy attributes provide the strongest rationale for a fundamental
reliance on risk assessment in guiding environmental policy-making.
In addition to providing a structured guideline for decision-making,
some risk frameworks also allow the flexibility needed to address
specific health hazards on a case-by-case basis. In the United States, the
use of 1983 NRC framework for the risk assessment and management
of environmental contaminants has become institutionalized over time,
serving as an official protocol with an important legal status in
regulatory decision-making. In practice, such a legalistic approach in
the development of environmental standards has created a relatively
inflexible decision-making structure not originally envisaged by its
proponents. Risk frameworks in most European Community countries
tend to be more flexible; they are accorded formal recognition by
governments, but are typically used as guidelines rather than
regulatory protocols. The use of risk frameworks by Health Canada,
Environment Canada, and other Canadian regulators is usually cited by
foreign observers as representing a reasonable middle ground—less
cumbersome than the U.S. EPA’s legalistic risk framework, and more
consistent than the consensus-based decision-making processes
employed by many European organizations.





In Canada, a broadly-applicable risk framework, CAN/CSA-Q850-97
Risk Management: Guideline for Decision-Makers, has recently been
developed and approved under the auspices of the Canadian Standards
Association (CSA). The Q850 framework is based on the collaborative
efforts of representatives from government, industry, research, and
environmental organizations. The most important of the major steps in
the Q850 risk framework will now be described.






.

The NRC (1983) report has in turn
provided the conceptual and
operational foundation for the later
development of a series of risk
assessment guidelines, most notably
the series of EPA guidelines for the
risk assessment of environmental
carcinogens, mutagens,
developmental toxicants (teratogens),
male and female reproductive
toxicants, immune system toxicants,
and neurological toxicants.


0
1
!%!


The Risk Identification step assesses the likelihood that an
environmental agent might constitute a potential health hazard,
based on its physico-chemical properties, its toxicological effects in
test animals, and its observed human health effects. Because the
terms risk and hazard are often used interchangeably in different
parts of the world, both risk identification and hazard identification
denote the same type of activity within the Q850 framework.

0
2
"
)%!

The Risk Estimation step attempts to quantify the amount of risk
incurred by various levels of exposure to environmental
contaminants. The Risk Estimation stage is a multiphasic process
which consists of the following three activities:
Dose-response assessment
The dose-response assessment step estimates the probability of
occurrence of harmful effects that might arise from a specified
level of exposure to a substance, based principally on
epidemiological studies in human populations and toxicological
studies in test animals.
Exposure Assessment
Exposure assessment determines the expected patterns and levels
of exposure to a substance in various segments of the population,
including estimates of the dose level, duration of exposure, and
continuity of exposure over time for various environmental
media—particularly air, water, soil, and food. Of special concern is
the exposure incurred by possibly susceptible subgroups within the
population—the young, the aged, women of reproductive age,
those with pre-existing health problems, and visible minorities or
disadvantaged sectors of society.
Risk Characterization
This step combines the qualitative findings about the nature of a
hazard as determined by the risk identification (hazard





/

stakeholders


0
2
*
3)!
!3
4
"

! &%!


$567+8
,)
-%"


%
"
!(
)
!


Risk Communication

This first step defines the context and organizational structure under
which a specific risk management problem will be resolved, including
such issues as: the scope of the problem, the terms of reference under
which the problem will be addressed, the concerned parties or
stakeholders who will be invited to act as participants in the risk
management process, the decision-making bodies responsible for
resolving the problem, the legislative and legal mandates for
anticipated regulatory actions, and the time frame under which the
process will operate.

Initiation

Preliminary
Analysis
Decision

Risk
Estimation
Decision

Risk
Evaluation
Decision

Risk
Control
Decision

Action/
Monitoring

identification) step, together with the quantitative estimates of
harm as determined by the dose-response assessment step and the
exposure assessment step. The risk characterization specifies the
likelihood of harmful health effects for various levels of exposure
within a range of likely exposure scenarios. The risk
characterization provides a well-documented summary of the
critical findings of the risk assessment process, which serves as the
scientific basis for decision-making in the later risk evaluation and
risk control stages. Ideally, it also serves as the basis for effective
risk communication with scientists, regulators, stakeholders, and
the general public.

0
#2
"
'39%!

The major issues to be addressed at the risk evaluation stage include
social factors, economic factors, political factors and legal factors. A
balancing between the costs of control and the predicted health benefits
from reduced exposure are estimated, informally by consensus, or
analytically by cost-benefit analysis and similar economic methods.
Factors not readily quantifiable, such as the societal acceptance of a
risk in affected groups, and the legal and political aspects of regulation
within existing federal-provincial jurisdictions are also reviewed.

