SRA / Society for Risk Analysis / SRA 2009 Annual Meeting

Poster Session

Room: 6:00-8:00 pm

P.1 Identifiability of Bioaerosol Transport and Risk Models by Environmental Sampling. Hong T*, Gurian P; Drexel University th339@drexel.edu

Abstract: In the aftermath of a release of a microbiological agent sampling must be conducted to identify the amount of agent released. Mathematical models can then be used to model the environmental transport and human health risks due to the exposure. Because both the dose response and transport of aerosolized microbiological agents are affected by the effective aerodynamic diameter of the particles, environmental sampling should identify not the total amount of agent but the size distribution of the particles. In this study a completely mixed compartment model is evaluated for the identifiability of parameters describing the amounts of different size fractions of particles initially released. Different sampling schemes are considered and condition numbers are calculated to assess the identifiability of the initial release. The results indicate that if samples are taken in a single well mixed room then neither a 4-particle-size model nor a 3-particle-size model can be identified by aggregate sampling of available environmental compartments. If sampling is extended to a second room, which has been contaminated by air flow through the HVAC system, then it is still not possible to identify a 4-particle-size model. However, it is possible to identify a 3-particle-size model. The three particle sizes modeled are 1 μm particulates, which would account for much of the risk outside the building due to their potential for long range transport, 3-5 μm particles which are respirable but are removed more readily by deposition and HVAC filters than 1 μm particles, and 10 μm particles which are not readily respirable and settle relatively quickly. Modeling these three categories is proposed as a means to distinguish between the hazards posed by different sizes of aerosols while reducing model complexity to a point where necessary inputs can be identified by aggregate sampling methods.

P.2 Microbial risk assessment of exposure to biosolids-associated pathogens. Teng J*, Gurian PL, Olson MS, Kumar A, Zhang H, Harte C, Olson B, Downs K; Drexel University tj62@drexel.edu

Abstract: Current U. S. Environmental Protection Agency regulations for the land application of biosolids are not based on quantitative microbial risk assessment. This study is developing quantitative models to estimate microbial risks to human health from exposure to land-applied biosolids. The first step was to precede the typical risk assessment framework with a consideration of the decisions to be informed. In this case setback distances from Class B land application sites are a key decision variable. To inform setbacks to groundwater and surface water a consideration of wet weather events is crucial. This initial step guided the nature of the subsequent risk assessment, as modeling of precipitation and infiltration needed to be developed to inform estimates of the appropriate setback distances. The subsequent work followed the established risk assessment framework. The hazard identification consisted of developing a list of possible pathogens of concern from biosolids, consisting of bacteria, virus, and parasites. Four exposure routes were considered: (1) Inhalation of aerosols, (2) Consumption of surface water, (3) Consumption of ground water, and (4) Direct ingestion. The pathogen doses were estimated from transport and fate models for each of the four exposure pathways. These doses were used to calculate daily risk of infection and risk of infection per application period. Different exposure rates and durations were assumed for the three subpopulation types: (1) Residential-adult, (2) Residential-child, and (3) Occupational-worker. Once coupled with site-specific data, this model can provide a technical basis to interpret the risk of biosolids-associated pathogens in terms of health impact. It can be used by environmental managers to set criteria for allowable pathogen concentrations in biosolids and setback distance, based on corresponding acceptable levels of risk.

P.3 Acceptable Microbial Risk: Benefit-Cost Analysis of a Boil Water Order for Cryptosporidium. . Ryan MO*, Duzinski P, Gurian PL, Haas CN, Rose JB; Drexel University mor23@drexel.edu

Abstract: Cryptosproidium is a naturally occurring pathogen and is of grave concern in water treatment due to its extreme resistance to disinfection by chlorination. It is typical for plants to issue a boil-water-notice (BWO) to a potentially affected populace when Crypto is detected in the finished product. Boil water orders involve costs as well as potential health risks and may not be justified for very low risks of infection; however, there is no predetermined risk level or environmental concentration that triggers such an order. In this study a cost-benefit analysis, conducted in advance of the traditional risk assessment framework, is used to identify a threshold level of risk for issuing a BWO, taking into account cost/values associated with: a statistical life, the expected cost of treatment of the disease and boiling water for the duration of the BWO. A decision tree populated with costs and probabilities of infection and illness is constructed and an exponential dose response model is used to solve for the dose of Crypto that can cause disease and possibly death of the exposed individual. Using this exposure dose, an environmental or raw water concentration is determined using the expected contact rate from drinking water given a typical water treatment plant and no point-of-use treatment. Monte Carlo techniques are used to conduct sensitivity/uncertainty analysis on several parameters used in the threshold risk, dose response and environmental concentration models.

P.4 Valuing Environmental Detection of a B. anthracis Release. Madsen JM, Gurian PL*; Drexel University pgurian@drexel.edu

Abstract: This study evaluated the benefits of early detection of a B. anthracis release that might be provided by an environmental detection system. A decision analytic model was constructed to estimate time to detection and treatment with and without the environmental monitoring program. The model considers a range of different exposure scenarios and assumptions as to the time at which an attack would be detected without environmental monitoring. Health benefits are monetized using a quality adjusted life years approach. Calculated benefits range from roughly $1 per person per hundred risk of attack for low dose releases with optimistic assumptions about the effectiveness of medical surveillance to $1000 per person per hundred risk of attack for high dose scenarios with conservative assumptions about the effectiveness of medical surveillance. These values do not take into consideration a system’s capacity for detection of other biological or chemical agents. The model results are influenced by uncertain factors, such as probability of attack and probable exposure levels and are intended only to serve as an initial effort at developing a framework for deciding when environmental detection efforts are appropriate.

P.5 Development and Evaluation of a Mechanistic Dose Response Model for Inhalation of Bacillus anthracis Spores. Weir M.H.*, Haas C.N.; Department of Civil Architectural and Environmental Engineering, Drexel University mw88@drexel.edu

Abstract: In dose response modeling currently does not include any depiction of the internal host responses from exposure to the pathogen. One of the main problems introduced into the current dose response paradigm is the dose itself. The dose used currently in dose response modeling is the exposed dose, meaning the dose that the host inhales is considered that which causes disease. This work models the bulk fluid transport and deposition in the respiratory system, and main pathogenesis of Bacillus anthracis (B. anthracis). This research is concerned with adapting the exposed dose to take into account the in-vivo host responses to pathogen exposure. A two-stage model has been developed and evaluated, the first stage being a stochastic model to track the bulk fluid transport and spore deposition throughout the respiratory system. The second stage is deterministic which models the interaction between B. anthracis spores and germinated vegetative cells with alveolar macrophages. The affect of this two-stage model in the dose response relationship of a set of animal models will be demonstrated. It is intended that the framework demonstrated will assist in the development of a set of physiologically based pathogen transport and kinetics models (PBPTK). This set of PBPTK models will be analogous to the physiologically based pharmacokinetic (PBPK) models, which have been incorporated into the chemical risk assessment toolbox over twenty years ago

P.8 Concepts and characteristics of causation, a clarification of Hill’s considerations. Cormier SM*, Suter II GW; U. S. Environmental Protection Agency cormier.susan@epa.gov

Abstract: It is more than 40 years since Sir Bradford Hill’s address to the Royal Society of Medicine on The Environment and Disease: Association and Causation. Hill’s intent and valuable contribution was to develop some “viewpoints” to consider when reviewing causal evidence. However, we believe that it is now time to explain and differentiate concepts that underlie the assessment of causation. In particular, we make a distinction between the basic characteristics of causation and the evidence that affirms or discounts specific possible instances of causation. We also explain how evidence of causation is drawn from different sources of information, and show how these sources are distinct from the qualities of the evidence. This differs from Hill’s causal considerations which implicitly combine characteristics of causation, types of evidence, sources of information, and qualities of evidence. We argue that there are six fundamental and useful characteristics of causation: time order, co-occurrence, preceding causation, sufficiency, interaction, and alteration. The cause precedes the effect (time order). The cause co-occurs with the unaffected entity in space and time (co-occurrence). Causes and their effects are the result of a web of causation (preceding causation). The intensity, frequency, and duration of the cause are adequate and the exposed entity is susceptible (sufficiency). The cause effectively interacts with the entity in a way that induces the effect (interaction). Finally, the entity is changed by the interactions with the cause (alteration). Evidence of these six causal characteristics can form the basis for assessment of causal relationships. The list of characteristics provides a template for developing an explanatory narrative that is causally relevant and substantive. Using six core characteristics may be easier for some assessors to organize, evaluate, and communicate their case, and for decision makers to assimilate, remember, and inspire action.

P.10 PROGRESS ON CUMULATIVE RISK ASSESSMENT GUIDANCE AT EPA. Bangs GW, Bollweg G, Galizia A*, Lowit A, Maurice C, Serveiss V, Victery WW, Galizia A, Hubal E, Payne-Sturges D; U.S. Environmental Protection Agency galizia.audrey@epa.gov

Abstract: In 1997, the EPA Science Policy Council issued guidance on planning and scoping for cumulative risk assessments. The Science Policy Council called upon the Agency’s Risk Assessment Forum to develop Agency-wide cumulative risk assessment guidance that builds from the Agency’s experience. This is being done in three phases. The first phase culminated in the Framework for Cumulative Risk Assessment in 2003. Building on the Agency’s growing experience, the Framework identified the basic elements of the cumulative risk assessment process. It also identified questions that needed to be answered before, or during, the development of specific guidance. Therefore, the technical panel convened to address the specific issues raised in the Framework; to identify or develop case studies exemplifying elements of cumulative risk assessment; and to identify remaining research needs for the Agency, in support of guidance development. This second phase of the technical panel’s effort resulted in several products: five issues papers, originally published in 2007; 12 case studies illustrating aspects of the Framework implementation; a Report that follows the Framework section by section and draws examples and key issues from the issue papers, case studies, and other literature, and also discusses research to improve cumulative risk analysis. The next Phase, developing guidance for cumulative risk assessment, will draw on this report, as well as a number of other efforts inside and outside EPA: a Regional Workshop being held in 2009; developing chemical mixtures methods; community risk assessment tools; European efforts such as NOMIRACLE and REACH; and geospatial analysis.

P.11 Nuclear threat risk assessment for Azerbaijan. Bayramov AA*; Institute of Physics Azerbaijan National Academy of Sciences azad.bayramov@yahoo.com

Abstract: The IAEA recommends six general nuclear security objectives: 1) performance of a risk based threat assessment, 2) establishment of appropriate legal and regulatory measures that promote nuclear security, 3) prevention, 4) detection, 5) response, and 6) the development of human resources. The mean of a nuclear security threat assessment is to better understand the existence and nature of the possible dangers presented by insiders, internal groups, and external groups that may seek unauthorized access to nuclear materials from sources inside the nation or that may try to use the nation’s transportation systems or geography as part of a illicit transport system to move contraband nuclear materials from one point to another. The aim of our investigations is not to predict terroristic events in Azerbaijan but only to indicate which scenario has the higher utility from the point of view of a terroristic organization in terms of balance between factors favoring and discoursing the attack, respectively.

P.12 Statistical aspects of risk assessment of chemicals, using graphical modeling. Fujii T*, Kageyama M, Gamou M, Kanafuji K, Tsubaki H; The institute of statistical mathematics t-fujii@ism.ac.jp

Abstract: Our aim is to construct a statistical model and the inference algorithm that provide a theoretical proof for the risk assessment of chemicals. In this talk, we introduce a graphical modeling that is suitable for the representation of the causal relations with uncertainty. It becomes better combining existing statistical tools such as EM-algorism, latent variables and so on. Our approach is challenging, but substantial progress can be made.

P.13 Derivation of QALY based dose-response relationship from animal data for the purpose of risk trade-off analysis of chemical substances. Gamo M*, Fujii T, Kageyama M, Hojo R, Gamo Y, Kishimoto A, Kanefuji K, Tsubaki H; National Institute of Advanced Industrial Science and Technology (AIST) , The Institute of Statistical Mathematics masashi-gamo@aist.go.jp

Abstract: When risk due to a chemical substance is reduced by substituting the substance with another, we need to consider risk trade-off problem. Although quantitative evaluation of risk trade-off is desired, conventional risk assessment methodologies, which utilize indices such as hazard quotient or MOE (margin of exposure) for health effects with threshold and lifetime probability of cancer for cancer risk, are not suitable for quantitative evaluation of risk trade-off. QALY (Quality Adjusted Life Years) is regarded as one of the risk metrics that enable us to compare various types of health risk quantitatively. However, derivation of QALY requires human dose-response data from epidemiological studies. This requirement seriously limits the applicability of QALY to risk trade-off analysis. We have developed a prototype framework of derivation of QALY based dose-response relationship from animal data. First, with consistency with the concept of cumulative risk assessment in mind, the endpoints found in repeated dose toxicity database were categorized by (sub-) organ and type of effect. The relationship of LEL (lowest effect level) among categorized endpoints were analyzed to develop an inference algorithm based on graphical modeling. The graphical modeling enables us to infer the relative toxicity value with confidential interval between arbitrary two substances for each organ. Second, for each organ, we explored for a reference substance for which human epidemiological data is available, and derived a QALY based dose-response relationship for the reference substance. Finally, the QALY based dose-response relationship for a substance under assessment is obtained by combing the relative toxicity value and the reference QALY based dose-response relationship. This framework was applied to the case studies of industrial detergents and additives of plastic. This study is funded by NEDO.

P.14 DECERNS: New Web-based Software Tool for Multicriteria Decision Analysis. Gritsyuk S, Tkachuk A, Babutski A, Vasilevskaya M, Mirzeabasov O, Didenko V, Yatsalo B*, Sullivan T; IATE yatsalo@obinsk.ru

Abstract: Decision-making support on risk management should take into account not only risk and cost criteria, but also different technological, ecological, geological and social objectives along with the stakeholders’ interests. DECERNS (Decision Evaluation in Complex Risk Network Systems) is a Web-based Spatial Decision Support System, WebSDSS, for multi-criteria decision analysis of a wide range of spatial and non-spatial alternatives. It integrates basic and some advanced GIS functions and implements several Multi-Criteria Decision Analysis (MCDA) methods and tools. DECERNS can also be integrated with a model server containing generic and site specific models for in-depth analysis of environmental risks as well as other decision criteria under consideration. DECERNS is an MCDA-oriented SDSS with implementation of MAVT, MAUT, AHP, TOPSIS, PROMETHEE and some other multicriteria models. Special tools allow extensive sensitivity analysis both for weight coefficients and partial value functions, and uncertainty analysis with the use of probabilistic approaches and fuzzy numbers (Probabilistic Multicriteria Acceptability Analysis, ProMAA, Fuzzy-MAVT, and Fuzzy-MAA). The structure of DECERNS WebSDSS, features of integrating different MCDA methods, GIS-technologies and math models, and implementation within the case studies will be presented. This work is carried out within the DECERNS project (IPP/ISTC #3549, www.decerns.com ).

P.16 Searching and Pruning Risk Factors in the Logit Model. Kawasaki Y*; The Institute of Statistical Mathematics kawasaki@ism.ac.jp

Abstract: The logit model is one of the most common tools in risk analysis. This paper proposes a method to search higher order interactions in the right hand side of the logit model. Taking many combinations of risk factors into account often leads to combinatorial complexity. We also propose some pruning methods to narrow the list down to a tractable size. Suppose we have a set of categorical data, some of which may not be exactly categorical but are continuously distributed. We want to discriminate each record either to positive or to negative using other observations as explanatory variables. In such a case, it is natural that we want to expand the usual logit model so that it should include two-factor interactions or higher as well as original data. The problem is how we find promising interactions effectively. To do so, we begin by searching all the three-way contingency tables. This can be done by multinomial model fitting with model selection by AIC. According to the size and dimension of the data set, this approach inevitably causes a sort of combinatorial explosion. As a pruning method to reduce the number of candidate interactions, we propose a test whether the distribution of the response variable given a set of the conditioning variables is significantly different from the marginal distribution given the whole data. Only survived interactions are used in the logit model. Regularization methods (L2, L1, elastic net) are also considered firstly to mitigate the multicolinearity, secondly for an efficient variable selection. An application in credit scoring will be shown as a real world example.

