VI Risk and uncertainty
9.43 We see the assessment of risk and uncertainty dealing with two important components:
The relative importance of these two aspects will vary according to the circumstances, and in any given case one of them may predominate.
- looking at broader uncertainties about the current issue which extend beyond the available scientific evidence and considering a wider range of possibilities; and
- where sufficient data are available, quantifying and analysing the risks associated with the issue under consideration.
9.44 Risk assessments prepared in support of decisions on environmental polices on standards should start with information about the nature of the hazard which the policy or standard seeks to address and the extent and quality of the evidence available for assessing the risks it poses. This part of the analysis should indicate whether the hazard is of a relatively well-understood type; if it is unfamiliar, an attempt should be made to identify the most nearly analogous hazards and the aspects which are not understood.
9.45 The limitations and uncertainties in any estimates of risk must always be made clear in ways which are meaningful to people without particular specialist knowledge.
9.46 Risk assessments should identify the uncertainties which have the largest implications for actions that would need to be taken to reduce or resolve them. However, it would be inappropriate and misleading to attempt to incorporate into risk assessments estimated probabilities for the correctness of particular scientific theories or interpretations.
9.47 Whatever action is taken in the name of precaution (from the use of worst-case scenarios and safety factors in assessments through to application of the precautionary principle in decision making) should be transparent and subject to review in the light of development of understanding. Relevant data should be collected and reviewed on a continuous basis; and if a standard has been set, it should be revised up or down as necessary.
9.48 If there are sufficient data, and sufficient knowledge of the underlying processes, quantitative risk assessments should cover not only risk of human deaths but risks of other harmful outcomes. For each estimate the assumptions should be made explicit and clearly stated.
9.49 No satisfactory way has been devised of measuring risk to the natural environment, even in principle, let alone defining what scale of risk should be regarded as tolerable.
9.50 As well as distinguishing between different types of effect from a hazard, risk assessments should also:
(a) indicate the distribution of risks (whether they are especially high for people in certain localities, age groups, or occupations, or people with certain medical conditions or genetic pre-dispositions); and
(b) characterise as far as possible the respective perceptions of the risks held by relevant groups, the meanings the risks will have for them, and their views about the tolerability of the risks.
Quantitative information on these points should be provided where it is available, otherwise qualitative assessments should be given.
9.51 For risks of the same general type, and where data are available and the processes sufficiently understood, direct comparison between options may be useful in informing decisions, for example;
(a) between the risks from the hazard being addressed and other risks of the same general type affecting the same group of people or compartment of the environment, so that estimates can be made of the total risk of that type to which these will be subject;
(b) between the risks from the hazard being addressed and the risks from different sources or pathways for the same pollutant or different pollutants from the same source, in order to identify any options for risk reduction that might obtain a larger benefit for a similar cost.
However, making comparisons between risks which the public does not perceive as comparable can undermine the credibility of regulators and governments.
67. Risk and its communication are high on the Government's agenda. The DETR Guidelines for Environmental Risk Assessment and Management (first published in 1995 and now being revised) provide a common framework for use in environmental decision making. They set out some basic principles which DETR and the Environment Agency normally use in the assessment and management of environmental risks and, more generally, provide decision makers, practitioners and the public with a consistent language and approach for environmental risk management.
68. The guidelines describe a framework based on science-based risk assessment, stakeholder involvement and risk management. They describe ways of estimating the probability of harm and of evaluating the severity of that harm, including how to handle the uncertainty likely to exist in the quantification of the probability and consequences of any hazard. They emphasise the importance not only of communicating the scientific aspects of risk in a coherent and transparent way, but of understanding underlying concerns, particularly those of the general public who may be affected by decisions based on risk assessments. This enables decisions to be made on whether a risk is acceptable, whether and how it should be managed, or whether it should be removed.
69. The framework identifies the key stages in risk assessment:
- hazard identification: identifying the potential for harm. This will have an important bearing on the overall assessment. It is important not to overlook secondary hazards that may arise - for example, during flooding contaminated sediments might be deposited on agricultural land;
- identification of consequences: identifying the potential consequences that may arise from any given hazard - at this stage no account is taken of likely exposure and therefore likely consequences;
- estimation of the magnitude of consequences: these can be actual or potential harm to human health, property or the natural environment. Estimation needs to take into account geographical scale and duration of consequences, and how quickly harmful effects might be seen. Long-term problems must be considered as well as immediate risks;
- estimating the probability of consequences: taking into account the probability of a hazard occurring, and of harm resulting;
- evaluating the significance of a risk: having determined the probability and magnitude of the consequences, it is important to place them in context. At this point some value judgements are made, either through reference to some pre-existing measure, such as a toxicological threshold or environmental quality standards, or reference to social, ethical or political standard. In some circumstances, a formalised quantitative approach to determining significance may be possible, for example the tolerability of risk frameworks developed by the Health and Safety Executive. In other instances, the risks of various options might be compared against one another;
- options appraisal: considering whether to reject, accept or reduce the risk, or to mitigate the effects.