0
+2
"
$%!%3

Under the Q850 framework, the Risk Control step consists of several
major activities: Identifying Feasible Risk Control Options, Evaluating
Risk Control Options, and Stakeholder Assessment of Options. This
involves a process of evaluating alternative regulatory and
non-regulatory options and selecting the most appropriate option. The
option selection task entails the use of value judgments on such issues
as acceptability of risk and the reasonableness of the cost of control.

0
.2
)03
)
!%!
!
&%!4%!%!(

This final step includes the implementation of regulatory and voluntary
actions, and monitoring of the compliance with and effectiveness of the
actions.
33

02
"
$%))9!&%!

In addition, the Q850 framework emphasizes the critical role of Risk
Communication process in the decision-making process. In earlier
frameworks, this activity has often been overlooked, or treated merely
as an information dissemination process, intended to explain and justify
risk management decisions to concerned stakeholders and the general
public. In contrast, risk communication under the Q850 framework is
conceived as a bidirectional consultative exchange that works at
virtually all stages within the risk management process. Risk
communication involves extensive stakeholder participation,
encouraging of an ongoing dialogue between the risk managers and
concerned participants regarding all aspects of the entire risk





7

management process. Its primary objective is to mutually inform, not to
persuade or bargain, nor to act as a public relations vehicle. Risk
communication activities thus run parallel to the entire spectrum of risk
assessment, risk evaluation, and risk control activities.
The Q850 framework also includes a series of discrete decision points
interposed between each of the major stages. When critical information
is deemed adequate for decision-making, an informed judgement can
be passed to the following stage in the form of a summary of findings
or as a recommendation for action. Conversely, when inadequate
information is available, a query can be referred back to a previous
stage for further data collection and analysis. This type of risk decision
system is therefore constructed as an iterative process—one which can
be repeated, if required, as long as essential information or analysis is
missing.
Accordingly, any decision step in the framework can be revisited
whenever additional information or analysis is required, and the entire
process can be restarted when new information becomes available at a
later time. The multiple decision points also allow for intermediate
decisions to terminate the risk management process when existing
information indicates that further data collection or decision analysis is
unwarranted.
For example, if a newly detected environmental contaminant were
found to be produced exclusively by natural processes which were not
amenable to control measures, the risk management process could be
terminated at the risk identification stage. Similarly, if a decision were
made by a manufacturer to permanently withdraw a potentially toxic
substance from production and distribution, then the decision process
concerning possible regulatory actions could be terminated at the risk
control stage.

*;
"
,)
-%"

The Health Canada risk management framework, Risk Determination:
A Model for Risk Assessment and Risk Management, was developed in
the early 1990’s by the Health Protection Branch (HPB). It has served
as a guideline to assist Health Canada in protecting Canadians against
environmental hazards—such as chemical pollutants and food
contaminants—and other public health activities to control disease and
injury.
In the environmental health arena, the influence of the HPB risk
framework carries beyond the confines of Health Canada, as this
agency also plays an advisory role for health risk assessment of
environmental substances regulated by several other departments, most
notably trans-boundary environmental pollutants (with Environment
Canada) under the Canada Environmental Protection Act (CEPA).





:

The Health Protection Branch (HPB)
Transition will result in a number of
proposals to renew the health
protection program:
- a decision-making framework and
process that is transparent and open
to public and stakeholder input;
- a legislative framework that
provides the necessary authorities to
protect the health of Canadians;
- a comprehensive Risk Management
Framework that is clear on the
accountability of HPB scientists and
managers;
INFORMATION October 1998
Health Protection Branch
Branch-Facts

The HPB framework reflects several significant differences between
Canada and the United States in government decision-making practices.
Most notably, the distinction between risk assessment as an objective
scientific enterprise and risk evaluation as a subjective non-scientific


*;
"
,)
-%"
9


process is less strongly emphasized in Canada than in the United
<

3
$! 

States. The HPB framework reflects a sense that a prescriptive
scientific methodology for risk assessment inevitably contains hidden
value assumptions and inherent societal biases. Hence,
Risk Determination: A Model for Risk Assessment and Risk Management (HPB, 1990)
any attempts to place a ‘firewall’ between the
scientific analysis and the consideration of societal
concerns is deemed neither desirable nor entirely
Hazard Identification
achievable in a consensus-driven regulatory regime
Risk Analysis
such as the Canada’s. Under the HPB framework, Risk
Risk Estimation
Assessment is structured to include both consideration
of scientific evidence in a Risk Analysis step, and
Risk Assessment
analysis of socioeconomic concerns in an Option
Development
of Options
Evaluation stage. The HPB Option Evaluation step is
generally equivalent to the Q850 Risk Evaluation step,
Option Evaluation
and is treated by HPB as part of the overall Risk
Option Analysis
Assessment process.
The HPB framework also interposes a Decision step
between the risk assessment and risk management
phases. The decision step is necessary to connect the
risk assessment activities with the risk management
activities through a decision pathway. Initially, the
Risk Management
decision step was represented by a blank box, which
contained no defined activities. Later versions of the
HPB framework include some specific provisions for
the decision step. A possible weakness in the HPB
framework is the assumption that only one decision
point exists in the flow of risk information. In
practice, however, incremental decisions within the
Health Protection Branch are typically made at several points
throughout the risk assessment and risk management process.