P.17 Analysis of low dose synergy literature for use in screening chemical co-exposures for risk assessment . Mumtaz M*, HESI Risk Assessment Methology Committee Mixtures Project , Embry MR; ATSDR, 1600 CLIFTON ROAD, F-32, ATLANTA, GA; ILSI HEALTH AND ENVIRONMENTAL SCIENCES INSTITUTE, 1156 15TH STREET, NW, WASHINGTON DC 20005 membry@ilsi.org

Abstract: There is a critical need to better understand the consequences of co-exposures to substances at environmentally relevant concentrations. Default assumptions of dose or response additivity, as appropriate, have been used to assess the joint toxicity of chemicals, due to the fact that it is not possible to perform toxicity tests on all combinations. In order to develop a screening approach to prioritize chemical mixtures or co-exposures for more detailed assessment, it is essential to know whether synergistic interactions can occur at low, environmentally relevant exposure levels. Studies present in the literature demonstrating synergism in mammalian test systems were identified, with emphasis on studies performed at doses close to the individual chemicals’ NOAELs or LOAELs in the test systems. This literature search identified data for 204 chemicals by in-depth critical review of 90 unique references. Few studies, however, included quantitative estimates of low dose synergy; interaction magnitude calculations were included in only 11 articles. Quantitative methodology varied across studies in terms of the null hypothesis, response measured, point of departure (POD), and whether the slope of the dose-response curve was considered. The values reported are not directly comparable because of the various definitions and quantification of synergy used. In general, the magnitude of any low dose synergy reported was relatively modest. It is anticipated that this analysis will be considered in developing an approach for quantitative risk assessment of chemical mixtures and co-exposures, including use of a Threshold of Toxicological Concern for screening and prioritization. Prior to this, it is recommended that consistent approaches be applied in synergy quantification, including that: synergy be defined in terms of departure from dose additivity; uniform procedures be developed for assessing synergy at low exposures; and the POD for calculating synergy be standardized.

P.18 CORRELATION BETWEEN PREDICTION ACCURACY AND NUMBER OF EXPERTS: AN EMPIRICAL STUDY . Shirazi C. H., Mosleh A.*; Center for Risk and Reliability, University of Maryland, College Park Mosleh@umd.edu

Abstract: Because expert information is considered uncertain, it seems logical to consult multiple experts in an attempt to have a more dependable database or at least gather more evidence. Speculations made about the correlation between accuracy of experts’ estimates and the number of experts elicited, have led many to conclude that the more experts are elicited, the higher accuracy of estimates can be reached. This may seem similar to increasing the sample size in an experiment. However, questions still remain whether empirical data actually support this idea, and if it does, to what extend this association has a notable impact in a real world application. In other words, is there minimal number of experts needed to obtain an optimal accuracy of estimates? The primary focus of this study is to explore possible relationship between the number of experts and overall accuracy of estimates using empirical data. To conduct this study, over 1500 publications analyzing or reporting the use of expert opinion since 1930s are examined, accumulating over 1900 experts’ point estimates in 27 different disciplines. In addition, another expert judgment database (TU Delft ) is used, which reports the assessment of over 800 experts on more than 4000 variables, representing about 80,000 elicited questions. The collected expert estimates are updated in a Bayesian aggregation framework and compared with true values. The average error of estimates is monitored as the number of experts increases, to uncover whether the increase in the number of opinions reduces the overall error of estimates and unveil the number of experts needed to obtain an optimal accuracy in practical applications.

P.22 Cost-effectiveness of investments in defense of critical infrastructure. Jamshidi T*, Bier V M; University of Wisconsin - Madison jamshidi@wisc.edu

Abstract: One of the most important parameters in determining the optimal allocation of security resources is the cost-effectiveness of security investments. However, the cost-effectiveness of investments in protection and resilience has not been rigorously studied, and as a result, we may be uncertain about this important parameter to within several orders of magnitude. In this talk, we plan to present first-of-a-kind quantitative estimates of the cost-effectiveness of our investments in protection and resilience, based on observed reductions in estimated risk or infrastructure vulnerability after the expenditure of security funds. We believe that the development of estimates of cost-effectiveness can support improved decision making in the various agencies that distribute security funds.

P.23 Deterring and Detecting the Smuggling of Nuclear Weapons in Containers Freight. Bier VM, Haphuriwat N*, Willis H; University of Wisconsin-Madison, RAND Corporation naraphoh@cae.wisc.edu

Abstract: Concerns about terrorists smuggling nuclear bombs into the U.S. in container freight have led to demands for 100% inspection at U.S. ports. However, under some circumstances, it is possible to deter smuggling attempts with less than 100% inspection. Based on available data, we quantify a model of inspection and deterrence to find the optimal level of inspection in the face of attempted nuclear smuggling, and explore the sensitivity of the model results to plausible changes in parameter values.

P.24 Port security (PortSec) risk analysis and resource allocation I: Methodology . Barrett AM*, Orosz MD, Southwell C, Bakir NO, Maya I, Chen J; University of Southern California anthony.barrett@usc.edu

Abstract: We develop and apply systems-based risk management methodologies and tools to inform port security resource allocations and investments. For tactical use, PortSec will provide up-to-date risk information for areas of interest and the overall port complex. For strategic use, PortSec will provide the port security analyst with tools to evaluate costs and benefits of counter-measures.

P.27 Setting a Standard of Proof: A Way to Introduce Risk Analysis Concepts. Brand KP*; University of Ottawa kbrand@uottawa.ca

Abstract: The author presents an `allegory' that has been found to be particularly effective (in his experience) as a `lesson plan' for introducing diverse audiences to key concepts in Risk Analysis. The allegory focuses on the rather general challenge of setting a “standard of proof.” It uses familiar examples drawn from law and medicine to interrogate how a “standard of proof” is set. This general problem context is used to draw sessons that can be applied to the domain of health risk assessment. The proposed lesson-plan, is particularly well suited for illustrating the important “is versus ought” distinction and the implied contingency upon both facts and (normative) values. A synopsis of the allegory is presented, with the balance of time being spent highlighting its ability to illustrate the key concepts in risk analysis, and enlighten contemporary debates.

P.28 Approximate infective viral load of non-pasteurized liquid egg from a highly pathogenic avian influenza-infected, undetected flock. Weaver JT*, Clouse TL, Malladi S, Ebel ED, Schlosser WD, Golden NJ; United States Department of Agriculture, Animal Plant Health Inspection Service, Veterinary Services todd.weaver@aphis.usda.gov

Abstract: In the event of a highly pathogenic avian influenza (HPAI) outbreak in the U.S., emergency response measures such as quarantine and movement control would be implemented towards limiting further disease spread. Requests for movement permits for birds or their products from premises identified in a quarantine order must be supported by a risk assessment demonstrating acceptable risk through their movement. Evaluating these risks before an outbreak can promote business continuity and recovery by reducing the time taken to conduct a risk assessment during the outbreak. Non-pasteurized liquid egg (NPLE) is a poultry product for which proactively evaluating the risk before the outbreak is advisable given its low shelf life and minimal industry product holding capacity. The objective was to evaluate the risk that movement of NPLE from an HPAI- infected, undetected flock would result in disease spread to another poultry premises. We used a stochastic disease transmission model to estimate the HPAI disease prevalence, disease mortality and fraction of internally contaminated eggs at various time points post infection of the flock. Using the model, we evaluated a targeted surveillance protocol where a pooled sample of 5 randomly selected birds from the daily mortality is tested via RRT-PCR to estimate the number of days to detect infection. Finally, we estimated the viral concentration in a transport tanker of NPLE while considering the mixing of the internal contents of contaminated and virus free eggs. Simulation results of the mean viral titer in the tanker of NPLE ranged from 1.15-4.3 EID50/ml under alternate scenarios. Analysis with an exponential dose response model indicates that the probability of the viral concentration in the NPLE being infective to an exposed chicken is less than 1%, suggesting that spread of HPAI among flocks via this transmission pathway is low. Use of this model can provide decision-makers with a framework for evaluating preventive measures.

P.29 Arsenic and tobacco-use related disease risk. Marano KM*, Wilson CW, Kathman SJ, Naufal ZS, Garner CD; RJ Reynolds Tobacco Company maranok@rjrt.com

Abstract: Arsenic is found in tobacco and mainstream cigarette smoke (MSS). Whether arsenic has an independent role in tobacco use-related disease, however, is not known. In this evaluation, the association between tobacco use-related disease and exposure to arsenic at concentrations relevant to cigarette smokers and smokeless tobacco (SLT) users (snuff and loose leaf) was assessed through a review of epidemiology and biomonitoring data and use of probabilistic risk assessment (PRA). Available epidemiological evidence suggests that oral and inhalation exposures to arsenic are associated with cardiovascular effects (e.g., stroke) in humans, effects which may also be associated with cigarette smoking and SLT use. Additionally, occupational studies show that high arsenic exposures are associated with lung cancer, an adverse health outcome also linked with cigarette smoking. For both SLT users and cigarette smokers, PRA of arsenic in SLT and MSS show that non-cancer risk estimates are below regulatory thresholds (i.e., Hazard Quotient <1) and cancer risk estimates are within the commonly accepted range of risk (i.e., 10-4 to 10-6). In environmental and occupational epidemiology studies, concentrations of arsenic associated with increased disease risk are higher than those observed in cigarette smokers and SLT users. Biomonitoring values from a representative sample of the US population (NHANES) suggest that total urinary arsenic in cigarette smokers and SLT users was not different than in non-users of tobacco. In addition, total urinary arsenic was not correlated with serum cotinine, a major metabolite of nicotine. Urinary arsenic does not appear to be elevated in cigarette smokers and SLT users compared with non-users of tobacco, suggesting that cigarette smoking and SLT use may not contribute significantly to urinary arsenic. Taken together, these results suggest that arsenic may not be independently associated with increased risk of disease in cigarette smokers and SLT users.

P.30 Advances and needs associated with the Technological Risk Prevention Plan in the French regulation. Alvarez A*; Worcester Polytechnic Institute al_alvarez@wpi.edu

Abstract: The 2001 Toulouse catastrophe was the dramatic catalyst that introduced, for French industries, the probabilistic evaluation of the likelihood of accidents in a purely deterministic regulatory framework. Before 2001, plant owners of facilities regulated by the SEVESO II Directive had to elaborate on safety reports so that the Authority Having Jurisdiction (AHJ) could establish an Emergency Response Plan based on worst case scenarios. After 2003 and the official publication of the “Risk Law”, the safety report became the center piece of another plan, Plan de Prevention des Risques Technologiques, or PPRT (in English: Technological Risk Prevention Plan). The PPRT appears as the industrial extension of the existing plans for natural risks such as flood or earthquake, eventually leading to the elaboration of a risk contour cartography for each of the 622 industrial facilities concerned. Ultimately, drastic measures will lead to land measures such as expropriation and relocation. In order to manage the PPRTs as efficiently as possible, the AHJ established specific groups including regulators, inspectors, industrial plant owners, and third party experts, for industrial categories such as ammonia refrigeration plants, chlorine plants, oil refineries or LPG transfer plants. A lot of work has been dedicated to the construction of quantified fault trees and event trees, which describe scenarios related to fire, explosion and toxic release hazards. These scenarios populate an official risk matrix, where the global risk of the considered industrial plant is assessed. This paper presents the current advances of the PPRT plans, illustrated with an LPG plant configuration, related to the scenario selection, the quantification of the reliability of safety measures, the consequence modeling and the account of the affected people. Different needs are also discussed in terms of ways to improve the decision making process.

P.32 Characterizing Risks of Oil Spills . Farber G*; U.S. Environmental Protection Agency farber.glenn@epa.gov

Abstract: EPA’s Oil Spill Prevention regulatory program dates to the early 1970’s, when risk analysis was not an integral component of development of regulations. The Agency is interested in developing a better risk management framework for the program, which affects over 600,000 facilities around the nation. In order to bring a risk management program up to speed, an initial task to is develop an understanding of the frequency of spills to land and water, the magnitude of spills, and characteristics of facilities and operations, and how these factors affect risks of spills. This session will present our approach to characterizing the risks of oil spills, the unique problems associated with this project, and some of our results.

P.33 A conceptual mapping model applied to Canadian regulations on dangerous goods transportation and storage. Abdelaziz Khadraoui AK*, Nathalie de Marcellis-Warin NM, Benoit Aubert BA; Centre interuniversitaire de recherche en analyse des organisations CIRANO khadraoui@cui.unige.ch

Abstract: Industrial activities require the production and the use of dangerous goods in chemical plants and the transportation between these installations. Dangerous goods induce risks on sites, but also during transportation. Regulations help chemical plants to minimize their risks on site and during transportation on dangerous goods. The legislative framework seeks to protect workers health, public health and the environment. In Canada, regulations on dangerous goods are emitted by several ministries (Environment, Health and Transportation) at the provincial and federal level. All companies must to conform to all these regulations. The task is not easy. A complete study of regulations on dangerous goods in Quebec (Canada) was carried out to better understand interactions between the various levels of regulation, to have a global vision of the lawful constraints associated to the logistic activities on dangerous goods and to identify inconsistencies and contradictions between each level of laws or in each regulation. In our approach, we used a conceptual mapping model describing concepts, rules, ontological roles and constraints governing the activities of dangerous goods transportation and storage. Using conceptual map permitted to identify informational overlap and zones of responsibilities, inconsistencies and contradictions in the Canadian regulations, Quebec’s provincial specificities and impacts of legislative evolution.

P.34 An alternative to the existing dominant of risk analysis & management in systems providing population safety: Human Centered Risk Approach vs Facility Centered Risk Approach. Eremenko VE*; International Nonprofit and Nonpartisan Organization “For Safety And Survival”, USA–RF vitalyanfi@yahoo.com

Abstract: Contemporary financial and economic crisis gives excellent occasion for critical estimation and revision of ineffective methods of management, including methods which base on the risk methodology. In turn the last one includes an area of our interest when risk analysis and management use for population safety. There is in view of using the risk methodology to provide protection and safety of population from accidents on potentially hazardous nuclear, chemical, biological facilities and also from industrial terrorism and natural catastrophes. Two alternative population risk methodology are considered and compared here: "Human Centered Risk Approach" (HCRA ) vs "Facility Centered Risk Approach" ( FCRA). Present structures of the systems of hierarchical / centralized regulation and management by population safety are based on the FCRA. Meanwhile the scope of such FCRA using proves to have very little effectiveness in the application to individuals. That as they are aimed at providing the safety not of a real, but a collective of hypothetical persons. However, not for the mass of hypothetical persons but for any concrete man- individuum such need in protection appears for the first several minutes after the start of majority of dangerous events . And this task is not possible to realize within a frameworks of the bulky and inertial hierarchical / centralized systems of population safety based on the FCRA methodology. This problem can be solved by means of compact systems kits for the individual usage only. The kits proposed are developed on the basis of the personal experience acquired with the protection of my own family during the Chernobyl' nuclear accident. It is the HCRA that can be a scientifically - methodological base for such kits of limited distributed management systems for every individual. These kits should be used while developing preventive protection measures as well as for ensuring safety of any concrete person and his/her family against industrial accidents, the natural calamities or terrorist acts. The distributed management systems, created on the base HCRA, makes possible to organize the protection of each individual, independent of distance to the epicenter and the time of the realization of danger. Moreover, the kits of individual protection, created on the basis of HCRA methodology, make it possible for each individual to purchase the measuring instruments in proper time, re-equip her / his dwelling and their personal transport, converting them into the reliable refuges from all natures of danger. Thus, these scientifically risk approach and distributed management systems are proposed as an alternative to the existing dominant of the FCRA and hierarchical / centralized systems of population safety.