70. The guidelines emphasise that evaluating the social significance of the risk is an important part of the process. The objectives of doing so are to engage stakeholders, help to identify difficult cases, aid the communication of risk messages, and not least, ensure that decisions on the acceptability of environmental risk recognise that environmental protection is part of the wider context of sustainable development, and so must be considered alongside other sustainable development objectives of economic growth, social progress, and prudent resource management. The guidelines outline factors which influence the perception of risk, and its communication. For example, they specifically recognise the problem of making comparisons between activities that have similar statistical probabilities and similar outcomes but are not comparable with regard to whether they are taken voluntarily or not. Such comparisons are likely to be viewed with scepticism by the public.
71. The Government notes the Royal Commission's view on the difficulty of measuring risk to the natural environment. The guidelines explicitly recognise that a complicating issue for environmental risk management is the lack of an easily defined measure of what constitutes harm to the environment. There is unlikely to be a single 'satisfactory way' of defining the tolerability of risk, simply because views on tolerability will depend on the risk concerned and public attitudes towards it. In some cases definitions of damage are laid down in statute, but in others criteria will need to be selected on the basis of scientific and social judgements. Legislation increasingly requires assessments of this kind to be undertaken (for example Part IIA of the Environmental Protection Action 1990, and the Habitats Directive). Criteria therefore need to be devised. To assist in this, the Environment Agency is collaborating on research to develop a methodology to evaluate risks to ecosystems from contaminated soils.
72. The Cabinet Office has been organising seminars on risk involving key Ministers, the Better Regulation Task Force, senior officials, scientists, professionals, consumer representatives and journalists. There is a recognised need for greater consistency in the Government's handling of risk, clarifying the role of experts, and improvements in the way in which the Government communicates about risk. Following the Cabinet Office seminars on risk, Government Departments are now required to prepare frameworks which set out their individual approach to risk. Departments will have to make their frameworks and procedures public by the beginning of September 2000.
73. HSE has produced for its stakeholders an account of how it goes about regulating risks. This account (Reducing Risks, Protecting People) shows how assessment of risk and uncertainty are part of a wider process of tackling a risk issue by defining and characterising the issue, examining the available options, adopting a decision, implementing the decision and evaluating the outcome. HSE has also commissioned research to review available techniques for comparing and ranking risks in order to set priorities.
74. The Environment Agency set up a National Centre for Risk Analysis and Options Appraisals in 1997 and, where appropriate, has prepared functional guidance on environmental risk assessment for the various legislation regimes it administers. The Agency's primary use of risk assessment is in supporting regulatory decisions on the issuing of environmental permits.
75. The Interdepartmental Liaison Group on Risk Assessment (ILGRA) has produced guidelines on Risk communication: a guide to regulatory practice. They are based on case studies of communication by Departments and set out four principles for successful risk communication with simple guidance on each. The guidelines have been widely circulated within Government, and Departments have been encouraged to benchmark their own approaches against this good practice.
76. Chapter 4 of the Government's Sustainable Development Strategy, A better quality of life, sets out the Government's approach to application of the precautionary principle. This takes as the starting point Principle 15 of the 1992 Rio Declaration - "where there are threats of serious or irreversible damage, a lack of full scientific certainty shall not be used as a reason for postponing cost-effective measures to prevent environmental degradation". It then elaborates the following key points:
(a) the precautionary principle means that it is not acceptable just to say "we can't be sure that serious damage will happen, so we'll do nothing to prevent it";
(b) precaution is not just relevant to environmental damage - for example, chemicals which may affect wildlife may also affect human health;
(c) at the same time, precautionary action must be based on objective assessments of the costs and benefits of action. The Government is committed to acting proportionately;
(d) there are no hard and fast rules on when to take action: each case has to be considered carefully;
(e) transparency is essential;
(f) decisions should be reviewed to reflect better understanding of risk as more evidence becomes available.
Published 21 November 2000
Royal Commission on Environmental Pollution Index
Environmental Protection Index
Defra Home Page