$
"
,)
-%"

One of the earliest and most influential risk management frameworks
was devised in 1983 by the U.S. National Research Council, and is
commonly referred to as the NRC “Red Book.” This framework was
presented in the report entitled Risk Assessment in Regulatory
Decision-Making: Managing the Process.
The NRC framework states unequivocally that scientific evidence and
political considerations should not be allowed to mix, and that the
scientific aspects of risk assessment should thus be completely
separated from the non-scientific aspects of risk management.
Adherence to “good science” as the foundation of the risk management





8

Decision

Implementation

Monitoring and
Evaluation

Review

Scientists assess a risk to find out
what the problems are. The process
of deciding what to do about the
problems is risk management...
Despite these often conflicting
pressures, risk assessment at EPA
must be based on scientific evidence
and scientific consensus only.
Nothing will erode public confidence
faster than the suspicion that policy
considerations have been allowed to
influence the assessment of risk.
William Ruckelshaus (1983), EPA
Director, addressing National
Academy of Sciences.

process is therefore deemed paramount. Without sound science in risk
assessment, it is thought that credible decision-making cannot be
achieved at the later stages of the risk management process.
The U.S. EPA states that its own risk assessment guidelines “… are
similar to those proposed by the NRC” and it specifically endorses the
NRC risk assessment terminology. As adopted and implemented for
regulatory standard-setting by the U.S. EPA, the NRC framework is
broadly viewed as a scientifically exhaustive, procedurally rigourous,
and highly transparent means of deriving environmental health
standards within the highly legalistic regulatory requirements of U.S.
government system. However, EPA-style risk management has been
frequently criticized as being excessively prescriptive in defining the
procedural norms for conducting risk assessment of scientific evidence
and as overly rigid in government decision-making, leading in many
instances to over-regulation or to the establishment of excessively
restrictive environmental exposure limits.
$
=:7>
"
,)
-%"

 %0

<
?22
*


RESEARCH

RISK ASSESSMENT

RISK MANAGEMENT

Development of
regulatory options

Laboratory and Field
Observations
Toxicity assessment:
hazard identification
and dose-response
assessment
Information on
extrapolation methods

Risk Characterization
Research needs identified
from risk assessment
process

Field measurements,
characterization of
populations

Exposure assessment;
Emissions
characterization

The risk management process in Canada is considerably more open to
informal information exchange between stakeholders during the various
steps of the process, in contrast to the United States—where the risk
assessment step for environmental hazards has been isolated from the
later steps by statutory or administrative restrictions on information
flow. However, the United States government is now actively pursuing
regulatory reform by moving gradually towards a more flexible and
better interconnected risk management system.







Evaluation of public
health, economic,
social, political
consequences of
regulatory options

Agency decisions
and actions

To a limited extent, such changes are already taking place. For
example, the 1993 Executive Order 12866 has proposed a new
regulatory plan which establishes principles concerning the analysis of
risks, benefits, and costs of revised environmental regulations. The new
regulatory plan would alter the current methodology for risk to allow
for the economic evaluation of risks, benefits and costs within the risk
assessment process. This marks a major policy shift, in that it reduces
the ‘separation’ between risk assessors and risk managers within the
U. S. regulatory system.


&%!5)"!(
!
"
!(
)
!

We often incur risk as the inevitable consequence of making decisions
with incomplete and uncertain information. It is relatively easy to make
decisions about risks in situations where reliable information is at hand
and there is minimum uncertainty. In these circumstances, the
likelihood of making a wrong decision is small. Unfortunately,
decisions about many aspects of environmental health must be made
under conditions of considerable urgency, where there is insufficient
time to completely investigate a problem and perform an exhaustive
analysis of available options. So it is not uncommon for risk assessors
and risk managers to find themselves in situations where quick
responses to rapidly unfolding environmental threats must be made
with minimal information and large degree of uncertainty.
Even in problem situations where months or years are available for
further investigations, the necessary information will never be entirely
complete or totally unambiguous. Risk managers must always be
prepared to make hard decisions under uncertainty. A maxim in risk
management states that: “not to decide is to decide”—in other words,
deferred decisions constitute implicit acceptance of the status quo,
including the health risks and adverse outcomes that may result from
the decision not to act. On the other hand, precipitate action may
introduce new hazards as the result of substitution of an agent with
known risks by another agent with uncharacterized, potentially greater,
risks.
In the longer term, further research and analysis can undoubtedly
reduce ignorance and uncertainty, but deferring crucial health
protection decisions to a later point in time may lead to disaster. This
dilemma is exemplified by the tragic outcomes of the HIV blood
transfusion problem in Canada and elsewhere, the “Mad-Cow” disease
outbreak in the U.K., and the seemingly pervasive health problems
associated with environmental contaminants—for example, asbestos,
heavy metals such as mercury and lead, or organochlorine compounds
such as PCBs and dioxins.








