HomeAbout SRAMembershipEventsNewsletterJournalOpportunitiesResourcesContact

Back to 2004 Meeting Main Page

Society For Risk Analysis Annual Meeting 2004

Session Schedule & Abstracts

P3 - Poster Session 3 Ballroom Foyer

Regulatory Policy & Decision Making

P3.1  Decision-Making Support on Risk Based Land Management with the Use of PRANA GIS-DSS. B.I.* Yatsalo, O.A. Mirzeabassov, I.V. Okhrimenko, I.A. Pichugina, D.V. Didenko; Obninsk State Technical University of Nuclear Power Engineering   yatsalo@obninsk.ru Abstract: Monitoring data and assessments demonstrate the situation on the territories of Bryansk region (Russia) contaminated after the Chernobyl accident (ChAcc) is far from satisfactory. Fraction of milk for 5 districts of Bryansk region with contamination above formal DIL exceeded in 2001 20% (in Novozybkov district - 60%); about 50% of the rural population live in settlements with a mean dose above 1 mSv/y. In this situation implementation of Decision Support System (DSS) for decision-making support (DMS) on Risk Based Land Management (RBLM) and optimisation of countermeasures (CM) within the rehabilitation of contaminated territories is topical. PRANA is a unique DSS developed with the use of GIS which comprises all the main territories of Russia subjected to radioactive contamination after the ChAcc and includes all the tools necessary for RBLM and CMs optimisation on local and regional levels. The main blocks of the PRANA DSS are: - vector electronic maps of landuse (more than 30,000 polygons); - databases (radiological, ecological, economic, demographic and other data and model parameters); - model blocks for assessing: contamination of agricultural produce, doses to the population, risk and results of CM implementation; - evaluation of CM effectiveness and DMS (local and regional levels - from separate settlement/field up to region as a whole). The following risk indices are considered within assessments of the radiological situation and DMS: - surface density contamination (137Cs, 90Sr); - contamination of agricultural produce; - external and internal doses to the population; - radiological risks caused by irradiation of the population; The assessments of risk indices (before and after CM) and evaluation of various CMs/alternatives are carried out. Ranking fields, farms, settlements and districts are carried out with the use of different criteria. Several versions of PRANA DSS have been developed: for practical implementation, for research, and for training and education.

P3.3  Insider censoring: Distortion of data with nondetects. D.R. Helsel; US Geological Survey   dhelsel@usgs.gov Abstract: Environmental data often include low-level concentrations below reporting limits. These data may be reported as “<RL”, where RL is one of several possible reporting limits. The data also may be reported as a single number, but flagged with a qualifier (J-values) implying a difference from values above the RL. A currently used method for reporting censored environmental data (“insider censoring”) produces a strong upward bias, while also distorting the shape of the data distribution. This bias results in inaccurate estimates of summary statistics and regression coefficients, distorts evaluations of whether data follow a normal distribution, and introduces inaccuracies into risk assessments and models. Insider censoring occurs when data measured as below the detection limit (<DL) are reported as less than the higher quantitation limit (<QL), while values between the DL and QL are reported as individual numbers. Three unbiased alternatives to insider censoring are presented so that laboratories and their data users can recognize, and remedy, this problem.

P3.4  Evaluating Risk/Risk Tradeoffs in Fish Consumption Advisories to Address the Risks of Methylmercury. J.T. Cohen, G.* Gray, C. Bouzan, G. Rice, A. König; Harvard Center for Risk Analysis   ggray@hsph.harvard.edu Abstract: Methylmercury is present, at varying levels, in virtually all fish and seafood. Concerns exist about potential neurodevelopmental and other effects of methylmercury exposure. Fish is also a food with potential health benefits to infants, children and adults. A common risk management strategy is use of fish consumption advisories to provide information about appropriate fish consumption to the general public. We have undertaken a quantitative analysis of the net public health effects of changes in fish consumption, weighing both the risks of methylmercury exposure and the risks of decreases in fish consumption. A group of experts helped develop dose-response relationships for the links between methylmercury and neurodevelopmental effects and fish (or fish components) and neurodevelopment, stroke, and coronary heart disease. These relationships were used to estimate the changes in population health in both natural units (e.g., changes in IQ) and changes in total population Quality Adjusted Life Years (QALYs) under different changed fish consumption scenarios. The National Food Processors Association Research Foundation funded this work.

P3.5  Lessons Learned from MTBE: The Need for a Comprehensive Assessment Approach. J.M. Davis; U.S. Environmental Protection Agency   davis.jmichael@epa.gov Abstract: Methyl tertiary butyl ether (MTBE) became a widely used gasoline additive during the 1990s because it was thought to reduce emissions and improve air quality. The experience with MTBE, however, has been marked by consumer complaints about health symptoms and unpleasant odors, as well as by many reports of MTBE-contaminated water. Although actions to ban MTBE and use alternative formulations have been taken or are being contemplated, any fuel choice has inherent trade-offs. Weighing the positive versus negative environmental impacts of various fuel options requires a comprehensive approach that combines a product life-cycle perspective with the risk assessment paradigm. Such an approach would evaluate the ecological and health risks associated with primary releases/emissions, along with secondary by-products, of different fuel formulations throughout their respective life cycles. This presentation will focus on the general features of a comprehensive assessment of oxygenate choices for reformulated gasoline: MTBE, ethanol, and no oxygenate. The potential for unintended consequences of new or alternative technologies such as fuel oxygenates will remain high unless they are examined in a systematic, comparative manner that identifies risks across environmental media for ecosystems as well as human populations.

P3.6  Potential Human Health Benefits of Antibiotics Used in Food Animals: A Case Study of VIrginiamycin. L.A. Cox; Cox Associates   tony@cox-associates.com Abstract: Risk management of food-animal antibiotics has reached a crucial juncture for public health officials worldwide. While withdrawals of animal antibiotics previously used to control animal bacterial illnesses are being encouraged in many countries, the human health impacts of such withdrawals are only starting to be understood. Increases in animal and human bacterial illness rates and antibiotic resistance levels in humans in Europe despite bans on animal antibiotics there have raised questions about how animal antibiotic use affects human health. This paper presents a quantitative human health risk and benefits assessment for virginiamycin (VM), a streptogramin antibiotic recommended for withdrawal from use in food animals in several countries. It applies a new quantitative Rapid Risk Rating Technique (RRRT) that estimates and multiplies data-driven Exposure, Dose-Response, and Consequence factors, as suggested by WHO (2003), estimate human health impacts from withdrawing virginiamycin. Increased human health risks from more pathogens reaching consumers if VM use is terminated (6660 estimated excess Campylobacteriosis cases per year in the base case) are predicted to far outweigh benefits from reduced streptogramins-resistant vancomycin-resistant E. faecium infections in human patients (0.27 estimated excess cases per year in the base case). While lack of information about impacts of VM withdrawal on average human illnesses-per-serving of food animal meat precludes a deterministic conclusion, it appears very probable that such a withdrawal would cause many times more human illnesses than it would prevent. This qualitative conclusion appears to be robust to several scientific and modeling uncertainties.

