Thesaurus of Terms Used in Microbial Risk Assessment - Chapter Three: Discussion of Selected Terms
Specialized meanings, relative to both common and technical usages, can present a barrier to clear communication within a field. A number of terms from this Thesaurus have been identified as warranting an in-depth discussion of the potentially contradictory usages, based on their importance in risk assessment and the potential for causing confusion. The discussions below are in addition to the commentary on definitions found at the end of some entries.
3.1 General Terms
A few risk assessment terms merit discussion because they are fundamental to the nature of risk assessment and are frequently misunderstood. These such terms addressed below include: dose-response, endpoint, exposure, hazard/agent/stressor, uncertainty, and variability.
dose and dose-response
Dose is a general term that refers to total amount of a hazard/agent ingested or absorbed. This total amount depends on both the concentration in the media and the amount of the media.
The term dose implies that an exposure to some concentration has occurred. In this respect dose and exposure overlap somewhat in meaning and in some cases they are used synonymously, often erroneously so. Generally, exposure refers to individuals or populations while dose reflects the exposure level of individual (see, for example, “average dose”).
Dose covers a variety of more specific measurements, so it is often qualified with a term that limits its scope. For example, the amount of a hazardous agent that comes in contact with the external body may not be the amount that reaches the target organ (i.e., the dose or dosage). Terminology that recognizes the possibility that the concentration of the hazardous agent can change throughout the exposure scenario is useful for reducing confusion. Dose can have a temporal element as well. In a given case, dose can be defined by the duration of a continuous exposure or the pattern of repeated distinct exposures. Clear definition of what constitutes a dose for a risk assessment scenario can be difficult because a variety of exposure scenarios may be of interest.
The concept of dose bridges the gap between exposure and response. For a concentration to become a dose, exposure must occur. In addition, the response of concern is directly related to the dose. The dose-response relationship characterizes the relationship between the dose level and a probability of response. The response can be individual- or population-based. For example, exposure to a higher level of hazardous agent could result in more total illnesses in a population. Mathematical models are usually used to estimate dose-response relationships.
The term exposure is used in a number of different ways in risk assessment. It is generally understood to mean relevant contact between a person and a hazard. Factors that may affect an exposure include the concentration of the hazardous agent, the duration of contact, the target tissues or organs, and the route of exposure. Exposure can refer to populations or individuals.
exposure route versus exposure source
Exposure route (also known as “route of exposure”) most commonly refers to the manner in which a hazardous agent moves from the environment to target tissues or organs. Examples of exposure routes include inhalation (nose and mouth to lungs), ingestion (oral), dermal (skin), eyes, ears, and sexual. Which routes are relevant for a given hazardous agent depends on both the host (or receptor in ecological RA) and agent (pathogen in MRA; stressor in ecological RA) properties and is situation dependent. An exposure source can originate from either natural or anthropogenic events, activities, or locations that generate or release hazards. Exposure sources can be classified as point sources or non-point sources. An example of a point source is an industrial facility that releases untreated water into a river; an example of a non-point source is agricultural run-off. For MRA, a sewage treatment plant would be a commonly considered point source, whereas, urban run-off is a possible non-point source of pathogens.
An exposure pathway encompasses both exposure source and route, and generally is described by a source, an exposure point, and an exposure route.
Endpoint is used differently in environmental versus human health risk assessment. In environmental risk assessment, endpoint refers to the environmental value that is being protected. Ecological attributes like biodiversity or a particular species like striped bass, are examples. In human health risk assessment, endpoint refers to a manifestation of adverse human health due to exposure to the hazard in question (also known as response in dose-response). Example MRA endpoints include infection, a defined set of symptoms, or death. Both environmental and human health assessments may address multiple endpoints. The significance of the endpoints are part of the justification for conducting the risk assessment. A poorly defined endpoint can lead to misunderstandings or inappropriate expectations from the risk assessment.
Hazard, agent, and stressor can sometimes be used synonymously. However, there are cases where the terms need to be differentiated. For example, “stressor” is used in ecological risk assessment and includes the connotation that the adverse response can be the result of a lack of something – such as a habitat – which would be called a “stressor.” The term “agent” does not have this connotation. “Agent” is used to denote a causative entity that actually physically exists as part of the environment and can be used in either ecological or human health risk assessment. “Hazard” is used primarily in human health risk assessment, although “hazards” are not limited to “agents.” For example, the number of days spent in a hospital may be a hazard that correlates with risk of nosocomial infection.
uncertainty and variability
Uncertainty and variability are two frequently cited causes of uncertainty in the outputs of risk assessments or in the answers to questions posed by risk managers. It is sometimes important to be able to distinguish the causes of the uncertainty in order to better manage risks. Many different sources of definitions are available and several of them offer conflicting definitions of variability and uncertainty. In some of these variability is considered a source of uncertainty. An exception to this view is found in the EPA Exposure Factors Handbook (1997a), where the authors state:
While some authors have treated variability as a specific type or component of uncertainty, the U.S. EPA (in Guidance for Risk Characterization. Science Policy Council, 1995) has advised the risk assessor (and, by analogy, the exposure assessor) to distinguish between variability and uncertainty. Uncertainty represents a lack of knowledge about factors affecting exposure or risk, whereas variability arises from true heterogeneity across people, places or time. In other words, uncertainty can lead to inaccurate or biased estimates, whereas variability can affect the precision of the estimates and the degree to which they can be generalized.