P.35 Pilot Review Process for Submissions to EPA’s High Production Volume Chemical Challenge. Patterson J*, Franz C, Dourson M, Matthews H, Sandusky C; Toxicology Excellence for Risk Assessment (TERA), American Chemistry Council, Matthews Toxicology Consulting Company, Physician’s Committee for Responsible Medicine patterson@tera.org

Abstract: In 2008 the American Chemistry Council, the Physician’s Committee for Responsible Medicine, and the U.S. Environmental Protection Agency (EPA) participated in a peer consultation pilot to review test plans (TP) and robust study summaries (RSS) from the High Production Volume (HPV) Challenge and Extended HPV programs. The purpose of the pilot was to test whether stakeholder review might be improved by implementing a focused approach designed to emphasize stakeholder and expert involvement. During the original Challenge program, stakeholders, including EPA, posted comments on the submissions to EPA’s website and sponsors would respond by posting public comments. Resulting comments were limited and provided little opportunity for interaction between stakeholders. The pilot process provided an opportunity for stakeholders and appropriate experts to be involved in face-to-face panel meetings to review TPs and RSSs prepared by the primary producers. Submissions on 22 chemicals were reviewed and discussed by the panel and stakeholders in public meetings organized by Toxicology Excellence for Risk Assessment (TERA). Meetings facilitated the exchange of expert opinions on the technical soundness of the submissions and ensured full discussion of comments for the sponsors’ consideration. Summary reports including individual panel members’ comments were prepared, indicating where panel members agreed (and disagreed) on issues and recommendations. Recurring scientific issues included evaluation of aquatic toxicity information for poorly soluble compounds, applications of read-across data from studies using different doses and routes of exposure and extrapolation of data from studies of structural analogs. A post-pilot evaluation found that most participants thought the pilot process an effective approach to provide real-time feedback and that direct interaction between sponsors and panelists helped to resolve issues more quickly than possible under the original process.

P.37 Exposure to Nanoscale Materials Via the Oral Route. Abbott LC*, Deerfield K, Froggett S; United States Department of Agriculture lcabbott@starpower.net

Abstract: Human exposure to nanomaterials through food is of interest to the United States Department of Agriculture. Exposure scenarios involving foods that are subject to the Federal Meat Inspection, Poultry Products Inspection or Egg Products Inspection Acts (meat, poultry, eggs and catfish) are of particular interest to the Department. Although the oral route of administration has not received the same degree of attention as inhalation or dermal routes, there are several toxicological and exposure studies that address the oral exposure pathway. We review these published studies and present a strategy to relate the types of effects observed with those exposure characteristics likely to aid in prioritizing nanomaterials by their potential to result in possible risk of adulteration to meat, poultry, eggs and catfish.

P.38 Risk Communication and Challenges of Nanotechnology in Third world Countries. ADESINA S.A*; Florida International University sades001@fiu.edu

Abstract: With the rapid development of nanotechnology and its wide application in contemporary society. Recent argument amongst academia, publics and government posed much questions on how to educate the populace and acceptable communication strategies of emerging industrial revolution called nanomaterials. Their interactions with socio-ecological systems and their environmental health risk have generated a lot of debates ranging from absolute cynical to outright mysterious. The potential scientific prospects have posed new challenge in the 21st century. Despite the burgeoning literature on the emerging nanotechnological impact, the question of what is in store for humanity still beg for answer. When many nations are still navigating themselves in the alarming environmental situation, little or no attention is paid to the environmental health dimension, risk perception and public health communication amongst vulnerable groups in the global south. The paper offers special insight into communication strategies and public perception that make general acceptance of nanotechnology more complex and skeptical in Africa. Also, environmental awareness and public health safety issues concerning nanomaterials were examined by empirically soliciting information to elicit opinions on nanotechnology. Since risk perception involved socio and ecological dimensions, a unique conceptualization of risk communication strategies in developing world is therefore articulated. A multidimensional perspective from health belief model through media advocacy and perceived vulnerability of the subject is used to evaluate the potential for harmful effects. The paper has implication for policy formulations and novel understanding of multidisciplinary approach to nanotechnology. It seeks to contributing knowledgeable insights to social component of nanotechnology communication and risk governance globally.

P.39 What has Life Cycle Assessment Told Us about Nanotechnology? Scanlon KA, Lloyd SM*; Concurrent Technologies Corporation lloyds@ctc.com

Abstract: Increasingly, there are calls to apply a life cycle perspective to better understand both the potential benefits and risks associated with nanotechnology. Life cycle assessment (LCA) is a systematic, analytical process for quantifying the inputs and outputs for each life cycle stage and assessing the total environmental impact of a product. It considers the complete life cycle of a product. Raw materials are extracted and used to produce useful materials, including nanomaterials. These materials are incorporated into products, which are used and eventually retired. Throughout the life cycle, the use and transportation of materials and energy generates useful products, but also results in environmental discharges and waste generation. In the case of nanotechnology-based products, the environmental discharges and waste could include nanomaterials. Evaluating the total environmental impacts of a product requires analysis of the inputs and outputs associated with each life cycle stage, including nanomaterials. Several LCA-based studies focused on nanomaterials have been conducted to date. Most of these have focused on specific stages of the life cycle (e.g., nanomaterial production). A few have evaluated the full life cycle of commercial products. The types of impacts evaluated ranged from single impacts (e.g., greenhouse gas emissions and the related global warming potential) to a more complete set of life cycle impact assessment categories. None of the LCAs conducted thus far have evaluated the impacts and risks associated with nanomaterial releases. There simply isn’t enough information to accurately quantify nanomaterial releases, fate, or environmental impacts and risks to human health and the environment. While LCA has provided important insights on certain life cycle impacts from producing and using nanomaterials, it has not informed the discussion about the specific risks from nanomaterial releases or waste products.

P.40 Analysis of risk assessment of some nanomaterials. Bayramov AA*; Institute of Physics Azerbaijan National Academy of Sciences azad.bayramov@yahoo.com

Abstract: Nanomaterials are commonly described as having at least one dimension smaller than 100 nm. A broader definition, though, refers to those materials that are manipulated at the atomic, molecular, or macromolecular scales in order to achieve functionality that is different than that found in the bulk. Many consumer items are available that contain nanomaterials, such as electronics components, cosmetics, cigarette filters, antimicrobial and stain-resistant fabrics and sprays, sunscreens, and cleaning products. In addition, some nanomaterials are now being used or are under development for use as therapeutic or diagnostic agents. Although the potential for human exposures has not been fully evaluated and is likely to be low in many cases, the safety of nanomaterials should be characterized to ensure consumer, occupational and environmental health protection. In our work we have carried out analysis of risk assessment of some nanomaterials in the context of chemical risk assessment, consisting of four parts – hazard identification, dose-response assessment, exposure assessment, and risk characterization.

P.41 Del Plata Basin coastal monitoring program: the use of native species as novel sentinels of anthropogenic impact (a preliminary approach). Dopchiz LP, Michieli JL, Razetto G, Asaroff P, Santa AM, Demichelis SO*; J.F. Kennedy University Argentina inv.sdemichelis@kennedy.edu.ar

Abstract: Del Plata Basin is distributed throughout several countries in South America. Its area surpasses 3.000.000 km and contains a great variety of natural environments and resources. The main rivers which form the Basin are: Pilcomayo, Bermejo, Paraguay, Iguazú, Paraná, Uruguay and, finally the Río de la Plata. Several areas of the basin are highly impacted by human activities while other remains pristine. The aim of the present work is to find vegetal species widely spread in coastal areas which could be suitable as sentinels of pollution using exposure and effect biomarkers. Known as Sagitaria, Sagittaria montevidensis Cham. & Schltdl. (Alismataceae) and bulrush, Schoenoplectus californicus (C.A.Mey.) Soják are very abundant well distributed in this river basin. We sampled several locations covering 1000 km of coastal area looking for differences and similarities in proposed novel field biomarkers. The lack of knowledge about sensibility of this species to toxicants constitutes a gap and could be novel tools in risk assessment cause of their presence from deeply polluted to clean areas. We analyze cytogenetic aspects in pristine areas. Chromosomal counts were made being 2n= 22 in Sagitaria and 2n=ca. 34 for the bulrush, this data did not show differences with previous data. Mitotic Index calculated (MI) = number total of cells in division x 100 x( total number of observed cells)-1, were 6.8. Neither micronuclei (0%) nor other abnormalities were observed anaphases or telophases and both species showed 100% fertile pollen which was coherent with the absence of abnormalities. These results indicate absence of clastogenic and aneugenic effects. Other cytogenetic biomarkers are under study. This information complemented our in course studies, indicating that these species could be used as sentinel in these areas.

P.42 Modeling the Uptake of Polynuclear Aromatic Hydrocarbons from Soil into Plant Foliage, Roots and Seeds. Burris JB*; Syracuse Research Corporation burris@syrres.com

Abstract: As part of the EPA’s effort to establish ecological soil screening values (Eco-SSLs), models were developed to estimate the uptake of PAHs from soil into plants. A literature search was completed and data were extracted from published literature and studies were deemed acceptable if they presented measured analyte concentrations in soil and tissues of plants grown in that soil (paired data) and provided details concerning exposure duration, sample collection and handling and analytical methods. Rinsed and unrinsed data were segregated as were studies on plant foliage, roots and seeds. Estimation of chemical concentrations in plant parts was accomplished by chemical-specific regressions relating the concentration in soil to the concentration in plant. Regression models were constructed for ten separate PAHs, for low molecular weight (LMW) PAHs, high molecular weight (HMW) PAHs and total PAH. The models were included as part of the Eco-SSL guidance revisions issued by EPA in 2007.

P.43 Perception of ecological risks to water environments in southern Nevada. Kiriscioglu T*; University of Nevada, Las Vegas --Graduate Student tkirisci@hotmail.com

Abstract: Freshwater shortage, exacerbated by urbanization, global climate change and drought, is a severe problem in arid southwestern United States. Southern Nevada Water Authority (SNWA), in order to meet southern Nevada’s growing water needs, plans to bring interbasin water (groundwater) from rural Nevada (basins-of-origin) to Las Vegas Metropolitan Area (receiving basin). Concerned citizens, environmental groups, and scientists contend that SNWA’s plan will have adverse social, economic, fiscal, ethical, and ecological implications. This on-going, survey-based study examines southern Nevadans’ perception of ecological risks to water environments and how it relates to water resource management in the region. Consistent with the inverse relationship between perceived risk and perceived benefit, preliminary studies suggest that people in southern Nevada’s urban communities perceive ecological risks to water environments (due to drought, urban water consumption, and interbasin water transfers) lower than people in rural communities. The findings of the study should help us shape sustainable water resource policies in the region, ensuring the rights of the stakeholders and natural systems are not violated, valuable resources are not wasted, and the true costs and benefits of interbasin water transfers in the region are accurately quantified so that an informed decision can be made to benefit both the rural and urban communities of southern Nevada.

P.44 The Development of Computational Tools for Risk and Consequence Analysis. Rosas CA*, Fajardo HC, Munoz F; Universidad de los Andes ca-rosas@uniandes.edu.co

Abstract: An important branch of risk and consequence analysis is the development of computational tools. The characteristics of these tools can vary depending on the results obtained during the simulation of a major critical event. Considering that computational tools were created in order to show the magnitude of different effects caused by major industrial accidents (i.e. radiation, overpressure, among others), most of these programs do not establish the relationship between risk and consequence analysis. If they did do this then the price would be unaffordable for small and medium industries, especially in emerging countries. We have developed different tools that simulate four different major critical events, such as: • BLEVE • UVCE • Pool Fires on land • Jet fire These programs were made using Visual Basic; a program which uses basic language programming and then presents the results on Microsoft Excel, an easily accessible program for most companies and universities. This is one of the main advantages of Microsoft Excel in comparison to other programs or tools because in emerging countries such as Colombia there are not enough economic resources to invest in expensive programs. Another benefit of Microsoft Excel is that the tools are easy to understand. Even people with little or no knowledge about the topic are able to get the results and establish a direct relationship between the consequence analysis and the operating conditions of a plant. This makes it easier to make different decisions throughout the process of constructing a plant. In conclusion, the development of these computational tools has enabled us to establish a driving force within the country in order to generate an understanding of different sectors such as: • Public • Academic • Industrial Life holds the same worth in the third world (emerging and developing countries) as anywhere else, meaning that risk analysis should be available to every industry worldwide.

P.45 Influence of heliogeophysical and space factors on failures of systems of power supply . Bayda S*; Federal Center of Science and High Technologies, All Russian science research institute for civil defense and emergency, EMERCOM of Russia oktaedr@yandex.ru

Abstract: Influence of change heliogeophysical and space factors on occurrence of failures in system of power supply is investigated. We count, that the reason of failures are extremums of transient processes of change of these factors creating variable energy loadings on an environment and on objects and networks of electrical power supply. Change of solar activity (Wolf's numbers) causes change of electric parameters of an ionosphere and an atmosphere. Non-uniform change of position of the Earth’s axis causes additional mechanical pressure and deformations of lithosphere and, as consequence, telluric electric currents and pressure. Influence of change of the Moon's phases which, owing to gravitational interaction in system the Sun – Earth – Moon, makes active electric processes and the phenomena in lithosphere is similar. The technique of research of these influences is developed on the basis of statistics of the last failures. The analysis of influence of heliogeophysical and space factors was carried out on mathematical model: ΔTn = Tn – To where ΔTn – is the interval of time between extreme value of the factor of influence and time of failure; To – the time of extreme value of the factor of influence; Tn – the time of failure; n – statistic registration number of failure. The total number of the failures which have got in a consecutive daily interval of time allows to calculate a time-and-frequency spectrum of distribution of failures number from the moment of extreme change. The basic hypothesis of presence of spectral laws is display of effect of return of oscillatory activity Fermi-Pasta-Ulam (FPU) in the system "atmosphere – object of power supply system – lithosphere" being complex oscillator with nonlinear connections.

P.47 Results of the SRA Specialty Group Study. Guikema SD*, LaRocca S; Johns Hopkins University sguikema@jhu.edu

Abstract: This talk presents the results of an in-depth study of SRA Specialty Groups and how they may be better able to meet the needs of SRA members. The study consisted of three phases. First, we analyzed all papers published in the last three years of several journals, including Risk Analysis, together with all abstracts from the last three years of SRA Annual Meetings and the first two World Congress on Risk meetings in order to classify them into methodological and topological areas. Second, we surveyed current SRA members to determine why they belonged to specialty groups, whether or not they perceived the current specialty groups to be meeting their needs, and what additional activities the specialty groups should be undertaking. Finally, we synthesized the results of the literature and abstract survey, an assessment of the current specialty group coverage, and the results of the member survey to make recommendations for changes in the current specialty group structure and operation. This research was sponsored by the SRA council as a New Initiative.

P.48 Assessing the risk of foodborne illness caused by norovirus transmitted in foodservice systems . Li D*, Schaffner D; Graduate Program in Ecology & Evolution, and Department of Food Science, Rutgers University dili@eden.rutgers.edu

Abstract: Norovirus (NoV) is currently regarded as the leading cause of gastroenteritis in the United States. The Centers for Disease Control and Prevention (CDC) estimate that 23 million cases of acute gastroenteritis are caused by NoV annually and 39% of all recorded NoV outbreaks occur in restaurant settings. In a foodservice system, good hygiene practices can help to reduce the virus’ spreading rate and risk of cross contamination. The purpose of this study was to build a simulation model that mimics the complex interactions involved in NoV transmission, which may take place in a foodservice system and to use the model to study interventions that might reduce risk. Data from the peer-reviewed literature were collected and used to build the model. Due to the limited availability of published data, our model focused primarily on quantifying the effects of specific steps on NoV transmission including (1) transmission between food, hands and other possible sources of contamination, (2) virus survival on different surfaces and (3) the effect of different hand treatments on virus reduction. The model was built with the discrete-event simulation package, Arena® (Rockwell International), and used to simulate foodservice worker movement, as well as virus transfer and survival. The model shows key points in the virus transmission process that lead either to further spread or to reduction in the spread of the virus. Effective hand washing may prevent some but not all of the exposed population from getting infected with norovirus. The result has the potential to reduce the morbidity and economic loss currently associated with NoV outbreaks, and to identify key foodservice worker behaviors that affect virus transmission.

P.49 EMERGENT SKEWED DISTRIBUTIONS OF ILLNESS SEVERITY AND A GENERAL MIXTURE DOSE-RESPONSE FUNCTION. Englehardt JD*; University of Miami jenglehardt@miami.edu

Abstract: Linear, additive probabilistic systems are everywhere, leading to the ubiquity of the Gaussian distribution as explained by the Central Limit Theorem. However, many complex systems consistently produce highly skewed distributions of the magnitudes of outcomes such as hurricanes, solar flares, oil spills, pathogen counts in drinking water, and, apparently, human illnesses. These power law-like distributions, L-shaped on linear axes, plot nearly linearly on log-log scale. That is, the vast majority of such outcomes are very small, and the rare events quite large. In this talk we will consider how such distributions emerge. In particular, perhaps the most general nonlinear mechanism is that of first order growth (e.g., compound interest, population growth, chemical reaction kinetics). Recognition of the multiplicative and autocorrelated nature of this process, and its application to the growth of complex system outcomes from their causes, will be invoked to explain the L-shaped distribution, revealing an emergent, attracting Weibull form (continuous and proposed discrete). The form is nearly log-log linear, converging further with variability in the number of outcome causes, and is verified for simulated pathogenesis. From this form, an emergent dose-response function for chemicals, mixtures, and non-chemical stressors is derived. The use of predictive Bayesian techniques to address uncertainty in the parameters of the general dose-response form will then be discussed. The approach can be used to assimilate limited and potentially conflicting toxicity data, together with new information types such as genetic and other biomarker data, to extrapolate assessments to low doses of regulatory interest. In this way, the unconditional probability of illness may be assessed from available information, as a basis for the regulation of approximately 700 new chemicals introduced to commerce each year.