P3.7  A Risk based Priority Ranking and Scoring Method and its Applicability to Hazardous Air Pollutants in Korea. HS Park, YS Kim, DS Lee, SS Choi, DC Shin; Yonsei University   mars0809@hotmail.com Abstract: It has recently been suggested that there is a need to set risk-based plans for the effective management of air pollutions, but to evaluate all pollutants is impossible due to the lack of resources. Therefore, the setting of a priority ranking method should be proposed for the effective management of air pollutants. In this study, a human health risk-based priority setting method has been developed for hazardous air pollutants, which was derived from the priority lists for substances in the TRI (Toxic Release Inventory) of Ulsan Metropolitan City, in Korea. Finally, the applicability of the developed method was evaluated. This was achieved by a comparison of two priority lists, one from a priority setting method developed by us, and the other being a quantitative risk assessment. The priority lists were categorized into 3 groups (high, medium, low), and Kappa coefficients of concordance were calculated. As results, the concordance was very low as negative Kappa coefficient, but the explanation rate of chemical ranking and scoring method to quantitative risk assessment was over 50%, and when separating human health effect into carcinogenic and non-carcinogenic, the concordances were a little higher than that for combined effects. Therefore chemical ranking and scoring method proposed in this study could set an appropriate priority list needed to conduct risk assessment in further.

P3.8  Risk versus Safety: the Use and Abuse of Precaution in Regulation. P.K. Keese; Office of the Gene Technology Regulator   paul.keese@health.gov.au Abstract: Precaution, once a fairly benign term, has been reincarnated with new teeth in the form of the Precautionary Principle (PP). The PP emphasises the primacy of protection (safety) over uncertainty of serious harm (risk). Consequently, it resurrects an underlying tension for regulation of new technologies, whether risk or safety assessment is the most appropriate driver of decision-making. One response has been to place protection of human health and the environment as the objective and use risk assessment/risk management as the mechanism to achieve that objective. The unstated assumption is that Safety and Risk are complementary. However, this is not entirely valid. Safety is deterministic whereas Risk is probabilistic. Safety is concerned with the absence of harm whereas Risk looks at the balance of harmful and beneficial outcomes. Safety is not directly observable and is applied where not all undesired outcomes may be known. In contrast, Risk is perceived as objective and measurable. Consequently, we construct some unwieldy chimera and PP its unlikely offspring. However, precaution in the old fashioned sense is just one of several protective measures available to a regulatory body. These can include, prevention of the activity, or some part of it, from occurring (prohibition); cautionary steps before (precaution) and while (caution) the activity occurs; monitoring for adverse effects arising from the activity (surveillance); and rectifying any adverse effects that do arise (remediation). But these regulatory tools come at a price. Precaution incurs direct economic costs of implementation; loss of potential benefits; increased regulatory burden; constraints on creativity; and reduced civil liberties in choice and accessibility to the products or applications of the technology. The task facing a regulatory body is marking that boundary between protection and opportunity, guided by new insights into uncertainty in its broadest sense as a form or source of doubt.

P3.9  Proposed Public Health Goals for Radium-226 and -228, Strontium-90, and Tritium in Drinking Water. R.A.* Howd, J. Jowa, A. Avalos, A.M. Fan; Office of Environmental Health Hazard Assessment, Cal/EPA   bhowd@oehha.ca.gov Abstract: Risk assessments were performed for the radionuclides radium-226 and –228, strontium-90, and tritium for development of public health goals (PHGs) in drinking water. PHGs are estimated concentrations in water that are not anticipated to produce adverse effects following lifetime exposure, and are advisory levels based exclusively on health considerations. Literature review for toxicological evaluations included both animal and human studies, and carcinogenic and non-carcinogenic endpoints. The proposed PHGs are based on a 1 in 1 million lifetime cancer risk from radiation, which is calculated from the radionuclide potency values in US EPA’s Federal Guidance Report 13 (1999). We also determined non-cancer health-protective levels. The proposed PHGs are: radium-226, 0.069 pCi/L; radium-228, 0.019 pCi/L; strontium-90, 0.35 pCi/L; and tritium, 400 pCi/L. These levels are judged to be protective against all non-cancer adverse effects, for which the health-protective values are 200 pCi/L for both radium isotopes based on bone necrosis; 600 pCi/L for strontium-90 based on shortened lifespan; and 6x10^7 pCi/L for tritium based on shortened lifespan and hematopoietic effects. The assessments included consideration of sensitive populations and drinking water consumption rates. In the past 10 years, the two radium isotopes have occasionally been found in California at levels above their combined MCL of 5 pCi/L (with a detection limit of 1 pCi/L), while strontium-90 was reported only once above its MCL of 8 pCi/L (detection limit 2 pCi/L), and tritium was not reported above its MCL of 20,000 pCi/L (detection limit 1,000 pCi/L). Monitoring methods are not sensitive enough nor are the data extensive enough to conduct a detailed exposure and risk assessment. Comments on the risk assessment methods and the proposed PHG documents (see www.oehha.ca.gov/water) are invited.

P3.10  A Framework for Environmental Microbiological Hazard Evaluation: Public Health Investigation and Microbial Risk Assessment of Municipal Solid Waste Truck Leachate. JA Shatkin, N Moyer; The Cadmus Group, Inc.   jshatkin@cadmusgroup.com Abstract: Fecal coliform counts in municipal solid waste (MSW) and construction and demolition (C&D) leachate from waste hauling suggest the potential for public health and environmental threat to exposed communities. To evaluate this potential hazard, traditional public health investigation and risk assessment were used together to evaluate whether, or under what conditions, waste hauling truck leachate might pose a threat to public health. One approach reviewed the current state of knowledge on the contents of MSW, leachate, and indicator bacteria, and their potential health implications; the other used quantitative risk assessment to estimate potential exposure to microbial pathogens and compare exposure levels to infectious dose levels. This dual framework of evidence-based and risk-based approaches offers a novel approach for investigation of microbial hazards in environmental media with increased confidence for regulatory decision making under uncertainty. Results of this investigation will be reported, as will the implications for future evaluation of potential biological threats within the framework.