Recently, the report of the
Presidential /Congressional
Commission on Risk Assessment and
Management (1997) has proposed
the use of stakeholder involvement:
“Involvement of stakeholders
—parties who are concerned with or
affected by the risk management
problem—is critical to making and
successfully implementing sound,
cost-effective, informed risk
management decisions.”

 
The weight-of-evidence concept constitutes a foundation for risk
decision-making under uncertainty, because it allows risk managers to
assess the credibility of conflicting evidence about the harmful
properties of an environmental contaminant in a systematic and
objective manner. In order to ensure that the process of weighing the
evidence is scientifically defensible, the weight-of-evidence concept
requires that the available evidence is of sufficient strength, coherence,
and consistency to support an inference that a serious hazard may exist.
Hence, the weight-of-evidence assessment is an ongoing activity where
new scientific evidence will be continually incorporated within the
decision framework as it becomes available. This process can be
frustrating to risk managers because earlier conclusions about a
potential hazard supported by existing evidence can later become less
convincing, whenever new evidence contradicts the scientific
underpinnings of a decision previously taken. Another source of
frustration is the inability of such inference processes to achieve a final
irreversible decision with a high degree of confidence in its correctness.
There is a natural desire among risk managers for achieving “closure”
in risk decision-making, so that controversial issues in environmental
health protection can be resolved once and for all on the basis of
compelling scientific evidence. Unfortunately, it is the fate of most
environmental contaminants to be subjected to repeated cycles of risk
assessment and risk management review over many years or decades.
While expensive and time consuming, such a periodic reappraisal of
scientific evidence helps to ensure that risks are managed in accordance
with the best available scientific information.

 
 !
Risk managers tend to rely on several broad principles of risk decisionmaking which may become explicitly stated as a formal regulatory
policy for a health protection organization. Or they may play a less
visible role in the day-to-day subjective judgements of individual
decision-makers and decision-making bodies. The risk management
process relies on two major operating principles to accomplish its
health protection mission under conditions of incomplete and uncertain
information. These two principles—the Precautionary Principle and
Sound Science—represent a pair of conceptual anchors placed at
opposite ends of the spectrum of possible decisions.
The Precautionary Principle impels decision-makers towards action in
situations where a serious or irreversible health hazard is believed to
exist as a possibility, although the exact probability of the suspected
hazard is imperfectly understood. The principle of Sound Science seeks
to restrain decision-makers from attempting premature or fallacious
judgements about hazards which may not constitute meaningful threats







Association and causality
While risk assessment and management
of discrete hazards, such as
manufacturing processes within
industrial facilities or the
transportation of dangerous goods, is
itself a difficult and complex
undertaking, at least the major causal
linkages between a potential hazard
and its associated risks to health safety
are visible, concentrated in one
locality, directly quantifiable, and
reasonably well understood.
In contrast, environmental
contaminants represent hazards that
are diffuse, mobile, not easily
quantifiable, and often poorly
understood. Another major obstacle in
establishing the causal linkages for the
harmful health effects of environmental
contaminants arises from the need to
account for the impact of other causal
agents such as smoking, radiation, and
natural toxins, which can also cause ill
health, either separately, or sometimes
in combination with manufactured
chemicals or their by-products.
Moreover, there are usually large gaps
in time between an exposure to a
contaminant, the observation of
possible ill effects, and a medical or
scientific assessment about association
and causation. Therefore the risk
assessment and management of
hazardous environmental contaminants
usually requires the use of weight-ofevidence methods for determining the
nature and extent of population health
risks, relying on scientific inferences
derived from studies both in exposed
human populations and in test animals
exposed in toxicological studies.

to human health. It requires that the decision to act should be based on
a reasonable probability of harm. Ideally, both principles would guide
the decision-maker towards a middle ground, where the majority of
substantial risks are effectively addressed and the majority of
insignificant ones are dismissed. Frederick the Great of Prussia is
quoted as exclaiming during the Seven Years War that: “He who would
defend everything defends nothing.” Deciding what to defend and not
defend is an essential task of risk decision-making.
2
*
&9%!
*!&03