P.50 Transport related noise exposition in Santiago and the social benefits of the implemmentation of the new city transport system. De La Maza C*, Toha E, Cifuentes L; Pontificia Universidad Católica de Chile cdelamaza@ing.puc.cl

Abstract: Noise is an unwanted product of the modern lifestyle. It is an annoying auditory sensation and one of the environmental disturbances that significantly affect human health and quality of life. The main responsible of urban noise is traffic flow. As such, the overall objective of this study is to: first determine the base line noise level to which people are exposed at the sidewalk in different municipalities of Santiago and second to estimate the social benefits associated with the changes in exposure to noise due to the implementation of the new public transport system in the Metropolitan Region of Santiago, Transantiago. The methodology developed to accomplish the study objectives includes the utilization of the software ESTRAUS, that represents the state of the art in strategic transport models (de Cea (2001)), to determine the vehicular flow and speed in each arc of the city in the scenarios with and without Transantiago.To characterize traffic emissions at different speeds of travel, noise emission factors were estimated for different types of vehicles (public buses, heavy duty trucks and light passenger vehicles). The total emissions area calculated then integrating total traffic flow (depending on its composition) and unitary emission factors. Regarding the base line, the highest levels of exposure are located, as expected, in the center of the city and in the outlying areas close to major urban highways. The average exposure at the sidewalk is estimated to be above the values recommended by WHO (50 dB) and reaches levels considered dangerous in some parts of the city (> 70 dB). The changes in exposure to noise due to the implementation of the new public transport system range between 2.5 to 4 dB depending on the area. The social benefit are estimated multiplying this reduction in exposure (dB), with local unitary willingness to pay values estimated. The total benefit associated with the reduction of noise produced by the project is equivalent to MUSD$(2008) 168.1 per year.

P.51 Policy-induced risk transfer of lead in Asia: An analysis using the combination of the CGE model and the environmental dispersion model. Makino R*; National Institute of Advanced Industrial Science and Technology ryoji-makino@aist.go.jp

Abstract: In Asia, there have been serious problems of health hazards and environmental pollutions associated with lead at the secondary refining (recycle) sites in some countries. The lead pollution level of each country might be affected by the environment-related regulations. If each country adopted its own lead-related regulations without any international coordination, it might result in “risk transfer” among the countries, which means alteration of risk level due to the change of emission amount because of the regulation change. This could cause the international conflict. In this study, we simulate how various combinations of economic policies and environmental policies of each Asian country affect the international distribution of health and environmental risk of lead by using the computational general equilibrium (CGE) model based on economics and the environmental dispersion model. The CGE model can simulate the variation of the economic variables when the combination of hypothetical policies is implemented. We focus the variation of the production volume of the secondary refined lead of each country, which is seen as significant emission source. The health risk of lead in each country is quantified by using the hazard assessment data and the concentrations of lead estimated by the environmental dispersion model after the implementation of the policies. This study contributes to the quantification of the policy-induced international risk transfer by combining the economic model and the environmental dispersion model. An international policy coordination might be needed in order to improve the distribution of risk, that means to realize some kind of equality, subject to a given risk level in the whole Asia. This study provides the valuable information for environmental policy makers to establish the scheme to recycle lead internationally managing the level and distribution of risk of lead.

P.52 The air quality and human health effects of integrating utility scale batteries into the New York State electricity grid. Gilmore EA*, Adams PJ, Apt J, Lave LB; Carnegie Mellon University eagilmor@andrew.cmu.edu

Abstract: In a restructured electricity market, utility-scale batteries can generate revenue by discharging when electricity prices are high and charging when prices are low. This strategy, however, also changes the magnitude and distribution of air quality emissions, ambient concentrations, human health effects and social costs and benefits. We evaluate these effects with a case study of a 500 MW sodium sulfur battery displacing peak electricity generators in New York City from 1 – 5 pm and charging using off-peak generation in the New York Independent System Operator (NYISO) electricity grid from 1 – 6 am. First, we map displaced and charging plant types to generators in the NYISO. Second, we convert the changes in emissions into ambient concentrations with a chemical transport model, the Particulate Matter Comprehensive Air Quality Model with extensions (PMCAMx). Finally, we transform the concentrations into their equivalent human health effects and social benefits and costs. Focusing on the relationship between premature mortality and fine particulate matter (PM2.5), we calculate a benefit of 4.5 ¢/kWh and 17 ¢/kWh from displacing a natural gas (NG) and distillate fuel oil (DFO) fueled peaking plant, respectively, in New York City. By contrast, ozone (O3) concentrations increase due to the decrease in nitrogen oxide (NOx) emissions, although the magnitude of the social cost is less certain. For most charging plants, there is a net social benefit when a DFO peaker is displaced. By contrast, we find a net social cost if a NG peaker is displaced. By using the present base-load capacity for charging, the upstate population will experience an increase in adverse health effects. Newer wind generation, however, which could charge the battery, would ensure benefits to the upstate charging location and in New York City.

P.53 Benefits and Costs of the new national fine particulate matter standard for Chile. Rodriguez M*, De La Maza C, Cifuentes L; Pontificia Universidad Católica de Chile mayo.rod@gmail.com

Abstract: Following the international trend, the Chilean Environmental Protection Agency (CONAMA) decided to prioritize the establishment of a standard for particulate matter with an aerodynamic diameter less than 2.5 μg/m3 (PM2.5). The overall objective of this study is to assess the social desirability of adopting a quality standard for this pollutant. Internationally, the PM2.5 annual average standards range between 30 and 8 μg/m3. Although, the WHO recommends an annual average 10 μg/m3. Based on this information we proposed and evaluated the social benefit for four different air quality standards alternatives for Chile accounting for different time schedules to reach the WHO recommendations. For each of the alternatives analyzed, we determined the desirability of establishing such a standard from two perspectives: individual risks and social costs and benefits. Regarding individual risks the deaths avoided and the increase in life expectancy for each alternative evaluated are estimated. The alternatives evaluated will avoid between 128.000 and 161.000 long-term exposure mortality cases in the whole evaluation period (2010-2040). At the same time life time expectancy at birth will be increase between 8 and 12 month depending on the standard considered. The social benefits of the PM2.5 concentration reductions required by the standard alternatives were estimated using the unitary benefits for each μg/m3 of PM2.5 reduced estimated locally. The total benefits (at present value) of the alternatives considered range between 55,900 and 45,100 MUS$(2008). To estimate social costs of achieving the new standard we built curves of total cost by μg/m3 of PM2.5 reduced based on real mitigation measures proposed for each city considering minimizing the costs of improving air quality. We consider that the compliance costs decrease over time. The cost of the program for the most stringent alternative will be in average 1150MUS per year. The total costs (at present value) of the program range between 11,200 and 23,600 MUS$(2008).

P.54 A complex adaptive systems approach to cost-benefit analysis. Russo HA*; George Mason University hrusso@gmu.edu

Abstract: When determining the feasibility of potential public policies, traditional cost-benefit analysis is performed based on general equilibrium theory, which relies upon structural parameters estimated from the pre-policy environment. Often, formal general equilibrium models are used that rely on the notion of an optimal, sustained equilibrium that is derived from simultaneously solving a system of linear equations. Although some of these models allow for market imperfections, they do not allow for a sustained disequilibrium to emerge, which is more often the situation. In fact, exogenous shocks to an economic system that are brought about by new or amended public policies do trigger disequilibrium and structural changes in the economy that can be sustained or even permanent. Traditional general equilibrium modeling is inadequate for discovery of such emergent and potentially negative results: a more effective alternative capable of facilitating discovery of such results is a complex adaptive systems approach from the field of computational social science. This poster represents a Doctoral dissertation in progress in the field of computational social science. It is intended to advance the current cost-benefit analysis literature by proposing a complex adaptive modeling approach in place of the traditional general equilibrium modeling, to overcome the limitations of the latter outlined in this proposal. In addition, the field of computational social science will benefit from research into a multi-resolution modeling approach where various sub-populations of the system are modeled at different resolutions to allow for abstraction where possible and detail only where necessary. Finally, it is anticipated that research on the case study to which the complex adaptive approach will be applied – proposed tax incentives for the film industry in the state of Maryland – will inform the ongoing debate taking place on that topic in Annapolis.

P.55 Food product safety and competition in the US food retail industry: Implications for public policy. Mojduszka EM*; USDA/OCE/ORACBA emojduszka@oce.usda.gov

Abstract: In this paper we evaluate incentives to supply food safety in the US food retail industry in the period of 2003-2008. We concentrate on market driven private incentives related to food product safety and competition (brand entry and exit, firm (brand) reputation, and the impact on social welfare). To date, in the food economics literature there are no studies that evaluate empirically food product safety and competition as private incentives to supply food safety in the US food retail industry. This topic is especially important to investigate because public incentives to supply food safety in the US food retail sector are weaker than in the other food sectors of the US economy (e.g., food production and processing) and also weaker than in the food retail sectors of many EU countries. The reason is that US food safety policies target mainly the food processing sector and to a lesser degree the food production (farm) sector through the requirement of minimum safety standards for food products and production processes. We develop and estimate a random coefficient, discrete choice demand model for two fresh food product categories, fresh meat and fresh fruits and vegetables, using individual consumer home scan panel data for 2003-2008. We obtain demand parameters with respect to food product price and food product attributes (especially safety attributes) offered by various retailers. We utilize the estimated demand parameters to calculate the welfare effects of safety and competition on consumer and producer surpluses. We compare and contrast the similarities and differences between the two categories of fresh meat and fresh fruits and vegetables. Our theoretical and empirical comparative analysis of the discrete choice model of individual consumer and welfare effects allows us to propose changes in the current food safety assurance system that could strengthen private incentives and performance of US retailers.

P.56 Presenting Risk Mitigation Cost TradeOffs to Privately Informed Decision Makers. Ballard BD*; GRA, Inc bdballard@gra-inc.com

Abstract: In some security and other applications of risk mitigation analysis, it may be necessary for full information on hazards and threats to be available to risk analysts, even though such information is available to decisionmakers. In spite of this, open ended cost and risk reduction benefit comparisons can be constructed to better inform decisionmakers about the risk mitigation choices they face. In this presentation a methodology for addressing such settings will be developed. Specific applications of the methodology have been used in Transportation Security Administration regulatory impact analyses, and these examples will be discussed. Finally, a "nested" approach to improving the methodology in the face of sequential nonindependent efforts at security risk mitigation will be developed and discussed.

P.57 Naphthalene research: The relevance of tumors in rodents to human risk assessment. LeHuray AP*, Bird MG, Hammon TL, Juba MH, Lewis RJ, Reitman F, Sun T-J, White RD, Wise K; 1 Naphthalene Council; 2, 5 ExxonMobil Biomedical Sciences, Inc.; 3 ConocoPhillips; 4 Koppers, Inc.; 6 Shell; 7 Chevron; 8, 9 American Petroleum Institute alehuray@comcast.net

Abstract: Industry associations and individual companies formed the Naphthalene Research Committee (NRC) to co-sponsor research that strives to improve naphthalene risk assessments. The NRC’s objective is to reduce the use of default assumptions in assessing cancer risks potentially posed by exposure to naphthalene. To further its objective, the NRC funds basic research that is published in the peer-reviewed scientific literature. Several NRC-funded studies were completed in 2007 and 2008 that could significantly change a quantitative risk assessment based solely on default values. Acute (single 6-hour) and 5 day exposure studies examined whether the National Toxicology Program (NTP) chronic studies were conducted at concentrations that exceed the maximally tolerated dose (MTD). The two inhalation studies, conducted in two strains of rats, demonstrated the high concentrations used by NTP in the two-year studies caused severe damage to the olfactory epithelium - a target tissue for subsequent tumor formation. Other NRC-funded studies have included respiratory uptake studies of naphthalene in rats, results of which demonstrate that use of the Category 1 gas model is an inappropriate default assumption in assessing naphthalene risk. Respiratory uptake data and histological lesion mapping of the rat nose after naphthalene exposure are being used to develop a PBPK model incorporating naphthalene-specific data rather than regulatory default assumptions. Physiologic and metabolic differences between rodents and primates exposed to naphthalene introduce significant uncertainty into the applicability of studies conducted in rats and mice to human risk assessment. NRC is using the Human Relevance Framework and the Hypothesis-Based Weight of Evidence paradigm to evaluate assumptions, alternative hypotheses, and guide development of data needed for a robust mode of action hypothesis.

P.58 Hypothesis-Based Weight of Evidence (HBWoE) Evaluation of Naphthalene – Carcinogenic Hazard Assessment and Mode of Action. Bailey L*, Rhomberg L; Gradient Corporation lbailey@gradientcorp.com

Abstract: The National Toxicology Program (NTP) inhalation bioassay data for naphthalene showed increased nasal tumors in rats and lung tumors in mice, each response associated with marked target-tissue toxicity that may depend on local metabolic activation of naphthalene, suggesting a mode of action (MoA) dependent on regenerative hyperplasia. Although cytotoxicity and hyperplasia were observed in mouse nasal and to some extent in rat lung tissue, tumors were not observed in these tissues. CYP2F metabolism of naphthalene to genotoxic metabolites in conjunction with cytotoxicity has been invoked as a potential carcinogenic MoA for naphthalene. CYP2F activity has been observed in mouse lung and in rat and mouse nasal epithelium. The lack of mouse nasal tumors – despite the observation in the mouse nose of other key events in the proposed animal MoA – needs to be carefully considered, as the weight of evidence for the proposed MoA depends on the degree to which a scientifically supportable explanation can be identified that makes sparing use of post-hoc assumptions to accommodate the apparent discrepancy. It is noteworthy that several structurally similar compounds show the same syndrome in mice of lung tumors and nasal toxicity but no nasal tumors. To effectively communicate the uncertainties in the naphthalene data, giving proper consideration to apparent contradictions, we propose an approach, Hypothesis-Based Weight of Evidence (HBWoE), to organize, evaluate, and communicate the large body of available relevant data, considering potential key events in the MoA in each species and tissue. We evaluate the weight of evidence in support of a cytotoxic MoA, in addition to alternative MoAs that have been offered in the scientific community, noting the explanatory power of the proposed hypotheses, considering consistencies and inconsistencies within the data set, and how compelling each endpoint is as evidence of a similar MoA in humans from low levels of naphthalene exposure.

P.59 Using the human relevance framework as a guide to naphthalene research. Bird MG*, Lewis RJ, Piccirillo VJ; 1, 2 ExxonMobil Biomedical Sciences, Inc; 3 VJP Consulting alehuray@comcast.net

Abstract: The International Programme on Chemical Safety (IPCS) International Life Sciences Risk Sciences Institute (ILSI RSI) Mode of Action/Human Relevance (MOA/HR) framework (Meek 2008) is finding increased acceptance as providing a structured approach to articulation of the scientific evidence upon which conclusions are based. The framework also offers explicit delineation of inherent uncertainties and has been incorporated into international and national guidance and has been adopted in risk assessments in many countries. It is relevant to the consideration of both cancer and non cancer effects and robustly sets the stage for informing dose-response for the general population, including sensitive sub-groups. The framework also provides for the efficient assimilation of the totality of available data in a research context, facilitating the identification a mode of action and delineation of key events. The Naphthalene Research Committee (NRC) - a consortium of Industry associations and individual companies - is co-sponsoring research to improve naphthalene risk assessment and as is using the MOA/HR Framework together with the Hypothesis-Based Weight of Evidence paradigm (Bailey & Rhomberg 2009) to evaluate assumptions, alternative hypotheses, but also to guide development of data needed for a robust mode of action hypothesis to assess human relevance of the rodent findings. Based on the current research, a developing MOA/HR framework for naphthalene will be described. This will set out, transparently, the weight of evidence of a hypothesized MOA for experimental animals and humans in the context of key events taking into account the anatomical, physiological and biochemical differences found to date, and will identify areas to be addressed in the ongoing research program.