P3.11  Addressing Cancer Risks from Screening Levels of Radioactivity in Drinking Water Supplies. L.* Jowa, R.A. Howd, A.M. Fan; Office of Environmental Health Hazard Assessment, Cal/EPA   bhowd@oehha.ca.gov Abstract: We reviewed and evaluated the categories of gross alpha and gross beta/photon emission represented by California maximum contaminant levels (MCLs) for these radionuclide groupings. The evaluation included toxicological and risk factors associated with the existing MCLs developed by US EPA (effective in 1977) for screening purposes, not as health protection standards. It also relied on the latest US EPA radioactive potency factors for carcinogenicity of the various radionuclides. The screening levels are 15 pCi/L for gross alpha and 50 pCi/L for gross beta/photon emission. Screening levels are by definition analytical tools, not final regulatory limits. In practice, if the screening levels are exceeded, then additional analyses are performed to determine specific radioisotope levels in drinking water. Public Health Goals are being developed separately for several of the more common radioactive elements, i.e., those that have specific MCLs. Our evaluation concluded that the present screening level MCLs for both gross alpha and gross beta/photon emission are far in excess of the de minimis risk level of 10-6 (one in a million) for lifetime cancer risks from radioisotopes commonly found in drinking water. Actual cancer risk from radionuclides in drinking water depends on the particular radionuclide(s) present, and the average consumption over a lifetime. The radionuclides detected in drinking water are largely natural radioisotopes, since the manmade radioisotopes resulting from fallout from atmospheric testing of nuclear bombs have decreased considerably since the discontinuation of testing. Lower screening levels, closer to the de minimis risk levels, would alert water providers to the presence of relatively high radioactivity risk levels in samples. This would trigger more extensive analyses, and collection of data adequate for risk assessment.

P3.12  Proposed Public Health Goals for cis- and trans-1,2-Dichloroethylene, 1,1,1-Trichloroethane, and 1,1,2-Trichloroethane in Drinking Water. C.A.* Caraway, T Parker, R.A. Howd, A.M. Fan; Office of Environmental Health Hazard Assessment, Cal/EPA   bhowd@oehha.ca.gov Abstract: Risk assessments were performed for the solvents cis- and trans-1,2-dichloroethylene (1,2-DCE), 1,1,1-trichloroethane (1,1,1-TCA), and 1,1,2-trichloroethane (1,1,2-TCA) for development of public health goals (PHGs) in drinking water. PHGs are non-regulatory concentrations in water that are not anticipated to produce adverse effects following lifetime exposure, and are based exclusively on health considerations. Literature review and toxicological assessments included both animal and human studies. The isomers cis- and trans-1,2-DCE both exhibit hepatotoxic effects in rodents; standard cancer bioassays have not been carried out. The proposed PHGs are 100 ppb for cis-1,2-DCE and 60 ppb for trans-1,2-DCE, based on hepatotoxicity. The most sensitive toxic effects of 1,1,1-TCA in the available studies are neurological. The proposed PHG for 1,1,1-TCA is 1,000 ppb, based on astrogliosis in the cerebral cortex of gerbils in a subchronic inhalation study. Three cancer bioassays have been performed, but were judged inadequate by both OEHHA and US EPA. Well-conducted cancer bioassays for 1,1,2-TCA showed increased incidences of hepatocellular tumors and pheochromocytomas in mice, but no increased tumor incidence in rats. The proposed PHG for 1,1,2-TCA is 0.2 ppb, based on an estimated one in one million lifetime cancer risk. The health-protective level for 1,1-2-TCA, considering only non-cancer effects, would be 50 ppb, based on hepatotoxicity. Comments on the risk assessment methods and the proposed PHG documents (see www.oehha.ca.gov/water) are invited.

P3.13  Uncertainty in Input Data on Contaminant Physical and Chemical Properties: Implications for Modeling and Risk Assessment . I Linkov, M Ames, E Crouch, L Green; Cambridge Environmental Inc.   Linkov@CambridgeEnvironmental.com Abstract: The models used in RA require input data on a variety of physical and chemical properties for each compound of concern. Small changes in some of these parameters can result in significant differences in estimated human health or ecological risks. The octanol-water partition coefficient (Kow) is one such uncertain parameter, particularly because it is often used to estimate additional partitioning and bioaccumulation parameters. Most risk assessments rely on deterministic calculations with parameter values taken from tabulations either compiled by the U.S. EPA or other regulatory agencies or from literature sources. The EPA has published or recommended the use of several databases and documents that report Kow values for compounds of interest, but for some chemicals there is a wide range of tabulated Kow values even among these sources. This paper assesses the implications of using various values of Kow to calculate sediment quality objectives (SQOs) for a PCB-contaminated site in the coastal waters of Puget Sound. A simplified food chain model was implemented using the range of Kow values available from, or recommended by, the U.S. EPA. The output of this model provides exposure point concentrations used for estimating human health risks through ingestion of locally caught game fish. SQOs were calculated as the maximum PCB concentration in sediments that would be associated with a health-protective level of risk. The SQOs obtained for the lowest Kow value reported in EPA tabulations are as much as a factor of five higher than those obtained with the highest Kow value. For the site considered in this study, this range of SQOs is estimated to correspond to a difference in remediation costs of $48 million. Standardization of Kow estimates among EPA databases and estimation programs would result in more consistent selection of remedial alternatives among sites and would ease the process of regulatory risk assessment review.

P3.14  Occupational and Hereditary Risk Factors for Chronic Beryllium Disease: a Monte Carlo simulation exploring the combination of exposure and genetic data from disparate sources. T. DeFoe; Carnegie Mellon University   ttummino@andrew.cmu.edu Abstract: Various researchers have identified a relationship between chronic beryllium disease (CBD), a serious and sometimes fatal respiratory disease associated with exposure to beryllium (Be), and certain inherited genetic factors, including an allele coding for glutamic acid at the 69th position of the Human Leukocyte Antigen molecule HLA-DP (HLA-DPB1(Glu-69)). OSHA is presently collecting data to evaluate risk to workers under the current permissible exposure limit (PEL) of 2 micrograms/m3 (8-hour TWA) and assess the potential effects of alternate exposure limits. The Agency’s risk assessment would benefit from information on risks experienced by beryllium-exposed individuals with hereditary risk factors; however, there is currently insufficient data to identify or characterize differential dose-response relationships among workers with and without these risk factors. This poster presents a Monte Carlo simulation modeling the occurrence of CBD under the current beryllium PEL and several alternate exposure limits, using assumptions based on information in the available literature to conduct analyses for workers with and without the HLA-DPB1(Glu-69) allele. The model draws on exposure, genetic, and disease data from various studies in the published literature and assesses the sensitivity of the results to several model inputs. The reliability of the results and their relevance to the regulation of occupational beryllium exposure are discussed.