The Precautionary Principle states that personal health is an
irreplaceable human good, and asserts that protection of public health
should be treated as the paramount concern for regulatory organizations
and governments. All other concerns, such as cost of control or adverse
economic impact, would then be accorded secondary importance in
public health policy, although consideration of the cost-effectiveness of
the proposed control measures necessarily need to be considered.
In order to protect the environment, the precautionary
approach shall be widely applied by States according to
their capabilities. Where there are threats of serious or
irreversible damage, lack of full scientific certainty shall
not be used as a reason for postponing cost-effective
measures to prevent environmental degradation.
UNCED Rio Declaration - Principle 15
In Canada, the Precautionary Principle is broadly supported as an
important guiding concept by most regulatory bodies at the provincial
and federal level, but is often interpreted in a more narrow sense than
the original Rio Declaration. For example, according to a recent
definition developed for Environment Canada under the CEPA
regulations, the Precautionary Principle should consist of four operative
rules:

•

the primary objective is minimization of harm to human health

•

appropriate remedial and preventive actions shall not be
postponed by the presence of uncertainty

•

evaluation of evidence shall be based on adherence to sound
science

•

decision-making shall include a careful consideration of the
cost-effectiveness of any proposed course of action

Decisions that require sound scientific evidence of harm are often seen
as operating according to the ‘weak’ or ‘utilitarian’ Precautionary
Principle. In contrast, the ‘strong’ or ‘ecocentric’ Precautionary
Principle tends to emphasize the need for environmental protection
even in the absence of convincing evidence. The distinction between





#

The Precautionary Principle has
become a declared policy issue for an
increasing number of risk
management deliberations within
Canada in recent years. For
example, the 1993 Comprehensive
Air Quality Management Framework
for Canada, agreed to by federal,
provincial and territorial
governments through the National
Air Issues Coordinating Committee
(NAICC), includes the precautionary
principle among its list of principles
of cooperation. The Framework
provides that: "Preventive and
corrective actions will be based on
the precautionary principle, sound
science, and consistency with
nationally acceptable data and
assumptions".

The Precautionary Principle, as
stated in the historic Rio Declaration,
contrasts sharply with existing
regulatory policy in many countries.
For example, in the United States,
many of the legislative provisions for
regulating environmental hazards
require that the legal burden of
proof to enact more restrictive
controls over pollutants rests with the
regulatory body, not the industry that
emits the pollutant. Under this
system, those regulatory actions not
supported by convincing scientific
evidence of a potential health risk
accompanied by a well-demonstrated
need for stricter emission controls,
are likely to be overturned in the
federal courts (e.g. rejection of the
1996 EPA proposal for a revised air
quality standard for particulate air
pollutants). Conversely, in the
European Union, application of the
Precautionary Principle is accorded
official status within the EU
Environment Health Policy, and is
incorporated into the environmental
laws of several of the member states.

‘weak’ and ‘strong’ is basically one of values, where strong precaution
favours more protection with less evidence and weak precaution
favours more evidence before protective decisions are made.
The Precautionary Principle is most appropriately applied in the early
stages of an unfolding risk problem, when the potential for serious or
irreversible health consequences is great but the likelihood of
occurrence is uncertain. In such cases, responsible risk managers may
formulate a reasonable worse-case scenario and act expeditiously to
provide interim controls to minimize the likelihood of anticipated
harmful consequences, until better information becomes available at a
later time. This cautious policy approach to health risk is generally
termed conservatism. Many conservative assumptions are included in a
typical risk assessment process, and all are inherently biased toward a
more pessimistic interpretation. Initially, the worst possible outcome
across a spectrum of uncertain contingencies is carefully considered.
As more credible information becomes available, the pessimism of
these conservative assumptions can usually be discarded as unlikely, or
redefined in a less pessimistic manner. Hence, preliminary risk
estimates for environmental hazards tend to be quite high, but many
will tend to diminish over time as better quality data become available.
Numerous conservative assumptions are purposely built into most risk
assessment methods, both to ensure public safety, and to encourage the
collection of better scientific data by organizations who wish to see risk
estimates adjusted downward, with a corresponding relaxation of the
more stringent environmental standards. However, many business
people in various regulated sectors of the economy remain concerned
that routine application of compounded conservatism within the risk
assessment process—always assuming the worst in every instance
where information is lacking or unclear—could result in significant
overestimation of health risks, thereby keeping many environmental
standards at excessively restrictive levels.
2
%9!
&
!&


The principle of Sound Science requires that an observed pattern of
events should not be accorded causal significance unless confirmed by
careful data collection and thorough statistical analysis. The Sound
Science requirement serves as the foundation of all of our modern
experimental sciences, because it ensures that misguided ideas arising
from chance occurrences, biased data collection, flawed experimental
design, or defective analysis will be rejected. Sound scientific ideas
based on confirmed observations from systematic experimental studies
will eventually be accepted, although it may require a longer time to
reach a firm conclusion.