P.60 Time-dose-response model . Huang Y*, Haas CN; Drexel University yh89@drexel.edu

Abstract: While microbial risk assessment (MRA) has been used for over 25 years, traditional dose-response analysis has only predicted the overall risk of adverse consequences from exposure to a given dose. An important issue for consequence assessment from bioterrorist and other microbiological exposure is the distribution of cases over time due to the initial exposure. In this study, the classical exponential and beta-Poisson dose-response models were modified to include a dependency of time post inoculation (TPI) to quantify the time effect on dose response. 24 sets of animal survival data administered with graded doses of various pathogens including Bacillus anthracis, Yersinia pestis, Francisella tularensis and Mycobacterium tuberculosis were drawn from open literature for model fitting. By using the maximum likelihood estimation (MLE) approach, the resulting time-dose-response models were found capable of providing statistically acceptable fits to all tested data sets. This is the first showing that a time-dose-response model of an appropriate form and flexibility can be developed to model the temporal development of animal infectious response to pathogens with different characteristics. It provides an advanced approach for future microbial risk assessment frameworks. The outcome may be used for the improved understanding of emergency preparedness and post-exposure decision making, or as a component to better assist epidemiological investigations.

P.61 Assessing the risks of chemicals in the environment: Are pharmaceuticals different? . Cragin DW*; Merck & Co. david_cragin@merck.com

Abstract: Currently, there is much debate about the significance to human health of trace levels of pharmaceuticals in the environment. One of the questions raised was whether pharmaceuticals differ from other chemicals found in the environment and if different methods are needed to assess their potential risk to health. To help answer this question, the NRC's Standing Committee on Risk Analysis held a symposium on whether new risk assessment methods are needed to assess the risks of low doses of pharmaceuticals in drinking water. This talk reiterates a discussion from the NRC symposium that compares pharmaceuticals with other chemicals as regards risk assessment. It shows how the wealth of animal and clinical data on pharmaceuticals can be used to address the hazard identification, dose-response, and exposure elements of the risk assessment paradigm.

P.62 Tolerance intervals on bioassay test results to assess total variability in unbalanced multi-components of variance settings. Feder PI*, Ma Z; Battelle feder@battelle.org

Abstract: Government scientific and regulatory agencies such as EPA, FDA, CDC, and NIH rely on the results of in vitro and in vivo bioassays as primary sources of data upon which to base inferences concerning safe doses or concentrations to which humans or animals can be exposed in the environment, in food, or in the workplace. Proper estimation of the uncertainties in these estimates is essential to the risk assessment process since they are the basis of inferences concerning lower bounds on permissible exposure levels associated with a specified level of risk or to upper bounds on risk associated with a specified level of exposure. It is thus essential to assure that the estimates of risk uncertainty reflect total variability. The total variability is composed of multiple components: Among laboratories, among tasks within laboratories, among runs within tasks, within runs. In particular, estimates of uncertainty based on just a single study at a single laboratory at a single time will lead to underestimates of variability because they do not reflect all the components of variation. This can lead to overestimates of safe doses or to underestimates of risk. This presentation will illustrate statistical methods underlying the construction of tolerance bounds on risk estimates in unbalanced components of variance settings that reflect all the components of variation. The lower tolerance bounds on reference doses will in general be more conservative than simple confidence bounds based on single studies and will provide more realistic inferences about reference doses. The tolerance bound methodology can also be used to screen future test results and additional test laboratories for acceptable conformance with past results. The results of the tolerance bound calculations and their application to screening test laboratories will be illustrated by example in tabular and graphical form.

P.63 Exposure to metal mixtures and kidney function: Investigating health disparities. Fox MA*, Chari R; Johns Hopkins University mfox@jhsph.edu

Abstract: African Americans, Hispanics, and Native Americans are more likely to develop chronic kidney disease than whites. Reasons for these disparities are not entirely clear and could include lack of health care, comorbidities, or chronic exposure to environmental toxins. We examined exposure to a mixture of nephrotoxic metals and measures of kidney function using urinary metal levels, health exam and questionnaire data from the National Health and Nutrition Examination Survey 2001-2004. The metals are barium, cadmium, cobalt, lead, molybdenum and uranium. We were able to include mercury and arsenic for subsets of the study population. We developed a mixture exposure metric (MEM) using the relative potency method as described in US EPA guidance on chemical mixture risk assessment. We conducted regression analyses to examine the association between the mixture exposure metric and estimated glomerular filtration rate, creatinine clearance, and albumin-creatinine ratio. Results of the exposure assessment and statistical analyses of exposure and kidney function will be presented and discussed.

P.64 Impact of integrating age-specific water intake rates into derivation of Minnesota groundwater guidance . Goeden H, Moyer P, Hassan I, Greene C*; Minnesota Department of Health paul.moyer@state.mn.us

Abstract: The Minnesota Department of Health (MDH) protects public health by establishing guidance values for contaminants found in groundwater, which is the major source of drinking water in Minnesota. This guidance, expressed as ug/L of groundwater, represents a concentration that is without appreciable risk to human health. The magnitude of these guidance values is a function of toxicity (indicated by a reference dose (RfD) or cancer slope factor (SF)) and the amount of water ingested. Historically, standard risk assessment methodology has followed the assumption that protecting against adverse health effects from chronic exposure would also be health protective of short-term exposures. As a result, health-based drinking water standards were derived by combining a chronic intake rate (2 L/day and 70 kilogram body weight) with an EPA IRIS chronic RfD. MDH’s recently revised process for developing guidance includes identifying less-than-chronic effects (e.g. immunotoxicity) as well as chronic health endpoints. The timing of the observed critical effect is then paired with the appropriate intake rate to ensure susceptible and highly exposed subpopulations are protected. MDH has found that in general, shorter-duration RfDs are higher than chronic duration RfDs. However, the magnitude of the health-based guidance value is also determined by the intake rate. Shorter-duration intake rates, particularly when factoring in younger age-groups, are up to seven times higher than average chronic intake rates. This combination of shorter duration intake rates and RfDs may result in a shorter-duration guidance value that is significantly lower than the chronic value. For example, an evaluation of chloroform using the historic chronic intake yields a value of 70 ug/L, while the MDH methodology derives a chronic value of 30 ug/L based on a short-term exposure and intake rate. Additional examples of assessments for specific perfluorinated chemicals, solvents and pesticides will be presented.

P.66 Risk Assessment of Lead Intake from Food in Japan. Koizumi Na*, Kumada Hi; Commissioner, Food Safey Commission, Japanese Government hirotaka.kumada@cao.go.jp

Abstract: Lead is known for its health effects, famously in the form of lead poisoning, the occupational disease known for decades among workers dealing with lead. In Japan, air pollution had been a serious problem until the end of 1960s associated to the use of lead gasoline, but the lead concentration in the air has been considerably improved today compared to 1970s, largely owing to the tightened regulation of lead use. In 2007, the earthen pots and the paints of toys imported from China showed above the maximum level of lead criterion of this kind, and the event triggered a call for reviewing and harmonizing the conventional Japanese lead standards to the international standards. Therefore, the Food Safety Commission decided to carry a risk assessment of lead in foods for effect on the human health. The lead working group is set up under the Expert Committee for Chemical Substances and Contaminant to deliberate the assessment. Previously, many in vivo and human epidemiological studies have reported various harmful effects of lead mainly in the areas of neurology, hematology and hematopoiesis, and reproduction. Among them, our working group focused on the neurological effects because a large number of lead-related neurological studies were published, and the neurological effects appeared to be the most sensitive toxicological aspect of lead. The assessment targeted children (including fetus and potentially pregnant women) as the high risk group. The studies used for children’s assessment include cohort studies determining IQ-blood lead level relationship (IQ served as the indicator of neurobehavioral development) and epidemiological insights drawn from cross-sectional studies. In contrast, for adults, there are only few epidemiological studies available to describe the impact of lead exposure in general environment. Consequently, the working group mainly evaluated epidemiological insights drawn from the high lead exposure studies among lead-handling workers.

P.68 Health Assessment of Occupational Exposure among Motorcycle-machine Repairmen. Ho WC*, Lin MH, Chen CY, Lin JD, Chen CJ, Lia JS, Wu TN; China Medical University whocmu@gmail.com

Abstract: Motorcycle is a very popular transportation tool in Taiwan and also a growing fast market in China. This study was to use a cohort data of 20,639 motorcycle-machine repairmen with certifications on their professional skills provided by Taiwan Labor Department to reach the following aims: 1) to establish a comprehensive data source providing health-related information on motorcycle-machine repairmen with professional certifications; 2) to describe the demographic features and the types of work-related injury, disability and disease; 3) to explore the relative occurrence of diagnostic diseases and cancers in the cohort of motorcycle-machine repairmen with the comparison to the general populations by identifying the indicator of the Standardized Incidence Ratios (SIRs). The data for analyses were drawn from the Worker Database in the Department of Labor Insurance and from the Cancer and Disease Registered Database in the Department of National Health Insurance in Taiwan. By using indirect standardization method for obtaining the annual and cumulative incidence ratios for the period 1997-2002 for cancers (SIRs), the results were shown that the motorcycle-machine repairmen were at greater risk than the general population for having the following four cancers: Lip, Oral Cavity and Pharynx、Digestive Organs and Peritoneum、Hematopoietic and Reticuloendo-thelial Systems and Thyroid Gland. The statistical results for analyzing prevalent diseases and the work-related injury among workers showed that for most of the workers the prevalent diseases were respiratory system disease and digestive system disease, and most injured body part was the fingers. The results in this study provided empirical evidence to support that motorcycle-machine repairmen comparing to general populations indeed have higher chances to have cancers. The chances could be possibly attributed to the exposure to the health hazards in doing the work. Advanced health hazard assessment is in need.

P.69 Demonstration of How Sensitive Subpopulations Might Be Considered in Regulatory Benefits Analyses. Greco SL*, Lynch MK, Corrales M; Abt Associates Inc. sue_greco@abtassoc.com

Abstract: Sensitive subpopulations may be defined by exhibiting a greater response after exposure to a particular pollutant. Several sensitive subpopulations, defined by socioeconomic (e.g., low income, low educational attainment) or demographic factors (e.g., children, elderly, African American), have been studied in previous human health benefit analyses; however sensitive subpopulations, as defined by gene-environment interactions have not. Gene-environment interactions (GxE) refer to the biological interplay between an individual’s genetic composition and environmental exposures due to toxins and other agents. In this example, we focus on how polymorphisms (the occurrence of two or more forms of a gene, or genotypes, in a population) contribute to interindividual differences and potentially greater susceptibility due to pollutant exposures. Emerging GxE research is making benefit analyses in sensitive subpopulations possible. Using a framework developed to investigate how fine particulate matter (PM2.5) mortality could be modified by socioeconomic and demographic factors (education attainment, Levy et al. 2002), this case study demonstrates how such a relationship might be modified by gene-environment interactions. We propose considering GxE interactions using a modified “impact function” framework. Such a framework would combine a) an effect estimate stratified by genotype from the epidemiological literature, b) baseline incidence rates for the health effect of interest for each genotype, c) the size of the exposed population and prevalence of the high-susceptibility genotype, and d) the change in the exposure metric of interest (e.g., reduction in air pollution attributable to a new rule). The main limitations of benefits analyses considering GxE interactions are data availability (e.g., epidemiological evidence for populations with different genotypes) and scale/resolution of the analysis (e.g., exposure assessment at the level of detail necessary).

P.71 Is it safe? Analysis of Compact Fluorescent Lamps Breakage and Potential Mercury Exposure. Nance P*, Patterson J, Willis A, Kroner O, Dourson M, Foronda N; Toxicology Excellence for Risk Assessment; New Zealand Ministry of Health nance@tera.org

Abstract: What are the potential risks when a compact fluorescent lamp (CFL) is broken? CFL’s are promoted as being more energy efficient and an ecofriendly replacement for incandescent lamps, but exposure to broken CFLs can potentially lead to inhalation and dermal health effects from exposure to mercury. The amount of mercury in CFLs can vary based on manufacturer and wattage. This concern is being raised around the world, as many countries are setting manufacturing regulations on mercury usage in CFLs. In this analysis, risk to infants and adults were evaluated based on two exposure scenarios: CFL breakage in a room with no ventilation and no cleanup, and CFL breakage in a room with adequate ventilation and cleanup. Concentration data from a study by Stahler et al. (2008) were compared to human toxicity benchmarks to calculate hazard quotients. For the no cleanup scenario, hazard quotients were generally less than one, an unlikely health risk. When the room was ventilated and the CFL was cleaned up, mercury concentrations were lower. A review of release scenarios, along with duration adjusted toxicity benchmarks, indicates that few releases produce levels of concern, but some scenarios can exceed risk targets and require further study. Uncertainties in the screening characterization include assumptions about room size, ventilation, age of bulb, the distribution of mercury in the room, and also the choice of the toxicity benchmarks used to develop the hazard quotients. This project was conducted for the New Zealand Ministry of Health, under the auspices of the Alliance for Risk Assessment (ARA), a collaboration of stakeholders representing government, academic, industry, environmental and consulting perspectives.

P.72 Inhibition of Acetylcholinesterase in a Farmworker Population Exposed to Organophosphate Pesticides. Griffith WC*, Vigoren EM, Coronado GD, Thompson B, Faustman EM; University of Washington, Fred Hutchinson Cancer Research Center, Seattle WA griffith@u.washington.edu

Abstract: A longitudinal study of farmworkers and non-farmworkers adults and children was conducted in the Yakima Valley in eastern Washington State to characterize exposures to organophosphate pesticides (OPs). Our analysis included evaluation of blood, urine and dust samples collected April to July 2005 when the farmworkers were employed as thinners in fruit orchards and for samples collected during the off season November 2005 to February 2006 when no work occurred in the orchards. Dust samples collected from homes and work vehicles were analyzed for 6 OPs; urine samples were analyzed for the diakyl phosphate (DAP) metabolites of OPs, and blood samples were analyzed for acetylcholinesterase (AChE), a biomarker for OP exposure. We are using Bayesian Markov chain Monte Carlo methods to build an integrative modeling tool describing the joint distributions of the data for predicting OP exposures. The analysis of our dust samples showed that the most common OP was azinphosmethyl (AZ), and the analysis of the urine showed that the metabolite with the highest concentration was dimethylthiophosphate (DMTP), a metabolite of AZ and other commonly used dimethyl OPs in the Yakima valley. Concentrations of AZ in dust were significantly higher in farmworker homes and vehicles, and concentrations of DMTP in urine were higher in farmworkers and their children than nonfarmworkers and their children during the April to July thinning season. During the November to February off season the farmworkers and their children had DAP metabolite levels in urine similar to nonfarmworkers and their children. Adult AChE inhibition was significantly correlated with increasing DMTP concentration during the thinning season for the farmworkers. (Supported by grant 2-P01-ES009601 from NIEHS and RD-8370901-0, STAR RD-83273301-0 from US EPA. Contents are responsibility of the authors.)

P.73 In vitro-in vivo extrapolation of the dose-response relationship for cellular perturbations by toluene using a cellular dosimetry model. Peyret T*, Krishnan K; UNIVERSITY OF MONTREAL, MONTREAL, CANADA thomas.peyret@umontreal.ca

Abstract: There is an urgent need to develop and validate tools and approaches for conducting in vitro-in vivo extrapolation of the data on perturbations of toxicity pathways and effects observed in cellular tests. In this study, cellular-level dosimetry models were developed to describe in vitro and in vivo systems, using toluene as the model substance. The tissue compartment of the in vivo PBPK model consisted of four matrices: cell, interstitial fluid (IF), plasma and erythrocyte; whereas the in vitro model contained two matrices, namely the cell and the culture medium (CM). Matrix (i.e. CM, cell, interstitial space, plasma and erythrocyte):water partition coefficients (PCs) were used as the basis to predict the cell:CM, cell:IF, and IF:blood PCs for the in vitro or in vivo models. The matrix:water PCs in turn were estimated on the basis of the physicochemical properties of toluene as well as the fractional content of neutral lipids, neutral phospholipids, water, and protein. Following the incorporation of the PCs and literature data on metabolic parameters within the cellular-level PBPK model, the human exposure concentrations of toluene that yield the same brain cell concentrations as in the in vitro SH-SY5Y cell cytotoxicity study (McDermott et al., Toxicol Appl Pharmacol, 219:85-94) were determined. For LDH leakage and intracellular Ca2+ effects, the in vitro concentrations of 5.64 (NOAEL), 13.1 (LOAEL), 22.6, 47.0, and 76.5 µM corresponded to the human exposure concentrations of 154.4, 256.8, 376.2, 678.3, and 1043 ppm based on identical brain cell concentrations at steady-state. In comparison, the less serious LOAELs for neurological effects have been reported to range from 35 ppm to 200 ppm for humans. Overall, the cellular dosimetry model developed in this study represents a potentially useful tool for conducting in vitro-in vivo extrapolation of the cellular-level effects and perturbations evaluated in high throughput assays. (Supported by AFSSET).