P3.15  The Food Safety Risk Analysis Clearinghouse: Practical Uses in Risk Analysis.. J Hinton, CF* McLaughlin, H Patel; University of Maryland, FDA, University of Maryland   cmclaugh@cfsan.fda.gov Abstract: Food Safety Risk Analysis is a multidisciplinary field with multiple dimensions. The Food Safety Risk Analysis Clearinghouse is an invaluable resource to professionals in food safety risk analysis, as it hosts information across the relevant dimensions within the Risk Analysis framework. The Goal of the Clearinghouse is to encourage the use of risk analysis in the development of food safety policy by facilitating access to information that might otherwise be difficult to find. Reasons that enable the Clearinghouse user access to more relevant data include the way it categorizes information and the two different search engines it uses. The Clearinghouse hosts data, tools, reports, select Federal Register documents, and presentations, and serves as a vast and authoritative portal to online resources that support the work of the food safety risk analysis professional. Its domain includes both chemical and biological hazards in foods as well as nutrition and weight management as related to food. Using both biological and chemical hazard examples, this presentation will provide an introduction to the Clearinghouse and will demonstrate how to effectively use the Clearinghouse to find resources and data that will support all steps of the risk analysis process.

P3.16  Creating resuable code for risk assessment of disease transmission in cattle. C.* Marooney, R.D. Jones, R. Clifton-Hadley; Veterinary Laboratories Agency   c.marooney@vla.defra.gov.uk Abstract: Standard methods of constructing risk assessment models using spreadsheets with specialist analysis add-ins are difficult to reuse for similar problems in the same technical area, in addition to not benefiting from full optimization of the calculation. Compiled code using object-oriented technology provides an improved route to reusing and enhancing calculation performance. This paper describes the implementation of an assessment of the risk of a bacterial disease being transmitted under a variety of testing regimes, using this alternative methodology, together with Monte-Carlo simulation applied to an individual animal model. The resulting model could be used to estimate the consequences of changing the current assumed frequency of testing dairy cattle once per month, to reduced frequencies of either every six months, or twice every two or every five years. The paper focuses on the flexibility of the software design for dealing with shifts of emphasis between within-herd and between-herd transmission, and between different aspects of the cow's lifecycle. This flexibility also means that the resulting model can readily be adapted to other diseases of cattle. In addition, alternative sources for statistical libraries and harnesses are also reviewed.

P3.17  An approach to assess the health risks associated with the remediation of contaminated sites . J David; Health Canada   jacinthe_david@hc-sc.gc.ca Abstract: In Canada, the federal government announced several initiatives with a total funds available of $3.5 Billions over 10 years, to clean-up federal contaminated sites. In this context, Health Canada is a key player in providing expert health advice to custodial departments and is promoting the consideration of health risks associated with living and working at/or near contaminated sites. Prior to remediation of contaminated land, the Regulations under the Canadian Environmental Assessment Act stipulates (article 41.1)that an environmental assessment must be conducted. An environmental assessment provides an opportunity to the federal authority to address the environmental effects of the project, i.e. the remediation. Health Canada has developed an approach to assess the risks associated with the proposed site-specific remediation technologies. This approach can identify the site-specific human health impacts related to the remediation technologies. This provides a means of weighing technology risks against site risks and the ability to balance risks that exist against those that could inadvertently be created through clean-up. This approach provides information on impacts/releases for each site remediation technology and methods for occupational and residential receptors.

P3.18  Development and Use of Risk Based Inspection Models by a Public Safety Authority in Canada. S Mangalam, R Mazumdar, R Feo, R Neate, M Scarland; Technical Standards and Safety Authority   smangalam@tssa.org Abstract: The Technical Standards and Safety Authority (TSSA) is an independent, not-for-profit organization that administers and enforces public safety laws associated with Amusement Devices, Elevating Devices, Boilers and Pressure Vessels, Fuels, Operating Engineers, and Upholstery and Stuffed Articles, under Ontario’s Technical Standards and Safety Act through an administrative agreement with the Ontario Government. TSSA conducts a wide variety of safety activities in each of the regulated sectors. One of these activities involves periodic inspections of facilities and equipment during operation to ensure that the device continues to operate safely. The inspection process reviews how the equipment and technology are being used, operated and maintained, and identifies any non-compliance to safety codes and regulations. TSSA has designed and implemented risk-based inspection scheduling models across most of its industry sectors to ensure that devices that present the highest risk to the public are inspected more often while low-risk units are assigned longer inspection cycles. The model development and implementation was carried out in three stages: 1) Concept development based on identification of risk factors representing the devices and equipment, human-device interaction, and location; 2) Model design using a semi-quantitative risk assessment approach; and 3) Implementation, including automated scheduling, monitoring, and measurement. This presentation will describe the three stages with specific reference to the models developed for the elevating devices and fuels sectors.

P3.19  Particulate Zinc Speciation In A Heavily Polluted Lake In The Nile Delta. M.A.H.* Saad, A.E. Safty; University of Alexandria, Faculty of Science   saad@internetalex.com Abstract: Lake Mariut, a northern Nile Delta lake south of Alexandria, is feeding mainly by anoxic and heavily polluted drainage waters from Qalaa Drain, beside domestic and industrial wastewater. Monthly variations of total suspended matter (TSM) and particulate zinc (PZn) speciation were investigated. The increase in TSM in the discharged drainage waters elevated the levels of TSM in the lake water. The TSM peak in July coincided with the high association of phytoplankton biomass in summer. Because the risk of a metal can be ascertained from studying the distribution of its different species, a five step sequential extraction scheme was applied to TSM to illustrate the contribution shared by each individual species in its total zinc concentration. Similar to the conditions in the lake sediments, the exchangeable fraction (F1) in the lake water was the least among the other zinc species, giving a negative poor correlation with pH. The percentage of the bound to carbonate fraction (F2) was less than 25%, giving a high significant correlation with organic matter / sulfide fraction (F4) and the empirical regression equation of: F2 = 420.7 ± 187.8 + 0.896 ± 0.26 F4. F2 showed the highest peak in March, due mostly to spring blooms of phytoplankton. The factor affecting the release of F2 is temperature, as confirmed from its correlation with F2. The part shared by the Fe / Mn oxides fraction (F3) in building up PZn was monitored through the positive correlation between PZn and this fraction. The correlation coefficient , similar to F2, indicates less contribution of F3 in building up the PZn in the wetland. The prevailing anoxia in this coastal wetland was the motive behind determining the definite regression relationship between F4 and each of dissolved oxygen, hydrogen sulfide, BOD and COD, which all proved to be non significant. The diminishing of the inert fraction (F5) in some months supported the increase in the other mobile dangerous fractions of zinc.

