Ranking Spacecraft Contaminants For Risk Assessment
RESPONSIBILITY for setting priorities and selecting chemicals for risk assessment rests with risk managers at the National Aeronautics and Space Administration (NASA), not with those doing human risk assessments for particular exposures. Setting priorities for risk assessment is a separate function from conducting risk assessments, which in turn is a separate function from making risk management decisions.
There are insufficient resources to conduct formal risk assessments on all potential water contaminants, so priorities must be established among them. The conclusion that this is necessary is based on several assumptions. First, the number of identifiable chemical species, and hence the potential number of hazardous substances in the space station, is large. Second, the number of potential substitutes for these chemicals is larger. Third, there could be several adverse outcomes for any chemical or mixture of chemicals. Finally, in the conduct of a formal risk assessment, the basis for choosing between chemicals and their substitutes, is a substantial exercise, particularly if it requires collecting new experimental data or new information on ambient exposure.
There are many approaches to setting priorities for choosing candidates for formal risk assessment. The approaches range in complexity.
At one extreme, priorities are set by expert judgment. Although expert judgment could result in acceptable choices, the information, assumptions, and logic involved are not necessarily specified. At the other extreme, priorities can be generated by complex and well-specified formulas. Substantial information and resources may be required. Each system has benefits and limitations, and a successful system for ranking candidates for risk assessment will combine approaches in establishing a strategy for establishing priority among chemical candidates that should be subjected to quantitative risk assessment.
APPROACHES TO RISK PRIORITIZATION
We describe three methods for setting priorities below. These examples are meant to describe methods of increasing formality for setting priorities.
Ad Hoc Approach
Candidate chemicals for risk assessment would be proposed as the candidates become of interest to NASA staff. As there would be more candidates than resources available to conduct risk assessments, there would be backlog of chemicals to which new nominees would be added. Periodically, candidate chemicals would be chosen for actual risk assessment. The decision makers would make subjective, qualitative, and, presumably, wise decisions. The parameters or the data elements upon which candidates would be chosen would not be specified, nor would chemical candidates competing for a formal risk assessment be weighed against each other in a quantitative sense. However, in this scenario the decision maker is none the less making difficult and complex choices.
Ad Hoc Approach with Factors Specified
A further step toward formality in setting priorities would be for NASA to specify the parameters it would consider. They might in-
clude evidence of exposure, magnitude of routine and accidental exposures, short- and long-term effects, ability to monitor and control exposure, and the need to have the substance on board. An example of a priority system in which the elements are specified but their weights and interrelationships are not specified comes from the National Research Council (NRC) report Carcinogens and Anticarcinogens in the Human Diet (NRC 1996). Nine criteria, including “extent of occurrence and use patterns; known human carcinogenicity, but no animal data” are listed. These are based on criteria used by the International Agency for Research on Cancer (IARC 1984) for selecting agents for carcinogenicity testing.
Formal System with Parameters, Weights, and Interrelationships Specified
In a formal system, priorities would be based on a specified set of parameters, a formula would be developed that combines scores for the various parameters, and the relationship and weighting of the parameters would be specified. The formula could be as simple as the sum of scores of various parameters or a more complex formula in which parameters are given unequal weights and their relationship would be other than additive. For example, two parameters could be the possibility for accidental exposure and the relative exposure that represents an immediate danger to life. These two parameters could be weighted equally, or danger to life could be given greater weight. Priority scores might be the mathematical sum of adding scores for the two parameters, weighting them unequally, and then adding them again, or a product derived by multiplying rather than adding. The parameters, weights, and relationships are set by the NASA authority responsible for setting priorities.
An example of a formula-based index for ranking carcinogens is the Permitted Exposure/Rodent Potency (PERP) (Gold et al. 1994, 1997). PERP is the result of division of the permissible lifetime exposure established by the Occupational Safety and Health Administration (OSHA) of the U.S. Department of Labor (in milligrams per kilogram per day) by the lifetime dose that induces tumors in 50% of animals. The PERP was proposed as a method for ranking potentially hazard-
ous exposures rather than a detailed assessment of risk. Similar indices could be developed to facilitate ranking exposures for risk assessment in other areas by using permissible exposure, such as in water, and measures of toxicity from experimental studies.