+

Some environmental advocacy
groups assert that the ecocentric
version of the Precautionary
Principle should be more strictly
followed, regardless of cost, through
the wholesale restructuring of
modern society to emphasize
environmental health over industrial
productivity. Such a position is only
defensible if one accepts the severe
strains on society that would result
from this policy. However, a more
balanced application of the
Precautionary Principle can also be
adopted, if risk assessment methods
are used to quantify the health risk
associated with a particular
environmental hazard, and risk
management procedures are used to
efficiently apportion societal
resources to address those hazards
presenting the greatest risk. In this
sense, risk assessment and
management methods should not be
viewed as antithetical to the
Precautionary Principle—rather the
risk paradigm attempts to define the
criteria under which the
Precautionary Principle can be
effectively applied to regulatory
decision-making.

Sound Science adheres to the concept of empirical conservatism, which
is entirely distinct from the policy conservatism inherent in the
Precautionary Principle. It is not surprising that research scientists and
risk managers frequently miscommunicate their underlying
assumptions, since both invoke conservatism in their professional
judgments, but in opposite directions. Neither type of conservatism is
inherently wrong, but simply
PRECAUTIONARY PRINCIPLE, SCIENCE, AND RISK ASSESSMENT
different.
Some commentators on the Precautionary Principle view it as separate from, and even

In empirical science, where
antithetical to, science. Hathcock, for example, has said the Precautionary Principle can
nature is studied without
"negate the input of science" and "overrule risk assessment," and allow
"arbitrary" food safety decisions whenever there is any uncertainty about a risk.
reference to policy consid…
erations, empirical conserSumming up, at the start of this section I cited the view that precaution is antithetical to
vatism attempts to minimize
science. The opposite is in fact true, and we must avoid the misleading implications of the
the likelihood of ‘false
perceived dichotomy between "science" and "precaution." Precaution is grounded in
alarms’, but accepts that a
scientific analysis, and applying precaution to decisions requires rigorous scientific input
from a range of disciplines. At times, precautionary approaches, with their emphasis on what
limited number of real
science does not know as well as what is known, may in fact require more rigorous science
effects may therefore go
than risk assessment, which has been known to brush aside uncertainties in order to answer
undetected for a substantial
too narrowly-drawn questions.
period of time. In mandated
science, which informs deciHathcock's quote could be turned around: Risk assessment can also be used to "overrule"
science-based precautionary judgments. But neither scenario represents sound decisionsion-making in the public
making. Done properly, Risk Analysis uses risk assessment and precaution together, as
health arena, policy conserinseparable and essential components of science-based decision-making.
vatism attempts to ensure
public safety by assuming
Science, Precaution, and Food Safety, Groth (2000)
the worst-scenario in uncertain situations—policy conservatism can thus occasionally fall prey to the ‘Chicken Little’
syndrome: “the sky is falling, and we must see the King.”
2


@
*!&03


ALARA stands for ‘As Low as Reasonably Achievable’—a concept
which asserts that environmental exposure to toxic substances should
be kept as low as is reasonably achievable, using pollution control
equipment and industrial processes that can be installed and operated at
reasonable cost.
The ALARA Principle thus occupies a broad middle ground between
the mutually opposing constraints of the Precautionary Principle and
Sound Science. Thus, an acceptable risk would conceivably be ensured
by adopting reasonably achievable control measures which balance
risks and benefits. In many jurisdictions, including Canada, the
ALARA Principle is frequently applied in formulating exposure
guidelines or regulatory limits for environmental hazards.
In practical terms, the question of defining ‘acceptable risk’ and
‘reasonably achievable’ must somehow be resolved. Environmental
risk management practices were initially developed under the





.

assumption that the public could be entirely protected from all risk, and
consequently, that no level of risk was acceptable. With the advent of
progressively more sensitive means of detecting miniscule quantities of
environmental contaminants, and with the realization that small
exposures to carcinogenic agents may
ALARA — Framework for Risk Criteria
entail some possible risk, the present goal
of risk management is to reduce risk to
the maximum extent deemed possible by
(A) Intolerable region
Risk cannot be justified
on any grounds.
practical considerations.

@
*!&03


!
"
!(
)
!


(B) As Low As Reasonably Achievable
(ALARA). Risk is acceptable only
if a compensating benefit is available.

The ALARA principle generally uses the
word ‘low’ to denote low exposure,
which is believed to correspond in a
(C) Broadly Aceptable Region.
No need for detailed work to
general way to a lower health risk,
demonstrate ALARA.
although this relationship is not
necessarily quantifiable in all cases.
Reasonable cost of control is evaluated
largely through technological criteria, such as best available control
technology (BACT) or best practicable control technology (BPCT),
although in recent years more consideration has been given to the
balancing of costs and benefits. Some industries feel it is unreasonable
to expect continual upgrading of environmental control technologies
when the corresponding reduction in health risks remains questionable,
or where persuasive evidence of a reasonable relationship between
costs and benefits of risk reduction is absent.
In practice, the ALARA principle is most readily applicable in largescale or high-technology industries, such as the manufacture of plastics
and specialty chemicals, where environmental control technology is
constantly being upgraded as new equipment and processes are
introduced. Because installing new equipment can, with relatively
minimal expense, incorporate recent technological improvements to
reduce emissions of hazardous substances, the cost of environment
control is partly covered within the capital expenditure budget
dedicated to increasing productivity. The ALARA principle works less
well in low technology sectors or in smaller industries, where there is a
slower turnover of capital equipment and where industrial processes are
relatively unsophisticated. In the latter case, sectoral approaches,
encompassing a group of related small- and medium-sized industries,
may enable more effective use of the ALARA approach.
#2