P3.20   A transdisciplinary study of consumer beliefs, behavior and microbial contamination in preparing a meal.. A. R. H. * Fischer, A. E. I. de Jong, L. J. Frewer, M. J. Nauta, R. de Jonge; Wageningen University Marketing and Consumer Behaviour Group and RIVM (National Institute for Public Health and the Environment)    arnout.fischer@wur.nl Abstract: The prevention of consumer food poisoning is an important issue from the point of view of public health. For this purpose, information interventions targeted at the behavior of consumers during food preparation are an essential part of developing an effective intervention strategy. However results from psychological research have shown that the development of effective intervention strategies is problematic, and indicate that consumers do not readily adopt safe food preparation practices. To be successful in communication relevant risk information, targeted campaigns should be developed in which technical risk assessments and risk perception/communication are integrated. To achieve this, close cooperation between, microbiologists and risk psychologists is required. The primary aim of the research is to explore and try to understand the relations between consumer practices, the resulting microbial contamination, and their risk perceptions regarding current food preparation practices. A study will be reported in which these relations were explored by asking Dutch consumers to prepare a chicken salad in their own kitchen. Cross-contamination during food preparation was microbiologically followed by contaminating raw chicken with a harmless tracer, as a substitute for Campylobacter. Participants were then interviewed to explore their beliefs with regard to food and food safety. By analyzing videotapes of the food preparation, the microbial contamination of the finished meal, and the interview texts, the relations between behavior and microbial effects and between food related beliefs and behavior are explored, which leads to identification of actual hazards and increased understanding of the relation between behavior and beliefs of consumers. Results and conclusions of this study will be presented, together with implications for future research.

P3.21  The Risk of Occupational Fatalities Unanticipated by Regulation or Codes of Practice. A.M. Finkel; U.S. Occupational Safety and Health Administration* (personal views only, not those of the U.S. government)   finkel.adam@dol.gov Abstract: A three-part series (December 2003) on OSHA in the New York Times highlighted the Agency’s long-standing reluctance to pursue criminal penalties against companies who willfully place employees in grave danger. Various legislators have introduced proposals to increase the severity of such penalties if issued, and to increase OSHA’s interest in the possible deterrent value of zealous criminal enforcement. Such measures, however, are essentially reactive and do not directly reduce the probability that a given workplace tragedy will repeat itself—especially in cases where the fatality was caused wholly or predominantly by factors other than non-compliance with regulations or recognized codes of practice. I will analyze several years of case reports of fatalities inspected by OSHA, in order to: (1) estimate the proportion of fatalities that occur because of product defect or design issues, or other factors that lie outside the existing (or reasonably plausible) scope of federal and state regulatory requirements; and (2) identify specific circumstances that have resulted in a pattern of fatalities, and where targeted educational and other outreach efforts might break this pattern. Based on this analysis, I will recommend process improvements whereby OSHA, NIOSH, and the state OSHA plans could share information and develop coordinated responses to fatalities that will likely recur absent some effort to inform employers and workers of unrecognized or unappreciated hazards.

P3.22  Contingent Valuation as a Tool to Link Risk Assessment and Economics. K.E.* von Stackelberg, J.K. Hammitt; Harvard Center for Risk Analysis and Menzie-Cura & Associates, Inc.   kvon@hsph.harvard.edu Abstract: There is a need for tools and methods that link human health and ecological risk assessment with economic valuation for the purpose of evaluating potential risk reductions associated with management actions. The objective of the proposed research is to contribute to the understanding of practical and credible approaches for estimating the benefits and costs of environmental policies and to improve decisionmaking regarding environmental issues by developing an integrated human health and ecological risk model using a case study approach. The case study focuses on potential human health and ecological receptor exposure to polychlorinated biphenyl (PCB) compounds via fish ingestion. The risk model integrates the results of two contingent valuation surveys to quantify the benefit of potential risk reductions for endpoints that are difficult (or impossible) to value by traditional, market-based economic methods. We describe the development of the two web-based contingent valuation surveys used to estimate willingness to pay of the general public to reduce potential risks associated with exposures to PCBs. The first survey focuses on potential developmental effects associated with in utero exposures, and the second focuses on potential reproductive effects associated with ecological receptor exposure. The results are used in an integrated risk model.

Dose Response

P3.23  Neighborhood Socioeconomic Environment and Allergen Asthma Associations: A Canadian Perspective. Sabit Cakmak, Robert Dales; Health Canada, University of Ottawa   sabit_cakmak@hc-sc.gc.ca Abstract: Cakmak Sabit*, Dales Robert** *Research Scientist, Health Canada, HECS Branch, 2720 Riverside Drive, Ottawa, ON. K1A 0K9, AL#6604C, E-mail:sabit_cakmak@hc-sc.gc.ca, Tel:613- 952 6913 ** Professor of Medicine, University of Ottawa, Canada Social status influences asthma morbidity but the mechanisms are not well understood. To determine if socioeconomics influence the susceptibility to ambient aeroallergens, we determined the association between daily hospitalizations for asthma and daily concentrations of ambient pollens and molds, in ten large Canadian cities. Daily time-series analyses was employed and results were adjusted for day-of-the week, temperature, barometric pressure, relative humidity; ozone, carbon monoxide, sulpur dioxide, and nitrogen dioxide. Results were then stratified by age, gender, and neighbourhood family education and income. There appeared to be an age-gender interaction. An increase in basidiomycetes equivalent to its mean value, about 300 /m3, increased asthma admissions among younger males (under 13 years of age) by 9.3 (95%CI 4.8, 14.1) v 4.2 (95%CI -0.1, 8.5) for older males. The direction was opposite among females with increased effect in the older age group; 5.0% (95%CI 3, 7.1) in those under 13 years v. 2.3%(95%CI -1.2, 5.8) for older females. No significant associations were seen between aeroallergens and asthma hospitalization in the highest education group, whereas in the lowest group, 4 of 6 associations were significant. In conclusion, our results suggest that younger males and those within less educated families may be more vulnerable to aeroallergens as reflected by hospitalization for asthma.