P.74 Development of community based risk assessment system for integrated environmental risk management in Korea. Shin DC*, Yang JY, Lim YW, Kim JY, Ann YJ, Bae HK; Yonsei University, Korea dshin5@yuhs.ac

Abstract: We have developed the 'system software' for health risk assessment of multimedia and multipathways scenarios including Geographical Information System (GIS) and have used it as Decision Support System (DSS) for efficient management of environmental risk in municipal and industrial areas in Korea. We assessed 7 metropolitan cities and 10 industrial areas. For risk evaluation of 120 chemicals including PAHs, VOCs, pesticides, POPs and metals, we used the simulated data by the multimedia fate model and field monitoring in study areas. As a follow-up project, we are conducting community based risk assessment (CBRA) research for integrated environmental risk management in Korea (2009-2010). We review the available tools (models, methods, data, approaches) for community-focused cumulative risk assessments and develop the guideline and tools to assess community cumulative risk through selected applications of case studies. The case study areas are an industrial area (urban type) and a mining area (rural type) for heavy metals and VOCs (volatile organic compounds). We developed the tools for CBRA including worksheet for identifying and ranking pollutants and sources, data collection and characterization system of pollutants at community level, personal data collection system of human exposure-related behavior, human and ecological risk analysis process for community population, environmental fate model for risk reduction planning, and documentation on how to set-up risk reduction scenarios. The developed CBRA documents and tools guide users through a comprehensive environmental health assessment that will provide an accurate and verifiable profile of the community’s environmental health status. And then community environment officials and stakeholders can use this profile for proactive, locally appropriate decision making.

P.76 Critical Evaluation of The Hazardous Substances Data Bank (HSDB) for Use in Current and Future Risk Assessments . Dourson M*, Jayjock M, Maier M, Willis A, Parker A, Haber L, Patterson J; Toxicology Excellence for Risk Assessment (TERA) dourson@tera.org

Abstract: We provide an assessment of the state of the science information in the current HSDB from a risk assessment/risk management perspective. We did this in two ways. We first conducted an informal, 2-question survey of forty-four experienced risk assessment scientists on HSDB use. Second, we analyzed individual HSDB categories in reference to a well studied chemical, acetone, and against our collective knowledge of different risk assessment disciplines and new science developments. The informal survey resulted in a mean frequency of use (on a scale from 0-3.0) of 1.8, indicating that HSDB was often used when starting a new assessment. The mean frequency of trust was a value of 1.7, indicating that HSDB was trusted-but-spot-checked. Two additional literature surveys testing different aspects of HSDB gave it high marks, and webstats indicate heavy use, especially by health care providers, toxicologists, other scientists and researchers, and citizens. HSDB contains a wealth of information making it a very useful starting place for chemical assessments. The overview and excerpted information in each category is very useful as a first approach to a chemical’s potential hazard, dose response assessment, exposure assessment and risk characterization. Areas for improvement were noted and include: • A standard group of categories, similar to other U.S. federal agencies allowing all agencies to use HSDB as a first stop in developing their chemical-specific information; • New categories on the latest science, such as genomics, nanoparticles, and modeling; • Regularly scheduled updates with exceptions for topics of high importance; • Better links to developing databases such as Chemical Hazard Emergency Medical Management (CHEMM), Wireless Information System for Emergency Responders (WISER) and Radiation Event Medical Management (REMM); and • A revision of the Occupational Exposure Limit (OEL) section that would allow HSDB to be the leading source of information for occupational risk assessors.

P.77 Application of disability-adjusted life year (DALY) in a food industry context. Kan-King-Yu D*, Moretti D, Membre JM; Unilever Safety & Environmental Assurance Centre (authors 1 and 3); Unilever Food and Health Research Institute (author 2) Denis.Kan-King-Yu@Unilever.com

Abstract: Since its first introduction in the World Development Report 1993 (World Bank, 1993), the Disability-Adjusted Life Year (DALY) has generated interest in summary health measures. Primarily used (World Health Organisation, 2001) as a metric for quantifying health impacts caused by environmental risks at a population level, the attractiveness of the DALY resides in the fact that it combines information about mortality and morbidity in one single number. In food safety, acceptable fortification levels of micronutrients are based on identified safe levels following the risk assessment framework. However, the number of people (in a target population/sub-population) affected by an adverse effect due to inadequate exposure to the fortified product is not estimated. In other words, the product of the dose-response probability density function by the daily intake probability density function is not mathematically integrated over the range of possible doses. Similarly, the benefits of the intended fortification are often not fully quantified. This paper assesses the use of the DALY as a single metric for integrating the risks and benefits of supplementing food products with micronutrients. One case-study is presented: a 250 ml drink fortified with iron at 15% above the Recommended Daily Allowance (RDA) in Argentina. The adopted methodology is based on the tiered approach recommended by BRAFO (ongoing ILSI project funded by the European Commission). First, the risks and benefits associated to the non-fortified product (reference scenario) are identified. Then the expected change in DALY is estimated after fortification. From an industrial perspective, the use of the DALY has the potential to better guide research priorities with regards to “health interventions” (here through the fortification of food products with micronutrients), as it allows the quantification of the effects of health interventions at the population level. The use of the DALY looks promising, as a rigorous, reliable and global health measurement. Advantages and disadvantages are discussed further.

P.80 Utility of a C6-glioma system for exploratory risk assessment of environmentally relevant mixtures of insecticides. . ROMERO DM*, ALAIMO A, GOROJOD R, KOTLER ML, WOLANSKY MJ; Argentine National Research Council - University of Buenos Aires mjwolansky@gmail.com

Abstract: Real environments and food products may carry low levels of multiple hazardous compounds having diverse modes of action. For instance, according to recent single-compound toxicological information from rats and environmental data, organophosphate (OP) and pyrethroid (PYR) insecticide residues pollute indoor and outdoor settings but in levels that imply no health risk for humans. However, there is a data gap on the neurotoxicology of complex insecticide mixtures. Here we present an exploratory, in vitro model, aimed to identify toxicologically relevant combinations of insecticides that require CRA-like research efforts in in vivo models of greater human relevance. We will show time (4-24-48 h)- and dose (0.5-250 uM; DMSO-vehicle control and five insecticide doses)-effect data for two OP and four PYR compounds. We use a C6-glioma culture system and a battery of effect measures to examine the individual and combined action of these insecticides. Moreover, two fetal calf serum (FCS) conditions are assayed (FCS 2-10%). Using 24h-exposed, C6 cultures in D-MEM supplemented with 2% FCS, followed by MTT assays for citotoxicity in 96-well plates, a high correlation between BMD15 estimates for cell viability and rat oral LD50s was apparent for three PYRs (deltamethrin>bifenthrin>tefluthrin). Chlorpyrifos (OP) and alpha-cypermethrin (PYR) did not fit well within this relative potencies trend. Acephate (OP) produced no evident C6-cell viability decrease in any exposure condition. We are presently testing if Hoechst-33258 histological observations and AchE assays in the C6 system may help interpreting above findings. These pilot studies will be used to appropriately design in vivo, environmentally relevant OP-PYR mixture studies aimed to test additivity using a battery of biochemical and neurobehavioral endpoints of neurotoxicity in the rat.

P.82 Toxicological analysis to determine the TDI of melamine in food. Hsieh DPH*, Chiang CF, Chiang PH, Wen CP; China Medical University (DPHH CFC), National Health Research Institutes (PHC CPW) dphsieh@mail.cmu.edu.tw

Abstract: Intensified food safety concern over melamine has prompted national authorities to assess its tolerable daily intake (TDI) for protection of general population including young children. TDI is calculated by dividing a no-observed-adverse-effect level (NOAEL) by a safety factor (SF). Based on appropriate choices of values, the U.S. Food and Drug Administration determined two TDI values in the unit of mg per kg body weight per day as first 0.63 and then 0.063, while the World Health Organization, 0.5 and then 0.2, as a result of increasing the SF values in calculation. We used the benchmark dose analysis method on the same set of toxicity data to determine an appropriate NOAEL, and then used an SF of 1000 to obtain a TDI of 0.0081. Arguments in support of this considerably lower TDI value were provided to alert the international community. (Sponsored by the NHRI Priority Action Plan Fund, 2008)

P.83 Development of dose-response curves using a Bayesian approach to model variation in inoculate infectivity. Mitchell-Blackwood J*, Gurian PL, Thran B, Taft S, Lee R, Hines S; Drexel University jade@drexel.edu

Abstract: An understanding of the dose-response (D-R) relationship is imperative for setting exposure guidelines, developing health-based detection systems, and establishing remediation goals. Because of many factors including (1) the lack of conclusive data, particularly at low doses, (2) the use of surrogates, and (3) the uncertainty and variability associated with environmental conditions– the analysis of D-R data from historical studies by classical methods have been limited based on the validity of many assumptions and the relevance of experimental conditions to overall conditions. The authors present an analysis by Bayesian methods, which relax some of these assumptions and produce a generalized prediction of infectivity. For the Bayesian approach a hierarchical model was fit to describe observations of survival or mortality of individual animals exposed to different environmental inoculates. This model describes the results from one experiment as being derived from an overall distribution of potential experimental outcomes. In this study the Bayesian hierarchical approach was used to analyze a rich set of historical data consisting of over 11,000 animals to obtain posterior distributions of exponential D-R parameters. The analysis for individual experiments usually produced parameter estimates with greater confidence than classical methods. D-R curves generated by the Bayesian hierarchical approach allowed for predictions to be made for unobserved or poorly observed pathogenic agents based on the observed data available for analysis with a quantifiable degree of uncertainty. A broad range of benchmark values for use in decision making have been generated. These values provide health-protective estimates of infectivity and have been informed by a large body of evidence. Bayesian methodology can be a promising approach for addressing and quantifying many experimental uncertainties simultaneously in D-R assessment and more broadly in microbial risk assessment.

P.86 Quantitative risk assessment on the effect of Listeria monocytogenes contamination in deli meats originating from manufacture and retail, on listeriosis cases. Pradhan AK*, Ivanek R, Gröhn YT, Bukowski R, Geornaras I, Sofos J, Wiedmann M; Cornell University, Texas A&M University, Colorado State University akp49@cornell.edu

Abstract: Listeria monocytogenes contamination in ready-to-eat (RTE) foods represents a considerable public health concern. For better control and effective allocation of resources, it is important to provide estimates on the attribution of human listeriosis cases to L. monocytogenes contamination of foods originating from manufacture and retail. The objective of this study was to estimate the relative risk of listeriosis attributable to L. monocytogenes contamination in ham and turkey, formulated without and with growth inhibitors (GIs), under two scenarios: (i) in prepackaged deli meats with contamination originating solely from manufacture at a frequency of 0.4% (based on reported data), and (ii) in retail-sliced deli meats with contamination originating solely from retail at a frequency of 2.3% (based on reported data). Using a manufacture to consumption risk assessment with product-specific growth kinetics parameters (i.e., lag phase and exponential growth rate), we estimated that 76 and 63% of listeriosis cases caused by ham and turkey (formulated without GIs) and 83 and 84% of listeriosis cases caused by ham and turkey (formulated with GIs) can be attributed to contamination originating at retail. Reformulation with GIs was estimated to reduce human listeriosis cases linked to ham and turkey, by about 2.8- and 9-fold, respectively, when contamination originated at manufacture and by 1.9- and 2.8-fold, respectively, for products contaminated at retail. Sensitivity analyses indicated that storage temperature was the most important factor on the estimation of per annum relative risk. Scenario analyses suggested that reducing storage temperatures to be consistently below 7°C would greatly reduce the risk of human listeriosis while reducing storage time appears to be less effective. Overall, our data indicate a critical need for further development and implementation of effective control strategies to reduce L. monocytogenes contamination at retail.

P.87 Variation in Listeria monocytogenes dose response in relation to Internalin A subtypes. Chen Y*, Ross WH, Whiting RC, Van Stelten A, Nightingale KK, Wiedmann M, Scott VN; Grocery Manufacturers Association, Health Canada, Exponent, Colorado State University, Colorado State University, Cornell University, Grocery Manufacturers Association ychen@gmaonline.org

Abstract: Recent advances in molecular subtyping and knowledge of pathogenicity have established a critical role for the surface protein Internalin A (InlA). Mutations in InlA leading to one of 13 premature stop codons (PMSCs) result in a Listeria monocytogenes (Lm) subtype with truncated InlA associated with a reduced ability to invade intestinal epithelial cells. The presence of PMSCs identified for 502 food isolates from a retail survey and 507 US clinical isolates (including 42 from two FoodNet sites in which the retail survey was conducted) allowed us to determine the prevalence of subtypes with or without PMSCs among the food and clinical Lm isolates, as well as the concentration distributions for these subtypes in foods. The new subtyping data were input into two published Models to generate an r-value (essentially the probability of a cell causing illness). Making the conservative assumption that the levels at retail were the levels consumed and using an Exponential approach, the mean log(r) values were -9.4, -10.7 and -8.1 for all Lm and for Lm subtypes with and without PMSCs, respectively. For one particular subtype, PMSC4, the mean log(r) was -11.4 using the Exponential approach. Similar r-values were obtained using a Subtype-specific Model based on a conditional probability approach. Growth between retail and consumption was also factored into estimating the r-value using growth rates and time/temperature data reported in 2003 and/or 2006 and 2007. With growth considered, the mean r-value was approximately lower by 1.6-2.5 log, 2.2-2.8 log and 2.1-2.5 log for all Lm and for Lm subtypes with and without PMSCs, respectively. Work is underway to further quantify uncertainty and to investigate the dose response relationship using a semi-parametric approach. Results from this study and further efforts integrating genomic data into hazard characterization will enable better attribution of risk to L. monocytogenes subtypes.

P.88 Quantitative microbial risk assessment for Salmonella in Peanut Butter. Schaffner DW*; Rutgers University schaffner@aesop.rutgers.edu

Abstract: Two large outbreaks of salmonellosis associated with peanut butter in the United States in 2006-2007 and 2008-2009 have recently underscored the potential of this shelf-stable food to cause large outbreaks. The objective of this work was to develop a quantitative microbial risk assessment (QMRA) that incorporates key variables such as: initial concentration of Salmonella the degree of thermal process applied, and the amount of peanut butter per finished product serving and the total number of servings on the market. The QMRA was developed using @risk (Palisade Corporation, Ithaca, NY). Non-linear inactivation kinetics for Salmonella in peanut butter was based on published models. These models were developed in model systems (not peanut butter) so they were also validated by comparison with death kinetics for Salmonella in actual peanut butter. A Salmonella dose response model was adapted from the one used for the FAO/WHO risk assessments of Salmonella in eggs and broiler chickens. Good agreement was seen when the thermal inactivation data in peanut butter was compared to the models. The QMRA demonstrates that even given a low initial concentration of Salmonella cells (1.5 CFU/g), what would normally be considered a severe process (60 C, 10 min), and a small amount of peanut butter in the finished product (~3 g/ serving), when a large number of servings (~1.5M) are considered, a significant number of illnesses (>10) are typically predicted by the risk assessment. An analysis of the publicly available data from the two outbreaks leads to similar findings Even low initial levels of Salmonella contamination may not be controlled by typical peanut butter thermal processes, and can result in large numbers of illnesses. The QMRA will be useful for food companies seeking to assess the risks posed by particular raw material standards or standard processes.