%!<3


3%!0

A more recent concept called the Reasonable Relationship approach
has proposed that the costs of control for environmental hazards should





/

Total
Risk

Tolerable only if risk reduction
is impractical or if its costs are
grossly disproportionate to
the benefits gained.

Tolerable if cost of reduction
would exceed benefits gained

Negligible risk.

'Acceptable' levels of environmental
hazards vary with the specific
application and substance being
considered and, in principle, should
be as low as reasonably achievable,
taking into account not only the
hazard but also the social and
economic benefits and the available
technology (the ALARA Principle).
…For chemicals, the ALARA
Principle is applied in setting the
guidelines or legal limits for the risk,
which thus constitutes the
'acceptable' risk. As a consequence of
the adherence to ALARA, the
'acceptable' risk varies from
application to application.
Health Canada, 1998

bear a “reasonable relationship” with respect to the corresponding
reductions in health risk. The approach does not strictly require that the
health benefits of risk reduction must match or outweigh the costs of
control, but it asserts that the balance of costs and benefits should at
least approximate each other. This approach requires a risk assessment
to estimate the expected reduction of the health risk, and a
socioeconomic analysis to estimate the associated cost of control. If the
anticipated costs and benefits are deemed in reasonable relationship,
either by formal calculations such as cost-benefit analysis, or by
informal consensus among stakeholders, then the proposed protective
measures would be implemented.

Under the ALARA Principle,
acceptable risk is commonly defined
as a lifetime risk which is so small as
to be considered 'essentially
negligible' by most people. A
quantitative criterion of one in a
million (10-6) for acceptable risk,
often termed the de minimis risk
level, has been adopted in many
jurisdictions, most notably the United
States, where it is routinely applied
for regulating carcinogenic
chemicals in the environment.
More recently, it has become
apparent that one in a million risk is
a very stringent criterion in situations
where relatively few people are
exposed. As a consequence of
conforming to the ALARA Principle,
the definition of a negligible risk
(not necessarily as stringent as de
minimis risk) may vary from case to
case, depending on the practical
limitations for controlling each
regulated substance.






7

SUGGESTED READINGS
Barnard, R.C. (1987). Scientific risk assessment and the regulation of
human cancer risks: background and new directions. Am Ind Hyg
Assoc J 48:798-803.
Barnard, R.C. (1994). Scientific method and risk assessment. Regul
Toxicol Pharmacol 19:211-8.
Barnard, R.C. (1995). Risk assessment: the default conservatism
controversy. Regul Toxicol Pharmacol 21:431-8.
Barnes, D. (1994). Times are tough - brother, can you paradigm? Risk
Anal 14:219-223.
Brunk, C., Haworth, L., Lee, B. (1991). Value Assumptions in Risk
Assessment: A Case Study of the Alachlor Controversy. Waterloo, ON:
Wilfrid Laurier University Press.
Canadian Standards Association (1997). CAN/CSA-Q850-97. Risk
Management: Guideline for Decision-Makers. Etobicoke (Toronto),
Canada: Canadian Standards Association.
Commission of the European Communities (2000). Communication
from the Commission on the Precautionary Principle. Brussels.
De Rosa, C.T., Stevens, Y W., Johnson, B.L. (1993). Cancer policy
framework for public health assessment of carcinogens in the
environment. Toxicol Ind Health 9:559-75.
De Rosa, C.T., Pohl, H.R., Williams, M., Ademoyero, A.A., Chou,
C.H., et al (1998). Public health implications of environmental
exposures. Environ Health Perspect 106:369-78.
Dyck, W., Del Bel Belluz, D., Craig, L. (1998). Current Directions in
Environmental Risk Assessment and Management. Waterloo, Canada:
Network for Environmental Risk Assessment and Management
(NERAM).
Environment Canada (no date). Tabs on Contaminated Sites #15: Risk
assessment principles and screening processes. Environment Canada Ontario Region, Environmental Protection Branch.
Environment Canada, Health Canada (1996). CEPA review: The
government’s response. Ottawa.
Goldstein, B.D. (1999). The precautionary principle and scientific
research are not antithetical [editorial]. Environ Health Perspect
107:A594-5.