P3.24  Linear vs. Nonlinear Models of Excess Lung Cancer Risks Induced by Carbon Filter Particles in Cigarette Smoke. L.A.* Cox, G. Charnley, C.L. Gaworski; (1)Cox Associates, Denver, CO; (2)HealthRisk Strategies, Washington, DC; (3)Phillip Morris USA, Richmond VA    charles.l.gaworski@pmusa.com Abstract: Risk assessment methods often emphasize conservative estimates of risks from exposures. Yet, some cancer risk models, especially biologically-motivated (e.g., MVK) models, can help quantify realistic risks in support of product design, development, and improvement. This risk assessment examines the extent to which carbon particles that could transfer to smoke from activated carbon filters during smoking might increase the risk of smoking-related diseases among smokers. Such transfers can arise as a side-effect of cigarette designs that attempt to filter smoke more effectively. We compare the results of a simple linear model (incremental risk proportional to incremental exposure), fit to epidemiologic data on PM2.5 and carbon black risks, to those from a more sophisticated, previously published MVK-type model developed by Leenhouts that models non-linear and dynamic relationships between cigarette smoking history and age-specific hazard rates for lung cancer mortality. Although the latter model over-estimates risks to non-smokers, it appears to provide realistic estimates of risks to continuing smokers for a broad range of exposure conditions. Moreover, it predicts an almost linear relationship between incremental smoking exposure and excess risk of lung cancer by any given “lifetime” age (e.g., 75). It thereby, retrospectively gives support to the simpler linear model. Both models indicate that excess risk from carbon particle transfer is proportional (with proportionality constant less than 1) to incremental exposure, and hence is small, even taking exposure uncertainties into account. This work demonstrates the potential value of biologically motivated cancer risk modeling in understanding the effects of an intrinsically hazardous product, while identifying challenges needed to develop better models of the differences in lung cancer risks between smokers and non-smokers.

Ecological & Environmental Risk

P3.25  Dicyclopentadiene Toxicity Assessment. K.M. Johnson, M.C. Rigby; Tetra Tech, Inc.   kay.johnson@tetratech.com Abstract: Dicyclopentadiene is a by-product of synthetic rubber production and was produced throughout World War II. Contamination from its production during World War II continues to be a concern in many locations as well as its use today in chemical manufacture. A toxicity assessment was conducted for this chemical. Only a limited number of animal studies were found on the chronic and subchronic effects of dicyclopentadiene inhalation. In particular, the inhalation reference dose (RfD) was based on a subchronic study where the only major effect was male rat nephropathy. No signs of toxicity were seen in female rats. The observed effects, including dose-related tubular hyperplasia, are linked to excessive accumulation of a specific protein (i.e., alpha-2u-globulin) in male rat kidneys. Synthesis of this protein does not occur in female rats, or any other mammal species. Thus, the male rat kidney dysfunction upon which the inhalation RfD was based is not an appropriate endpoint to determine noncancer effects in humans. For oral exposures, only a limited number of studies have been conducted. The oral RfD is based on another study using rats, although no adverse effects were observed in a three-generation reproductive study. This lack of a substantial toxicity database is of concern because the risks associated with exposures to dicyclopentadiene cannot be evaluated with a high level of certainty.

P3.26  Using Toxicity Test and Bioaccumulation Data for Site-Specific Soil Screening Values for Ecological Risk Assessments. R.T. Ryti, J.T.* Markwiese, R. Mirenda, P. Newell; Neptune and Company, Inc.; Los Alamos National Laboratory; Environmental Health Associates   jimm@neptuneinc.org Abstract: Baseline ecological risk assessments for terrestrial environments often use toxicity tests as measures of effects and analyze concentrations in biota as measures of effect and/exposure. For large facilities with ongoing risk assessments, it is appealing to utilize this information to develop facility-specific ecological screening levels. This poster considers the potential problems associated with utilizing toxicity test and bioaccumulation data and compares the screening levels developed with site-specific information with values from the literature and the values proposed as ecological soil screening levels (Eco-SSLs). The case study data are from baseline ecological risk assessments completed to support corrective action decision making for the Los Alamos National Laboratory. The primary data considered for developing facility-specific soil screening levels include toxicity test results using western yarrow (Achillea millefolium), a drought tolerant forb, conducted according to ASTM method E1963-98. The test provided 10 measures for evaluating possible adverse ecological effects of COPECs on plants, including: seedling germination, mortality, average shoot height, average root length, shoot wet weight, root wet weight, total wet weight, shoot dry weight, root dry weight, total dry weight (replicate [pot] or plant). Soils were also tested with the earthworm bioaccumulation test (ASTM method E1676-97) as a measure of effect (mortality and growth) and exposure for detritivores. Site-specific toxicity tests provide important information on the effect and bioavailability of COPEC mixtures. Other data from the baseline ecological risk assessments that can be used for facility-specific soil screening levels include contaminant concentrations in small mammals and bird eggs as measures of exposure and effect. Bioaccumulation characteristics of the site will be benchmarked against literature-reported uptake factors for points of comparison.

P3.27  Implementing a Decision Based Framework to Evaluate the Impact of Over- and In-Water Construction Materials along the Connecticut Coastline of the Long Island Sound. M.E. Kozlak; Clark University   mkozlak@yahoo.com Abstract: The choice of construction materials used for over- and in-water coastal structures (i.e. docks, pilings, bulkheads, etc.) is not regulated along the Connecticut (CT) coast of the Long Island Sound (LIS). The typical construction material currently used for structures along the CT coast is Chromated Copper Arsenate (CCA) treated wood. Other construction materials exist, but are used infrequently. Stakeholders have addressed concerns about the lack of regulation on construction material choice and the wide spread use of CCA treated wood for coastal structures on the basis that it potentially violates sections of the Connecticut General Statutes (CGS) by leaching toxins that can cause adverse effects and ecological risks to the LIS environment. The available data on the exposure to contaminants from treated wood and ultimate characterization of risks in estuarine and coastline environments is highly variable and uncertain. The Connecticut Department of Environmental Protection (CTDEP) is faced with the challenge of determining if the use of treated wood violates sections of the CGS and assessing the regulatory alternatives for construction material choice. This study elicits the need to put the full implications of decisions back into the assessments that help guide them by going beyond the traditional US Environmental Protection Agency ecological risk assessment paradigm. Analyzing, instead, both beneficial and harmful impacts of the regulatory alternatives that fairly represent the construction material choices facing CTDEP and other stakeholders. This study uses a precautionary guided approach that addresses the absence of empirical or predictive certainty among regulatory alternatives within a policy/decision orientated framework to assess the impacts of treated wood and its alternatives.

