Definitions of Risk Assessment

". . . the attempt to quantify the degree of hazard that might result from human activities ... an exercise that combines available data on . . . potency in causing adverse . . . effects with information about likely . . . exposure, and through the use of plausible assumptions, it generates an estimate of risk."

-William D. Ruckelshaus, 1985

". . . the scientific activity of evaluating the potential effects of an entity and its application in order to ascertain the likelihood that an adverse effect may occur, and to characterize the nature of that effect."

—Paraphrase from National Research Council, 1983

". . . the process of obtaining quantitative or qualitative measures of risk levels, including estimates of possible health and other consequences."

". . . an analytical tool that facilitates the organisation of large amounts of diverse data with the goal of estimating the potential risk posed by a process (or event) of interest."

". . . the measures to estimate what harm might be caused, how likely it would be to occur and the scale of the estimated damage."

—United Nations Environment Programme (UNEP), 1996

activity can proceed with an acceptable level of safety. Thus the process may be "put on hold" until the needed information is provided.

Organizing the Scientific Information

The very large and ever increasing amount of scientific information available warrants consideration of structured approaches to risk assessment. Indeed, risk assessment requires a different way for scientists to organize and evaluate information. They are asked to evaluate a product's safety as opposed to its potential contribution to scientific knowledge.

In this brief discussion we highlight some of the important aspects of the thinking that has gone into developing such structured approaches. Although these appear disparate in nature, they are consistent with the goal of defining and quantifying potential risks or supporting the notion of "no fore seeable risk." In reality, no single approach is best; the one used typically is the approach most suitable to the needs of the present circumstances. Reviewers will find themselves using different approaches to different applications, or even to different sections of one application.

Over the years, many approaches to biosafety analysis have been used by regulatory scientists or proposed in the literature.

Trait Analysis Approach

In trait analysis, the assessor categorically evaluates attributes of (1) the parental organisms, (2) the genetic construct, (3) the modified organism, and (4) the environment in which the organism is to be released for testing. The analysis uses pertinent criteria and an indication of levels of concern dependent upon the attributes. For example, an organism with a short survival time would be of less concern than one with a long survival time.


TRAIT ANALYSIS Characteristics of the modified organism including the transferred gene(s), the parental organisms, and the receiving environment


Works well when releases are small in scale, but becomes increasingly difficult and less certain as spatial and temporal scale increases

FAMILIARITY Comparison of modified organism to similar organism(s) that is (are) well known and of GM traits to similar traits derived through classical genetic methods

Based on the assumption that "small" genetic changes (one to four genes) will result in no significant change in a well-known organism (e.g., crop plant) and that phenotypic expression is the same regardless of how the modification was obtained

FORMULAIC Possible adverse effects (e.g., to the environment or human health) and the probability for their occurrence

Useful for organizing scientific information into two categories; facilitates consideration of risk-management options

INTUITIVE What is known or available to an

REASONING individual or group of assessors based on education, experience, and reason

May rely too much on what seems important as opposed to what should be considered (becomes less of a concern with training and experience)

Similarly, an organism with a narrow geographic range would be of less concern than one with a wide or unknown range.

Familiarity Approach

This popular line of approach advances the concept of relative risk assessment. The determination of level of concern is based not only on the genetic characteristics of the organism, its pheno-type, and the environment into which it will be released, but also on a comparison of the GM organism to the corresponding well-known non-GM organ ism, and the GM trait derived from classical genetic techniques. In other words, how "familiar" scientists are with a particular organism and trait helps them to determine the appropriate level of concern. The essence of the argument is that because most crop plants are genetically modified in increments, the amount of new genetic material is a very small percentage of the plant's genome, and, regardless of how the trait was derived (through classical breeding or by modern molecular techniques), it will phe-notypically be the same. For example, by comparing GM plants with the parental plants that, based on past introductions, have a safe history, it is possible to arrive at a reasonable assessment of how the modified plants will behave in the environment.

Formulaic Approach

Some regulatory agencies have modified the basic risk-assessment approach used for chemicals to use with biotechnology products. In essence, categorical considerations of hazard (H) ascribable to a chemical and the chemical's potential exposure (E) to individuals or groups of individuals are determined. In combination, they determine a level of risk (R). This is commonly described algorithmi-cally as R = H X E. The important insight this equation offers is its inherent organizational nature. The analysis may be subdivided into manageable parts. Using estimates of a potential impact (hazard) and the proximity of a material to the potentially affected component of the environment (exposure), an estimate of the level of risk is obtained. Both hazard and exposure are necessary for risk to be present. That is, presence of a hazard without exposure, or exposure to something that is not hazardous, poses no risk. Other considerations such as dose response (a measure of the level of potential impact) and risk characterization (severity of concern and level of uncertainty) complete the process. More recent thinking about this paradigm has led to minor alterations in the basic formula to recognize and account for the nature of organisms as opposed to chemicals. These alterations include the addition of terms for survivability (fitness), mutation, and reproduction.

Intuitive Reasoning

Assessors tend strongly to rely on their intuition when evaluating applications to release GMOs. Of course they are educated and have considerable expertise, usually in a specific discipline, but because they will have to make decisions with incomplete information, they tend to base decisions on what "feels right." Unlike the previous approaches, the intuitive approach has no structure per se on which to develop an assessment; individual assessors have differing intuitions. Because some measure of consistency is lacking, risk assessors using only this approach may find it more difficult to communicate with other assessors and decision makers.

Despite the inherent level of uncertainty involved in a risk-assessment process and the fact that, at present, assessors are addressing events with a low probability of occurring, using a systematic approach to risk assessment is a worthwhile exercise. When used appropriately, the approaches described above will help to organize scientific information, facilitate communication, and minimize paralysis in decision making.

Practical Considerations

0 0

Post a comment