P.89 Modeling and risk assessment for the growth of Salmonella on cut tomatoes. Schaffner DW, Pan WJ*; Rutgers University schaffner@aesop.rutgers.edu

Abstract: Outbreaks of salmonellosis associated with fresh tomatoes continue to be major food safety concern. When whole tomatoes are cut, any Salmonella present can be transferred from the surface to the inside of the fruit. Despite their low pH, cut tomatoes provide a favorable environment for Salmonella to grow when temperature abused. Few reports quantify the growth rate of Salmonella on cut tomatoes, and only at limited number of incubation temperatures. While modeling software (the pathogen modeling program - PMP and ComBase Predictor - CBP), can be used to predict Salmonella growth, no models have been validated for cut tomatoes. Our purpose was to build a mathematical model to describe the growth of Salmonella on cut tomatoes and use that model to assess the risk posed by incubation at different temperatures. Whole red round tomatoes were dip inoculated with a cocktail of four Salmonella strains obtained from the Centers for Disease Control and Prevention (CDC). Inoculated tomatoes were dried overnight, cut with a sterile knife, placed in a sterile plastic bag, incubated at 10 temperatures and plated on XLT4 agar for enumeration. Growth rates were calculated using DMFit. Relevant data on Salmonella prevalence and concentration were determined from the literature. Temperature data were obtained from a survey of holding practices in Rutgers University dining halls. The software @risk (Palisade Corporation, Ithaca, NY) was used for risk modeling. A plot of the square root (SR) of growth rate (GR) was linear with temperature (T), such that SR(GR) = 0.026T - 0.1065, (R2=0.9398). The growth rate was slightly faster than that predicted by CBP, but slower than that predicted by PMP. The Salmonella in cut tomato model created here provides a useful tool for estimating the risk posed by different abuse temperatures. The results show that a small number of food samples held at elevated temperature conditions contribute to the greatest risk.

P.90 Estimating Salmonella attribution from chicken, pork and ground beef in Waterloo Region using a quantitative risk assessment approach. Phinney R, Otten A*, Fazil A; Public Health Agency of Canada, 110 Stone Road West, Guelph, Ontario, Canada ainsley_otten@phac-aspc.gc.ca

Abstract: Salmonellosis was the second highest reported enteric disease in the Region of Waterloo, Ontario, Canada in 2006. Retail contamination data from June 2005 through June 2008 was available through C-EnterNet, a federally funded surveillance program currently operating in the Waterloo Region, for Salmonella in chicken breasts, pork chops and ground beef. This data allowed for a model to be developed from the retail level, predicting the proportion of illness attributed to each of the three fresh meat products. The model contains modules that take into account the behaviour of Salmonella on each meat product throughout the retail to consumption pathway, and was created using the @Risk v 4.5 platform for Microsoft Excel. The model estimates the probability of illness per serving of each type of meat; and then a relative risk calculation is performed to estimate attribution of salmonellosis from the three types of meat considered. The results indicated that if a person were to acquire salmonellosis from food consumption, and they had consumed all three types of meat, the likelihood that they became ill from consuming the chicken breast would be much greater than the chance of attribution from either the pork or beef. These results suggest that chicken breast consumption provides a greater risk of illness than pork chop or ground beef consumption in Waterloo Region when consumed in equivalent amounts. The model has the potential to be developed further, as additional transmission pathways could be included along with other chicken, pork and beef products, if appropriate data was available. Other pathogens could also be investigated using the developed model framework and regional data collected by C-EnterNet.

P.92 FoodRisk.org: an online resource for food safety risk analysis. Ruzante JM*, Mckillop K, Gendel SM; Joint Institute for Food Safety and Applied Nutrition, US Food and Drug Administration jruzante@umd.edu

Abstract: Science plays a vital role in food safety assurance through the application of risk analysis. FoodRisk.org is an on-line resource that supports the use of food safety risk analysis by i) providing quality risk assessment resources with a special emphasis on data and models, ii) providing an introduction to the theory, methods and practice of risk management applicable to food safety, iii) providing an introduction to the theory, methods and practice of risk communication applicable to food safety, iv) establishing and maintaining open channels of communication within the food safety risk analysis community, and v) promoting the use of food safety risk analysis globally. FoodRisk.org collects and catalogues available data and methodology on food safety risk analysis from the private sector, trade associations, federal and state agencies, and international sources. Each resource is evaluated for relevance, reliability, timeliness, quality and value before being included in FoodRisk.org. Examples of the data and tools that are made available through Foodrisk.org are the EcoSure dataset on cold temperature storage of food products during transportation and 24 hours after being placed in consumer home refrigerators/freezers, an Australian risk assessment model for Listeria monocytogenes in ready-to-eat meats and the updated version of the food handling practices model developed by the U.S. Food and Drug Administration. The audience for FoodRisk.org is comprised of scientist from government, industry, and academia, as well as consumers and members of the media. Users can search the contents of the FoodRisk.org database using either simple or advance search mechanisms that retrieve records using free-text keywords contained anywhere in the record. Approximately 6,500 individuals visit FoodRisk.org each month, and about 44% of those are located within the U.S., the remaining users are spread over 100 other countries.

P.93 Application of XML to improving information access for food safety risk assessment: AllerML - An ontology and markup language for allergens . Gendel SM*, Ivanciuc O, Power TD, Schein CH, Braun W; FDA/CFSAN and Univ. of Texas Medical Branch steven.gendel@fda.hhs.gov

Abstract: Food safety risk assessment requires the integration of numerous types of data from multiple data sources. Using data from different sources can be difficult because the terminology used to describe the content of each database varies widely. This difficulty can be overcome by developing and deploying subject-specific ontologies and Extensible Markup Language (XML) schemas. These tools permit clear and accurate descriptions of the information contained within a database and they facilitate the automated retrieval of data without requiring detailed knowledge of how each database is structured and implemented. As an example of how this approach can be used, we have developed AllerML, an ontology and markup language for protein allergens. AllerML is based on the International Union of Immunological Societies (IUIS) nomenclature (http://www.allergen.org/), the Structural Database of Allergenic Proteins (SDAP) database (http://fermi.utmb.edu/SDAP/ ), information from other sources including other allergen databases, PubMed, and primary protein structural and sequence databases. AllerML uses a controlled set of concepts and relationships to describe protein allergens, including their sequences and structures, the sequences and locations of IgE and T-cell epitopes, and cross-reactivity. AllerML can be used to automate the exchange of information between allergen databases and to support computational and bioinformatics analysis related to allergenicity assessments and allergen hazard characterizations.

P.94 Risk/Benefit Analysis for Food Supplements – The Lessons from Amino Acids. Rogers MD*; International Council on Amino Acid Science michael.rogers@skynet.be

Abstract: Food and dietary supplements are widely used. Recent EU regulatory initiatives aim at controlling health-benefit claims for supplements and fortified foods, and setting upper supplement intake limits. However, practical experience indicates that the major risk to the consumer arises from inadequate controls on quality rather that potential overdose or misinformation. Specific amino acids have been used to improve sport performance, to enhance the quality of protein, improve immune functions, reduce stress, or, in the case of a single amino acid tryptophan, to improve sleeping patterns. During the late 1980s, non-prescription tryptophan preparations made by a single maker were linked to a deadly outbreak of eosinophilia myalgia syndrome (EMS). Consequently, tryptophan was banned in both the US and UK, although food- or pharmaceutical-grade tryptophan had never been associated with symptoms of EMS and the disease was not linked to “overdose” of tryptophan. In 2005, the UK authorities re-approved the amino acid, while imposing strict quality controls. However, specification analyses of some tryptophan supplements found recently on the US market indicated that some of them contain impure, probably feed-derived, tryptophan representing a major risk to the consumer and raising risk management questions. The International Council on Amino Acid Science (ICAAS) was set up to address problems such as upper limits and quality standards. ICAAS sponsored research has indicated that (a) the major amino acids have a high margin of safety and (b) the upper limits cannot be established using conventional risk assessment methods. The reason for this fact results from a well-controlled metabolic homeostasis of amino acids in the human body, since they are not only elements of a normal diet but also make up a substantial part of human tissue. ICAAS is currently devoting attention to establishing globally applicable specification standards on amino acids used in the human food supply.

P.95 Identification information needs for risk management about methyl mercury. Yamaguchi H*; Kyoto Univercity h.yamaguchi@kw4.ecs.kyoto-u.ac.jp

Abstract: Currently, most of national agencies to manage toxic chemicals assess risks about methyl mercury in fish, and establish guidelines to communicate the level of safe maximum exposure to consumers. However, reducing the fish consumption to avoid the methyl mercury exposure might damage nutritional benefit of public health, as fish is the main source of high quality protein and omega-3 fatty acids. Therefore, the appropriate risk management of the methyl mercury requires to assess not only human risks but also nutritional benefits in fish consumption and to communicate effectively them depending on concerns of consumers. The purpose of this poster is to present the framework to identify information needs for methyl mercury. Studies have been systematically derived some existing scientific knowledge and stakeholder concerns, and identified more requested scientific knowledge on risk assessment. The method used to derive these knowledge or concerns is the fuzzy cognitive map. This method has been used for the policy analysis and the decision support system. Furthermore, it focuses on three uncertainty factors to extract the more requested information for the risk assessment; i) lack of knowledge, ii) indeterminacy of causal relation and iii) disagreement between stakeholders.

P.97 Renewable Fuels: Public and Environmental Health Model. Biksey TM*; WSP Environment & Energy and University of Pittsburgh thomas.biksey@wspgroup.com

Abstract: The proposed strategy by the U.S. Environmental Protection Agency (EPA) to increase the supply of renewable fuels, as mandated by the Energy Independence and Security Act (EISA) of 2007, will result in the production of approximately 36 billion gallons by 2022. The EPA strategy notes that increasing renewable fuels will reduce dependence on foreign oil, and will reduce greenhouse gas emissions. Cellulosic biofuels is one of the four categories of renewable fuels proposed by EPA; the other three categories include biomass-based diesel, advanced biofuels, and total renewable fuel. Life cycle assessments are a useful tool to address the potential environmental impacts from the increase production of renewable fuels. Greenhouse gases and fuel usage currently dominate life cycle assessment studies today, but inclusion of other public and environmental health impacts must be added to the suite of life cycle assessment studies if there is the expected increase in global biobased fuel agricultureal systems. Potential irreversible environmental impacts from excess nitrogen and phosphorus loads from agricultural systems, issues of changes of soil fertility, land use, erosion, and proportional increase in usage of pesticides and genetically modified crops are estimated to result in potentially unknown environmental burdens as tradeoffs associated with increased agricultural production of biobased feedstocks. This study proposes a public and environmental health model to evaluate the potential impacts from the increase production and processing of biobased renewable fuels. Potential impacts explored include consumption of water resources, land use management, releases to groundwater, application of fertilizers and pesticides, and erosion and sedimentation. The results of the public and environmental health model are used to formulate recommendations that can inform public policy regarding the implementation of EISA.

P.98 Food Risks: a case study about the food control system in Brazil . Cassiano AC*; Santa Catarina Federal University carolcass@hotmail.com

Abstract: The aim of this study is to analyze how the Brazilian food control system faced a case of food risk: the scandal of the processed milk contaminated by caustic soda and hydrogen peroxide, which came to public on the end of 2007. The case study was done by a methodology based on the sociology of the environment and social-constructivists risk analysis theories. To investigate the trajectory of the risk, we considered the identification of the actors involved, the conflicts and the negotiations between them. Through the study of those topics, we were able to examine how the food safety regulation is put in practice in Brazil. This is an absent subject on national academic literature about food safety. The type of risk studied is different from other non intentional food risks as it was a premeditated one, based on “bad faith” practices of milk processors in order to obtain more profits and/or to hide the bad quality of the product. A specificity of this event is that it was discovered and investigated by the Federal Police, due to an inefficiency of the food control system to detect it. As a consequence, the national system of food control created a new regulation to be able to avoid this kind of risk. But as risks like the one studied receive a punctual attention in the media and then fall back to an invisible position (although they do not disappear), we put in question how the new regulation can be effective.

P.99 Rocket science for the rocket scientist? An experiemtnal study of the effects of expert-layperson communicaiton on knowledge, attitudes, and levels of fear regarding nuclear energy. Evans SA*; University of Maryland sevans1@umd.edu

Abstract: This study integrated risk communication and cognitive psychology literature in an effort to contribute to the body of research on communicating to laypeople about risky topics that are particularly technical and scientific in nature. A pre-test, exposure, post-test experimental design was employed to determine the effects of message characteristics on learning about a scientific topic. Specifically, this study examined the role of source (scientist, public relations professional, science writer) and jargon (low jargon, high jargon) on learning, attitudes, and levels of fear regarding nuclear energy. Learning, proximity to a nuclear power plant, and involvement were also of interest. This was a two state study with half of the participants recruited from Spotsylvania, Virginia and half from Winston-Salem, North Carolina. A main effect of jargon on learning was found as well as main effects of source on attitudes and trust. Further, location significantly affected involvement, and there were significant effects of learning on attitudes and of involvement and prior knowledge on learning. The practical implications of these results and the potential for further research are also considered.

P.100 Air Pollution: Strong Social Distrust of Authorities. Gutierrez V.V.*; Universidad Diego Portales and Pontificia Universidad Catolica de Chile virna.gutierrez@udp.cl

Abstract: Air pollution in Santiago became a cause of concern in the mid-'80s, when concentrations of suspended particulate matter, sulfur dioxide, ozone, carbon monoxide and nitrogen dioxide far exceeded international environmental quality guidelines. After more than 20 years of environmental regulations and control, suspended particulate matter (PM10) still exceeds the maximum permitted concentrations in some periods of the year. Conditions are particularly severe in winter when, in the unfavorable geography of central Chile, pollutants are trapped in the Santiago basin during spells of temperature inversions. The situation is particularly problematic for the authorities involved; they are confronted with negative public attitudes and media coverage, which intensifies strong social distrust. In this study we show how much trust people place in the authorities in charge of regulating Atmospheric Pollutions. We also described how the level of compensation, people claim for the potential impacts of this hazard, influences the relations of perceived risk, public acceptability and social trust. In 2008, a survey was conducted in the city of Santiago, Chile and 520 subjects answered the survey. Data was analyzed using structural equation modeling procedures. The results show that the levels of trust participants had in authorities in charge of dealing with Atmospheric pollution is very low. The Acceptability of air pollution determines the extent of trust in regulatory institutions, as well as the public's perception of risk and level of compensation. Implications for decision makers and public policies are discussed.

P.101 ASSARIS: An information sharing system for disaster risk communication. Maeda Y*, Nakano T, Matsuda K; Shizuoka University, Ufit Co. Ltd., Ube Information Systems, Inc. tymaeda1@ipc.shizuoka.ac.jp

Abstract: For supporting risk communication with regard to a disaster, ASSARIS, Associative Search System Across Risk Information Sites was developed. The system uses blog articles about Tokai Earthquake, which is expected in the central Japan in near future in high probability, as input database. By using this system, relationships between articles across web sites are visualized as network diagrams, in which articles are shown as nodes, while similar articles are connected by edges. GETA, the Generic Engine for Transposable Association, is utilized for calculation of similarity between the articles. TouchGraph Java applet is used for rendering the article network.

P.102 Risk Communication for Microbial Risk Assessment. Marcum T*, Julias C, Luke N; Camp Dresser & McKee Inc., Denver CO and Edison, NJ luken@cdm.com

Abstract: After the 9/11 terrorist attacks, per the Public Health Security and Bioterrorism Act of 2002 and Homeland Security Presidential Directives 7, 9, and 10, EPA is charged to protect our nation’s critical water infrastructure and to monitor chemical, biological, and radiological terrorism threats to public health and the environment. Consequently, EPA is working closely with other agencies, e.g., DoD, USDA, and FDA, to develop a microbial risk assessment (MRA) guideline to estimate plausible levels of risk that might result from exposure to environmental pollutants, including microbes and pathogens. Prior to developing guidance, EPA’s Office of Water has established clear definitions of terms used for MRA. These terms enable risk assessors, managers, and communicators to better communicate with each other and avoid misunderstandings. This also helps to increase the transparency, contributes to a common understanding of the MRA process, and provides a better presentation of MRA results. Recently, many researches and different quantitative methods have been published. Similar to all models, there are advantages and limitations to the published methods. As a result, even though there have been numerous published documents on quantitative methods for MRA, there has been no specific regulatory guidance to follow in order to perform quantitative MRA. As of today, quantitative MRA guidance has not yet been published. It appears that a quantitative MRA approach will be adopted to frame an assessment to provide direct measurement of pathogens in developing acceptance and rejection guidelines for water sources of microbial exposure to human populations. Even though there is no specific regulatory guidance on quantitative MRA, the published documents can serve as a basis for risk management decisions. As a result, prudent advice to reduce and prevent exposure can be communicated to the public.