For example, a list of organic contaminants found in water during a 60-day chamber study is provided in Chapter 2 (Table 2-12). Dividing the value for each contaminant found by a corresponding acceptable level would provide a ratio, similar to a PERP, that could then be as part of a ranking system. There are several possible parameters:
Likelihood of routine exposure: Potential for routine exposure is based on a thoughtful examination of the spacecraft environment, including what chemicals are carried onboard; which chemicals are predicted to be produced by routine operations or as the result of short-term experiments; and what exposures occurred on past flights, including frequency, amount of exposure, and severity of consequences. Exposure can be expressed as a multiple of an existing appropriate U.S. Environmental Protection Agency standard, as on OSHA standard, or as a Threshold Limit Value (TLV) established by the American Conference of Governmental Industrial Hygienists.
Medical intelligence: Periodically, information will become available from ground-based experience, such as toxicologic testing, or from flight-based experience, such as occurrence of symptoms. This information should be reviewed and used in setting priorities for risk assessment.
Likelihood of unusual exposure: Unusual exposures can result from accidents and other unwanted episodes. Many of these exposures are predictable, such as those that would result from fires and leaks.
Severity of toxicity: Knowledge of the severity of the toxicity, including considerations of reversibility and ability to perform during and after exposure, is critical.
Design requirements: The physical design of the spacecraft should be a major consideration in the selection of chemical candidates for risk assessment. A primary concern is the effectiveness of systems in producing or limiting exposure in the spacecraft to specified levels. Clearly, a decision of what constitutes
acceptable exposure must be considered in the design phase of mechanical and other systems. It could be that the timing of the design phase will in part force selection of chemical species for formal risk assessment. Once the design is completed, and later when the system is built, the system's capacity for controlling and eliminating exposures should determine the choice and use of candidate chemicals during flight.
Special spaceflight considerations: The special circumstances and consequences of spaceflight, such as microgravity and its effects on calcium metabolism, should affect choice of candidate chemicals.
Spaceflight experience: This includes onboard and ground-based experience.
FLEXIBILITY OF RISK PRIORITIZATION
Flexibility in ranking is encouraged to increase the effectiveness of risk assessment. Flexibility should be manifest in several ways. New information might include changes in the parameters, changes in information about specific agents, and the addition of new substances for consideration. One might foresee a system that produces a series of priority rankings based on changes in parameters considered, their weighting, or their relationships. Priority rankings could then be compared in a way akin to sensitivity analysis in mathematical risk assessment.
We have described a range of approaches for choosing candidates for risk assessment. One choice is subjective, and the elements considered are not specified by the expert decision makers. In a second approach, the parameters to be considered are specified but their weighting and interrelationships are not. The third approach is more formulaic and involves specifying and quantifying the elements that are considered in the decision as well as the weighting of the elements in the decision making. The subcommittee recommends the use of a combination of these approaches to rank chemicals for risk assessment.
Gold, L.S., G.B. Garfinkel, and T.H. Slone. 1994. Setting priorities among possible carcinogenic hazards in the workplace . Pp. 91-103 in Chemical Risk Assessment and Occupational Health: Current Applications, Limitations, and Future Prospects, C. M. Smith, D. C. Christiani, and K. T. Kelsey, eds. Westport, CT: Auburn House.
Gold, L.S., T.H. Slone, N.B. Manley, G.B. Garfinkel, L. Rohrbach, and B.N. Ames. 1997. Carcinogenic potency database. Pp. 1-606 in Handbook of Carcinogenic Potency and Genotoxicity Databases, L.S. Gold and E. Zeiger, eds. Boca Raton, FL: CRC Press.
IARC (International Agency for Research on Cancer). 1984. Chemicals and Exposures to Complex Mixtures Recommended for Evaluation in IARC Monographs and Chemicals and Complex Mixtures Recommended for Long Term Carcinogenicity Testing. IARC Intern. Tech. Rep. No. 84/002. Lyon, France: International Agency for Research on Cancer.
NRC (National Research Council). 1996. Carcinogens and Anticarcinogens in the Human Diet. Washington, DC: National Academy Press.