:

Gori, G.B., Flamm, W.G. (1991). How sick a patient? Report of a
workshop on cancer risk assessment. Regul Toxicol Pharmacol
14:215-22.
Gray, J., Brewers, J. (1996). Towards a scientific definition of the
precautionary principle. Marine Pollution Bulletin 32:768-771.
Groth, E. (2000). Science, Precaution and Food Safety: How Can We
Do Better? Yonkers NY: Consumers Union of U.S.
Harrison, K., Hoberg, G. (1994). Risk, Science, and Politics:
Regulating Toxic Substances in Canada and the United States.
Montreal: McGill-Queen’s University Press.
Health Canada (1990). Health risk determination: The challenge of
health protection. Ottawa: Health Protection Branch.
Health Canada. (1997). Health and environment, partners for life:
Assessing and managing health risks, pp. 32-48. Ottawa.
Health Canada (1998). The health and environment handbook for
health professionals: Chapter 2. Risk. Ottawa.
Jarabek, A.M., Farland, W.H. (1990). The U.S. Environmental
Protection Agency’s risk assessment guidelines. Toxicol Ind Health
6:199-216.
Joint Committee on Health and Safety, Royal Society of Canada,
Canadian Academy of Engineering (1993). Health and Safety Policies:
Guiding Principles for Risk Management. Waterloo, ON: Institute for
Risk Research.
Joint Working Group — Health Canada, Atomic Energy Control
Board, Ontario Ministry of the Environment (1998). Assessment and
management of cancer risks from radiological and chemical hazards.
Ottawa.
Kates, R.W. (1994). Sustaining life on the earth. Scientific American
Oct;271(4):114-22.
Krewski, D., Birkwood, P. (1987). Risk assessment and risk
management. Risk Abstracts 4:53-61.
Krewski, D., Slovic, P., Bartlett, S., Flynn, J., Mertz, C. (1995). Health
risk perception in Canada II: Worldviews, attitudes, and opinions.
Human & Ecological Risk Assessment 1:231-248.
McClellan, R.O. (1994). A commentary on the NRC report “Science
and judgment in risk assessment.” Regul Toxicol Pharmacol
20:S142-68.





8

McClellan, R.O. (1999). Keynote Address: Human Health Risk.
Assessment: A Historical Overview and Alternative Paths Forward.
Inhal Toxicol 11:477-518.
Morgan, M., Henrion, M. (1990). Uncertainty: A Guide to Dealing with
Uncertainty in Quantitative Risk and Policy Analysis. New York:
Cambridge University Press.
National Research Council (1983). Risk Assessment in the Federal
Government: Managing the Process. Washington DC: National
Academy Press.
National Research Council (1994). Science and Judgment in Risk
Assessment. Washington, DC: National Academy Press.
National Research Council (1996). Understanding Risk: Informing
Decisions in a Democratic Society. Washington, DC: National
Academy Press.
Omenn, G.S. (1995). Assessing the risk assessment paradigm.
Toxicology 102:23-8.
Patton, D. (1994). The NAS risk paradigm as a medium for
communication. Risk Analysis 14:375-378.
Paustenbach, D. (1995). Retrospective on U.S. Health Risk
Assessment: How others can benefit. Risk: Health, Safety, and
Environment 6:283-332.
Power, M., McCarty, L. (1998). A Comparative Analysis of
Environmental Risk Assessment/Risk Management Frameworks.
Environ Sci Technol 32:224A-231A.
Presidential/Congressional Commission on Risk Assessment and Risk
Management (1997). Framework for Environmental Health Risk
Management, Final Report, Volume 1: Washington DC.
Presidential/Congressional Commission on Risk Assessment and Risk
Management (1997). Risk Assessment and Risk Management in
Regulatory Decision-Making. Final Report. Volume 2.
Washington DC.
Rodricks, J.V. (1994). Risk assessment, the environment, and public
health. Environ Health Perspect 102:258-64.
Schierow, L. (1998). The role of risk analysis and risk management in
environmental protection. Washington: Congressional Research
Service.








Thomas, S., Hrudey, S.E. (1997). Risk of Death in Canada: What We
Know And How We Know It. Edmonton: Alberta University Press.
Tickner, J. (1999). A map toward precautionary decision making. In
Protecting Public Health and the Environment: Implementing the
Precautionary Principle. Ed. J. Tickner, Washington, DC: Island Press.
VanderZwaag, D. (1997). CEPA and the Precautionary Principle.
Ottawa: Environment Canada.
Walker, V. (1996). Risk characterization and the weight of evidence:
adapting gatekeeping concepts from the courts. Risk Anal 16:793-799.
Wildavsky, A. (1988). Playing it safe is dangerous. Regul Toxicol
Pharmacol 8:283-7.
Zeckhauser, R.J., Viscusi, W.K. (1990). Risk within reason. Science
248:559-64.