P3.28  Aquatic Hazard Assessment for Selenium: Limitations of a Simplified Hazard Assessment Protocol. W.E. Wright, P.B. Stenhouse, B.N. Narloch; MWH   bruce.narloch@mwhglobal.com Abstract: Selenium is an essential nutrient at low environmental concentrations but has been implicated in several toxic episodes including avian malformations and population declines in Kesterson Reservoir, California; and fish depletions in Belews Lake, North Carolina. A simplified hazard assessment protocol was recently proposed to evaluate hazards of selenium in fish and avians. Evaluation criteria used in this protocol include toxic effect thresholds and ranges of effects-based concentrations. Selenium, represented by the maximum concentration measured in each of five environmental compartments, is compared to these evaluation criteria, and a hazard score is derived. As with any screening-level risk assessment methodology, this simplified protocol has several limitations including (1) use of maximum concentrations to represent environmental exposures, (2) failure to consider the selenium form, and (3) effect thresholds that may be inappropriate for species inhabiting a watershed. For example, the surface water toxicity threshold for fish and avians is given as 2 ug/L in the simplified protocol; 5 ug/L is assigned a hazard score of ‘high’. The reproductive threshold for selenium in dietary items of fish is given as 3 ug/g (dry wt), and 4 ug/g (dry wt) in whole body tissues. Investigations conducted on the Blackfoot River watershed in Southeastern Idaho identified a maximum selenium concentration of 6.8 ug/L in background surface water samples and up to 400 ug/L immediately downstream of seeps from phosphate mining overburden shales. Laboratory investigations on indigenous cutthroat trout failed to identify adverse effects on mortality or reproduction at up to 15 ug/g (dry wt) selenium in the diet and up to 11 ug/g (dry wt) selenium in whole body samples. These observations suggest the proposed hazard assessment protocol for selenium may be applicable as a screening tool, but should not be used for the determination of ‘safe’ watershed concentrations of selenium.

P3.29  Risk Estimation of Short Chained Chlorinated Paraffins in the Future. K Tsunemi, A Tokai, J Nakanishi; Research Center for Chemical Risk Management, National Institute of Advanced Industrial Science and Technology   k-tsunemi@aist.go.jp Abstract: The use of short chain length chlorinated paraffins (SCCPs) has been found to present a risk to aquatic organism in surface water. In addition, SCCPs have been shown to concentrate a large extent into sediment and may degrade very slowly in the environment, therefore there is concern that a risk to benthic organism increases in the future. Thus, in this study, the transport processes of SCCPs from their major sources to surface water and sediment were modeled to estimate the time course, from the past to the future, of the risk to aquatic and benthic organism. The margin of exposure (MOE) for the ecological risk was calculated based on the estimated concentration in surface water and sediment. Three scenarios were set as countermeasures for risk reduction. The first scenario was BAU (business as usual) scenario with no countermeasure for risk reduction, and the second scenario was the registration of PRTR list of chemicals aimed at reducing emission of SCCPs in production and use sectors of SCCPs. The third scenario was the implementation of marketing and use restrictions to limit the use of SCCPs. The result of the first scenario indicated that the MOE values were not be sufficiently large to guarantee the safety of aquatic and benthic organism in the future, whereas in the result of the second and third scenarios, the MOE values were sufficiently large to guarantee safety. Therefore, the second scenario may be the appropriate countermeasure for its reasonable cost performance.

P3.30  How few PCB congeners can estimate total PCBs in aquatic biota?. J.A.* Nedoff, L.J. Kennedy, B.A. Williams; Kennedy/Jenks Consultants, San Francisco, CA, and Portland, OR   judynedoff@kennedyjenks.com Abstract: There are 209 individual PCB congeners, not all of which were present in the technical Aroclor mixtures used for various industrial purposes. PCB production was banned in the US in 1975, but PCBs are persistent in the environment, as many congeners do not readily degrade. As a result of PCBs are often among the primary risk drivers at contaminated sediment sites. Shorter lists of representative congeners have been developed to streamline laboratory analysis and data evaluation, including a list developed by NOAA for the National Status and Trends Program. A short list of congeners that consistently represents a proportion of the total could be used, which would result in reduced costs and comparable datasets at sediment sites where PCBs are the focus. The total from the short list could be multiplied by a defined factor to estimate the total PCB concentration. To develop a proposed "short list," datasets from a variety of sites that include an extensive list of congeners in tissue samples were analyzed to determine the most abundant congeners among various tissue types. This analysis resulted in a list of congeners that represents a consistent proportion of the total PCB value. This list was compared to other short lists, including the NOAA list, the list of dioxin-like congeners, and other lists of congeners analyzed at sediment sites. The summed concentration of the 12 dioxin-like congeners and the associated TEQ also are compared to the estimated total PCB concentration for the different tissue types.

P3.31  Risk Assessment of Introducing Rabies into Taiwan Through Importation of Dogs. H Weng; University of California Davis   hweng@ucdavis.edu Abstract: No cases of rabies have been reported in Taiwan since 1959. However, with the exception of Japan, all other neighboring countries of Taiwan have been declared rabies-endemic areas, or have been considered to be high-risk areas likely to carry the disease. Given the extensive exchanges between Taiwan and the countries in this region, and its high population density of stray dogs, an outbreak of rabies in Taiwan could have serious consequences. The objectives of this study were to model and assess the risk of introducing rabies into Taiwan through importation of dogs under different quarantine policies and to select the optimal safety measure for prevention of introducing rabies into Taiwan. An innovative stochastic model was developed to assess the risk of a rabid dog penetrating different safety measures. The probabilities for a rabid dog to successfully penetrate different safety measures were computed from the results of 10,000 Monte Carlo iterations using @Risk 4.5. The parameters of interest were the expected number of years for introduction of a rabid dog into Taiwan to occur and the relative risks of introducing rabies using different quarantine policies compared with the current ones. The results of our model showed that except through smuggling, the current policy, a thirty-day quarantine, had the highest risk of introducing rabies into Taiwan. Requirement of a verified rabies vaccination certificate alone was the most efficient safety measure to prevent introduction of rabies into Taiwan, while control of smuggling was also a significant prevention measure. One additional benefit of requirement of a vaccination certificate was the motivation of no quarantine at all would dramatically reduce the current very real threat resulting from dogs being smuggled into Taiwan. We concluded that requirement of a verified vaccination certificate as an alternative to existing quarantine policies was favorable for prevention of introducing rabies into Taiwan.