Whether or not variability is considered a source of uncertainty, it is a theme in the definitions of both terms that variability should be distinguished from other sources of uncertainty. A practical reason for this is that other sources of uncertainty theoretically can be reduced through the collection of more data. Variability cannot be reduced by gathering more data, but can be reduced by narrowing the scope of the scenario that is being considered, or by dividing the variability into less variable subgroups and addressing the groups separately.
3.2 Framework Terms
Two of the most commonly used frameworks are the NRC model which has been adapted by Codex for MRA of food risks and the ecological model which has been adapted by ILSI and EPA for MRA of water related risks (CAC 1999; EPA 1998a; ILSI 2000; NRC 1983). The main terms that have different meanings between these two frameworks are analysis, characterization, and profile. How these differ is discussed below. Because these terms refer to different parts of risk assessment depending on the framework, there is potential for confusion. Therefore, clarification when using these terms should reduce confusion.
The EPA/ILSI MRA and EPA ecological frameworks use the term analysis differently than the NRC and Codex frameworks. Codex and NRC use “risk analysis” to refer to the whole field that encompasses risk assessment, risk management, and risk communication. The EPA/ILSI framework uses “analysis phase” to refer to the characterization of exposure and characterization of human health effects within a risk assessment. Therefore, there can be confusion if the term “analysis” is used without the appropriate qualifier, such as risk analysis or analysis phase.
The EPA/ILSI framework uses the term “characterization” in several places. The analysis phase consists of characterization of exposure and host characterization. The final phase is called risk characterization. Codex uses the term hazard characterization to refer to what the EPA/ILSI framework calls characterization of human health effects, which is also called dose-response in other frameworks (e.g., NRC 1983). Therefore the term characterization is commonly used in conjunction with other words as qualifiers. The term is used in its generic sense as well in specific context referring to a particular step in risk assessment. Confusion can be minimized if users are aware of the framework that is being used.
This term is used in the EPA/ILSI MRA and EPA ecological frameworks to refer to the last part of the analysis phase of risk assessment. In the EPA/ILSI framework, a summary of the results of the infectious disease hazard (pathogen) characterization and the host characterization is conducted to arrive at an exposure profile and a host-pathogen profile. The international community (Codex) uses profile to refer to the initial stages of risk assessment where data is gathered and the risk assessment scope is defined. In contrast, the EPA/ILSI framework refers to this initial stage as problem formulation. Therefore the term profile is used to refer to two very different phases of microbiological risk assessment.
3.3 Infectious Hazard-Specific Terms
A microbial risk assessments (MRA) differs from other types of risk assessment in a variety of ways, and MRA terminology reflects this. The ability of microbes to multiply or die in the environment and the dynamic nature of host-pathogen interactions add a unique complexity to MRAs. The dynamic nature of the interaction of host, pathogen, and environment is commonly referred to as the “epi triad.” For describing fluctuations of microbes in the environment, terms used in chemical risk assessment can be applied without causing confusion. Terms like fate and transport, persistence, background level, and lower detection limit are often appropriately used and understood. The dynamic nature of host-pathogen interactions, unique to infectious disease risk assessment, has spawned some terminology that is frequently used but poorly articulated and defined. Two examples are provided below.
No general definition for “immune status” was found in the development of this Thesaurus. Immune status is a frequently used term, so it is surprising that definitions were not readily available. This may be because there are many ways to characterize the human immune system. “Immune status” refers to an individual’s (or population’s) degree of immune system functioning. Immune markers can include, but are not limited to, general indicators, such as T-cell count, and myriad specific markers, such as antibodies that confer acquired immunity. In addition, the general strength and specific abilities of an individual’s immune system fluctuates through time. It can refer to either individual or population based measures of immunity characteristics. The EPA/ILISI framework identifies the following factors as influencing immune status: age, genetic background, pregnancy, nutritional status, concurrent illness, and medical treatment (ILSI 2000). Because this is a very common term in MRA, therefore it would useful to develop a definition that is broad enough that all the current usages of the term are encompassed.
secondary spread/secondary attack rate/secondary transmission
Definitions for these analogous terms were not found in most of the sources consulted. In individual studies, secondary transmission can be defined for the study in question, but those definitions may not be consistent across studies. The most restrictive definition limits secondary transmission to cases arising from direct human-to-human contact between a primary case (infected or ill) and the secondary case who becomes infected or ill from that contact. Broader definitions include secondary cases that arise from contact with fomites or contaminated food or water. In the cases where secondary transmission includes infection from pathogens from the environment, the case would not be considered secondary unless it occurs in the context of an outbreak where primary cases have already been identified. It may be difficult to distinguish between primary and secondary cases. For pathogens that infect only humans (no zoonosis or vectors) all infections ultimately arise from human to human contact or human to environment to human pathways. In these cases it is useful to limit the definition of secondary cases based on time and distance. For example, after a primary case is identified, only cases that occur within the time window of the primary case being infectious would be considered secondary transmission. For cases that occur during outbreaks a credible scenario for exposure between the primary and secondary cases would be important. Because there are many legitimate ways to define secondary transmission, it is important that the definition be elaborated for the situation being presented.