P3.33  Importance analysis and value of information analysis in air pollution risk modelling. M.* Tainio, J.T. Tuomisto; National Public Health Institute, Department of environmental Health, Finland   Marko.Tainio@ktl.fi Abstract: In this study, value of information (VOI) analysis and importance analysis were used to assess quantitatively uncertainties in a risk model. The scope was to compare probabilistically health effects of alternative bus types in Helsinki, Finland. Model included emission, exposure, dose-response and mortality sub-models. Parameter uncertainties and model uncertainties were defined quantitatively. The model was implemented using AnalyticaTM Monte Carlo simulation program. Importance analysis was used to identify which sources of uncertainty contribute most to the uncertainty in the model results. By doing so we could concentrate the modelling efforts to those parts of the model that affect the outputs most. We calculated rank-order correlations between the input variables and the model outputs. Value of information analysis represents a technique that evaluates the expected benefits for decision-making if additional information becomes available and reduces uncertainty. Importance analysis showed that the plausibility of the cardiopulmonary mortality and the emissions factors of alternative bus types had the larges impact to the model uncertainty. Value of information analysis showed that perfect information for this particular choice of bus types was expected to improve the decision outcome by 0.07 deaths per year. VOI analysis results imply that the remaining uncertainties are not important for decision-making. Both VOI and importance analyses were shown to bring significant information to the risk modelling and decision makers.

P3.35  Acrylamide and Furan in Heated Foods: A Toxicological Review and Risk-Benefit Evaluation of Heat-induced Reactions . J.R. Coughlin; Coughlin & Associates   jrcoughlin@cox.net Abstract: In 2002 Swedish researchers found acrylamide (a rodent carcinogen) in fried/baked foods (notably potato chips, French fries) from 30 – 2,300 ppb and reported that foods cooked or processed below ~ 180 ºC did not contain detectable levels. Acrylamide production was attributed to the high temperatures reached in Maillard, non-enzymatic browning (sugar-amine) reactions required for foods’ desirable color, flavor, aroma and texture. Acrylamide research is underway globally on: (1) mechanisms of formation and mitigation; (2) methods of analysis; (3) dietary exposure and biomarkers; and (4) toxicology and metabolism. In May 2004 the U.S. Food and Drug Administration (FDA) released preliminary data on the presence of furan (a rodent carcinogen) in a range of heated foods up to ~ 160 ppb, and FDA initiated a major review of the safety of furan. An important toxicological consideration that has received little attention to date is that many Maillard Reaction Products (MRPs) have been shown to be beneficial to health, including many antioxidants, anti-mutagens and anti-carcinogens. MRPs have been identified that induce the formation of carcinogen-detoxifying enzymes. In addition, furan itself has been shown to be a good antioxidant, as have many other volatile heterocyclic flavor compounds and the melanoidins, the complex brown Maillard polymers. Thus, dietary acrylamide and furan risk to humans must be evaluated in light of their metabolic disposition in the human body together with the beneficial effects (on cancer, coronary artery disease, immune function and aging) known to be produced by the Maillard reaction. The results of the global research efforts on the effects of heat-processing and cooking of foods must be used to carefully evaluate both the risks and benefits of heat-induced compound formation in support of any regulatory decision-making on the public health risks of dietary acrylamide and furan.

P3.37  Is There Really an Increased Crash Risk Associated with Cellular Telephone Use While Driving? . R Ray, T Fordyce, K Ramachandran; Exponent Inc.   kramachandran@exponent.com Abstract: The most recent studies examining cell phone usage rates in national accident databases date back to the late 90’s. These studies were based on analyses of the Fatality Analysis Reporting System (FARS) and the National Accident Sampling System (NASS). The number of wireless subscribers in the US has almost doubled since this research was completed, and while this does not appear to be associated with an accompanying increase in either fatal crashes or collisions, the availability and increased use of cell phones while driving has raised safety concerns from both the government and the general public. It is likely that prior studies were susceptible to an under-reporting bias as only a few states coded information on cell phone use. As of 2002, 16 states coded this information; however, we are unaware of any published studies that have re-examined FARS and NASS. The purpose of this study was to conduct examine these databases to determine if the data indicate there is an increased risk associated with cell phone usage while driving. Most crashes do not involve distractions, however, given a crash where distraction is a factor, we found that there is not an increased risk associated with cell phone use when compared to other distractions. Our analysis revealed that cell phone use while driving is no more dangerous than adjusting the radio or climate controls. As wireless subscribers nearly doubled in the time period analyzed, we expected to see an increase each year in crash rates where cell phone use was a factor. Rates were calculated by normalizing the crash frequencies using subscription information. We found that there was not an increase in crashes associated with using cell phones while driving, as the rates remained stable. All of the data currently available indicates that there does not appear to be an increased crash risk associated with cell telephone use while driving.

P3.38  Risk and Decision Analysis for the ESER Program, Idaho National Engineering and Environmental Laboratory (INEEL), Idaho. B.N. Narloch, W.E Wright; MWH   bruce.narloch@mwhglobal.com Abstract: Under sub-contract to Stoller Corporation, MWH was tasked to evaluate the adequacy of a $1 million per year environmental monitoring program at the Idaho National Engineering and Environmental Laboratory (INEEL). This half century-old program monitors potential releases of radiologicals, metals and other chemicals to the environment. MWH conducted a human health and ecological risk assessment to identify areas of the study area where potential environmental impacts could potentially occur. Primary contaminants of concern included radionuclides (e.g., cesium-134 and –137, cobalt-60, plutonium-239 and –240, radium-226, strontium-90, and uranium-234 and –238), metals such as hexavalent chromium and lead, and chlorinated solvents. The human health risk assessment considered direct contact pathways (e.g., oral, dermal and inhalation routes of exposure), as well as indirect exposure through the food-chain. Results of the human health and ecological risk assessment were used as inputs to a decision model that was designed to optimize resource allocation for future monitoring. Inputs to the model included human health and ecological risk, regulatory requirements, public perceptions, stakeholder values, and monitoring costs. Results of the risk assessment and decision analysis were used to rank monitoring media and sampling locations in terms of their relative importance. These results are currently being used for strategic environmental planning and future resource allocation.