P.103 Tall ladders, hot flames: Exploring the culture of occupational safety trainings of volunteer firefighters. Rickard LN*, Brown HC; Cornell University lnr3@cornell.edu

Abstract: From teaching members appropriate protocol to informing them about federal law, occupational trainings play a central role in any workplace. For occupations involving risk, such as firefighting, however, the role of trainings becomes even more critical. Research in the occupational health and safety literature has documented the extent to which such trainings can increase worker awareness of potential risks, as well as decrease workplace accidents. Whereas most firefighters face the potential of high risk to their safety (and the safety of others) in the course of their daily work, the frequency with which they encounter these risks is often less substantial; how, and by what specific methods, might trainings allow for the “simulation” of such risky events to occur? Beyond investigating trainings at a micro-level, we also consider their role in the overall functioning the organization. Given the “ritual” involved in “performing” skills alongside one’s colleagues, trainings may contribute not just information, but rather support an “occupational community.” Similarly, trainings may play a part in building and maintaining a particular “safety culture” underlying this occupational community. Using data collected in qualitative in-depth interviews and participant observation, this pilot study examines the role of occupational trainings at a volunteer fire department. More specifically, we argue that trainings, while diverse in nature, effectively utilize storytelling, perpetuate department-specific terminology, and cultivate commonly held occupational risk perceptions. Taken together, these traits contribute to members’ perceived “realism” of the trainings, while simultaneously providing a means of coping with complex and uncertain risk information. Moreover, trainings help to build and perpetuate an occupational community by initiating and socializing new members, constructing and defining an “in-group”, and building the trust essential to department functioning on-scene. Implications for both applied and theoretical risk communication are presented.

P.104 Risk information sufficiency in an emerging information environment. . Scherer C.W.*, Yuan C., Levitan L., Rickard L., LU L.; Cornell University cws4@cornell.edu

Abstract: The risk information seeking and processing (RISP) model attempts to better understand what motivates individuals to attend to information about a particular risk. This model has been successful in increasing understanding of motivation and processing of risk information, and the capacity to understand particular messages. The RISP model proposes that the drive to overcome information insufficiency motivates individuals to process risk-related information systematically. Typically, the model has been applied in conditions of high information availability with specific defined behavioral solutions. In contrast, the present study applies the RISP model to a high uncertainty situation where little information has been available, and solutions to the risk are unclear and compete with economic and cultural factors. The context for the study is agricultural plastic film, which is used in large quantities on many dairy farms. Until recently individual producers have been faced with less than desirable disposal options: (a) Burn the used film in an open fire or boiler on-farm, creating an environmental and health hazard, (b) Haul the waste film to the local landfill, with the associated disposal expense; or (c) Bury or store it on the land, creating a mosquito and vermin haven and an unsightly area. Recently information about plastic film disposal has become available, with recycling a growing possibility. This created a situation in which we could study how individuals seek and process emerging risk information. A mail survey of nearly 4,000 producers examined opinions in five areas consistent with the RISP model: perceived information sufficiency; information gathering capacity; affective response to the risk; perceived hazard characteristics; and individual characteristics. Combined with an earlier qualitative study of producers, this study suggests an increasingly complex picture of information seeking and decision-making about this emerging risk issue.

P.105 Seeing and believing: Exploring public understanding of environmental risk maps. Severtson DJ*, Vatovec CM; University of Wisconsin-Madison djsevert@wisc.edu

Abstract: Although maps are increasingly used to communicate environmental risk information to the public, very little is known about how map information is understood and potentially acted on. MacEachren adopted theories of spatial cognition to posit that cognitive representations derived from maps are a function of what is “seen” and prior knowledge. Visual features may influence discrepancies between the intended and imparted representations derived from a map. Representations encompass knowledge, beliefs, emotions, and behavioral intentions; beliefs are a stronger predictor of protective health behavior than knowledge. The enumeration unit used to display risk data may influence representations. For example, a map can depict risk data as points or as statistical information smoothed across a geographic area (a choropleth map). When data is sparse, hatching is recommended for conveying uncertainty on choropleth maps. For point maps, uncertainty is indirectly communicated by the density of points across an area. The purpose of this study was to apply MacEachren’s framework to assess what people “see” and understand from risk information designed to vary by format (map, alphanumeric table), enumeration unit (point, choropleth), data uncertainty (high, low), and risk (less, more). Risk information depicted water test results for private wells. Three sequential rounds of cognitive testing interviews among 13 private well users were used to collect data. Choropleth maps conveyed a stronger risk message than point maps or tables, likely due to visual salience. A viewer’s perceived proximity to mapped hazards relative to their perceived residential location influenced risk beliefs. Perceived hazard proximity effects were most evident for point maps and least for tables. These and other findings suggest that visual features and format shape knowledge and beliefs, beliefs derived from maps can be accurate when derived knowledge is inaccurate, and prior knowledge influenced what was seen and understood.

P.106 Caffeine Intakes from Beverages by Pre-adolescent and Adolescent Children. Storey M*, Coletta F, Anderson P; American Beverage Association, Consultant, Coletta Consulting mstorey@ameribev.org

Abstract: Little is known about current caffeine intake among pre-adolescents and adolescents. In this analysis, we examined caffeine intake from beverages by children 9-18 years of age – a timeframe when children begin to make their own independent food choices. Using data from the National Health and Nutrition Examination Survey (NHANES) 2003-2006, we used appropriate survey weights to calculate average caffeine consumption from beverages for males and females aged 9-13 and 14-18 years. Appropriate survey weights were also used to calculate caffeine intake from beverages per kilogram of body weight for each age/gender group. The data show that average caffeine intake from beverages is approximately 24 and 55 mg/day for females 9-13 and 14-18 years of age, respectively; for males in these same age groups, average caffeine intake is 30 mg/day and 74 mg/day, respectively. Caffeine intakes at the 90th percentile did not exceed 88 mg or 2.5 mg/kg/day among 9-13-year olds; for older adolescents, their intakes did not exceed 180 mg/day. These findings suggest caffeine intake from beverages by 9-18-year-old children is modest and is compatible with a total dietary intake of 250 – 300 mg/day, a level considered by health professional organizations to pose no health risk to healthy populations.* *Health Canada. Caffeine: It’s Your Health – New Recommendations. http://www.hc-sc.gc.ca/hl-vs/iyh-vsv/food-aliment/caffeine-eng.php#he. Accessed May 2009.

P.109 Value orientations and sustainable outdoor recreation and tourism. Winter PL*, Bricker KS, Schultz JR; US Forest Service; University of Utah pwinter@fs.fed.us

Abstract: Environmental problems remain among the most pressing social issues of our time (Zelezny and Schultz 2000); and climate change is expected to have myriad impacts on outdoor recreation resources (Morris and Walls 2009). Applications of socio-behavioral solutions to ameliorate environmental problems continue to be a promising area of psychological inquiry. Concern for environmental issues is typically high, however the reasons for that concern vary (Schultz 2002). Reasons for concern have been organized around sets of values described as biospheric (e.g., plants and animals), altruistic (other people excluding oneself), and egoistic (e.g. one’s own health or lifestyle). It has been argued that these value bases can expand our framing of messages focused on environmental responsibility, improving the fit between what matters to individuals and environmental appeals. Findings from a survey of recreation managers in a land management agency are reported in order to explore sources of concern and key management issues surrounding sustainability. Respondents (n=433) participated in an online survey regarding ‘sustainable recreation and tourism’. Perceived attributes of sustainable recreation and tourism were explored considering social, cultural, economic, and environmental impacts following the UNWTO (1996) definition of sustainability. Altruistic and biospheric values were more frequent sources of environmental concern than egoistic values. Biospheric and altruistic values were significantly correlated with perceived professional responsibility to practice sustainable operations, and support for additional agency investment towards sustainability. However, all three value orientations were significantly correlated with concern about climate change. In addition, all three were associated with greater support for agency adoption of actions that would reduce climate-changing gases and for anticipating the effects of climate change. The importance of pursuing overarching values as part of organizational efforts towards sustainability is discussed in light of our results.

P.110 Application of Mental Model in Climate Change-Induced Disaster Risk: Developing an Equitable Comparative Methodology. Dhar Chowdhury P, Haque CE*; University of Manitoba parnali@hotmail.com

Abstract: The normative mental model prescribes to determine the knowledge gaps between “experts” and so-called “laypersons” by mapping beliefs of specialists via creating “influence diagram” and then comparing the characteristics of laypersons’ mental model in relation to the specialists’ mental model. We argue that underlying this approach is the placement of specialists’ knowledge, information and belief systems as socially superior and desired, and of a subjugation of local and general publics’ knowledge and belief system to specialist’s system. By challenging the normative mental model approach, we adopted an equitable comparative approach to evaluate and extend a common methodology to analyze the different sets of knowledge and beliefs held about climate-induced environmental hazards. The present study mapped existing knowledge structures about the risks associated with climate change-induced environmental extremes such as, floods, droughts and heat waves among the residents of Manitoba, Canada. “Influence diagrams” were formulated to capture status of knowledge and beliefs held by specialists; “general public models” were developed from data elicited from face-to-face interviews and a questionnaire survey of the community residents, elderly and low-income families. The juxtaposition of specialists and local community residents’ understandings of climate change-induced disaster risks enabled us to determine knowledge gaps and misunderstandings and to reinforce appropriate sets of safety and mitigation beliefs and behavior pertinent to disaster risk communications. By designing safety, mitigation and prevention information to more relevant and effective to the socio-ecological context of the community residents, we assert that the public may gain improved knowledge about climate change-induced disaster risks, such that better emergency and disaster risk management, self and neighbor protection, and informed decision making develop over time.

P.112 Integrated Approach to Risk Management and Risk Communication: Cement Manufacturing Case Study . Palma-Oliviera J*, Abreu C; Faculdade de Psicologia e de Ciências da Educação Alameda da Universidade palma@veritas.pt

Abstract: Industrial growth required to sustain economic development can result in environmental contamination that nowadays can be significantly reduced by implementing state-of-the-practice controls. Still, facilities engender elevated risk perception and negative attitudes within the public, resulting in ineffective risk management among stakeholders. An integrated risk management and risk communication approach is necessary to sustain industrial operations, minimize environmental risks and increase public acceptability of an industrial presence in communities, especially in environmentally-sensitive areas and in industries treat industrial hazardous waste. We will present an integrated approach implemented at the SECIL-Qutão cement kiln in Portugal. The facility is near a populated area but also borders (in fact is within) a Natural Park that sustains unique plant communities and was recently placed under consideration for the UNESCO Heritage site. SECIL has implemented unique environmental management programs that include extended recovery of the quarries, biodiversity monitoring, use of two high-end particle control devices (an electrostatic precipitator followed by a baghouse), and on-line measurements. A human health and ecological risk assessment, conducted with extremely conservative exposure scenarios, indicates de minimis threats. As confirmation, the company maintains a series of air quality monitoring stations, and soil and lichens are also monitored for dioxin and heavy metals at strategic locations to check for evidence of facility impacts. Risk communication strategy includes an Environmental Stakeholders Committee and an annual “open doors” week. Local stakeholders and regulators not only have allowed continued facility operation, but are also considering permitted burning of hazardous waste in the kiln. Participation and support of the Committee has led to substantial diminution of perceived risk in the nearby community.

P.113 Development of training program for better understanding climate change risk: Applying focus group interview method. Aoyagi-Usui M*; National Institute for Environmental Studies, Japan aoyagi@nies.go.jp

Abstract: We have been investigating the public understanding of climate change, and we found that there is large gap between lay public and expert on climate change issue. We have been using both qualitative and quantitative survey method, and we began to think we need some method to bridge the gap between them. We used the focus group interview method for our qualitative survey, so we added “a series of lectures” that deal with causes, effects, and countermeasures for preventing climate change issues. Our program consists of five parts:1) Building a rapport among group members including their moderator. 2) Watching a short DVD movie that is about the effects of climate change such as melting ice of the Arctic Ocean, agricultural products which are damaged by high-temperature injury, invading tropic insects, then discuss about this movie. 3) The first lecture that deal with mechanisms of climate change, and discussion, Q&A. 4) The second lecture that deal with countermeasures against climate change and future perspectives, and discussion, Q&A. 5)Overall discussion. As focus group interview is usually carried out with six to seven participants with one moderator, participants with one lecturer result very concentrated one. Participants had to listen and think about what lecturer talked. If question came up, she/he can ask it immediately. This situation makes participants feel “involved” with lecture and lead better understanding of lecture. For evaluation of this Focus group interview with lectures, we asked participants to evaluate their satisfaction level before and after the survey. Almost of their satisfaction level went up. So, this focus group interview with a series of lectures has a possibility to be a effective training course for better understanding the climate change issue.

P.114 Communicating the risks of asbestos exposure: a review of workplace and community based programs. Nicol AM*, Hurrell AC; University of British Columbia hurrell@interchange.ubc.ca

Abstract: Exposure to asbestos is clearly linked to serious and deadly respiratory diseases such as mesthelioma, lung cancer, and asbestosis. While the production and use of asbestos has been decreasing since the 1970s, it is still mined in Canada, and asbestos containing products are still imported and used across North America. It is estimated that over 1.5 million workers in North America are currently exposed to asbestos. Although the link between exposure and disease is well known, only a handful of studies have attempted to evaluate how asbestos risks are communicated to people with known exposures. This research systematically assessed all peer-reviewed intevention studies that tried to inform workers or the general public about the health hazards associated with asbestos. Overall there were very few articles that focused communicating asbestos risks, although work in this area started in the 1970s. Most interventions were aimed at workers, including worker notification programs, medical screening programs and worker-led asbestos awareness projects. Worker notification programs were associated with behavior change, including obtaining medical and legal advice and increasing protective equipment use. More studies were found that dealt with the psychological impacts of learning about past asbestos exposure in the workplace. These studies provided mixed results, with some research noting that workers responded favourably, particularly as many of them had been unaware they had been exposed to asbestos. Other studies determined that subjects experienced considerable psychological distress when informed about their exposure history. Only one study aimed at increasing public awareness of asbestos was found. Overall, it is surprising that such a small body of literature exists on how to effectively inform people about the risks of asbestos, given its widespread use historically and the heavy financial and emotional burden associated with asbestos-related disease for patients and their families.

P.115 A Risk-Based Approach to Shuttle Transition Workforce Communications. Krishen L*, Jahn J; Futron Corporation lkrishen@futron.com

Abstract: As a cornerstone program in NASA’s rich history over the past 3 decades, the Space Shuttle Program has become part of the American cultural fabric. With the kick-off of a new vision for Space Exploration, the Shuttle Program is chartered with the delicate task of flying out its manifest safely to complete the construction of the International Space Station while balancing the retirement and transition of its assets to developing programs.One of the top challenges facing a monumental Program nearing the end is a massive and premature loss of critical skilled personnel. To begin tackling risk reduction in the area of mission success supportability, the Shuttle Transition and Retirement team initiated the formation of an effective Workforce Communications Strategy.The challenge presented to the Shuttle Transition and Retirement Communications team was to address and maintain the morale and focus of an aging, diverse, and geographically dispersed workforce under duress. To begin, the team released a survey designed to assess areas of weakness within Space Shuttle Program workforce communications. Data from the survey led to an approach involving information-sharing tools and process improvements which were used to keep track of workforce trends, metrics, and survey findings. From these strategies, a more creative, media-driven side of communications emerged. Some approaches were more successful than others, but overall initial feedback has shown positive trends. Insights and examples will be shared throughout the presentation.By sharing some lessons learned from the approach used by the Shuttle Transition and Retirement Communications Team, this presentation will provide insights in how to initiate and establish a working risk-based communications strategy to help offset premature loss of critical skilled personnel from an organization that is transitioning or retiring.

P.116 Psychological reactance and news coverage of health risks: Examining the moderating role of personality traits on perceived threat and freedom restoration . Fung T*, Hillback E; School of Journalism & Mass Communication, University of Wisconsin-Madison timothy.fung@hotmail.com

Abstract: This study attempts to advance the understanding of the role of psychological reactance (Brehm, 1966) in individuals’ reactions to news coverage of risks by examining how two personality traits (trait reactance and trait uncertainty) may influence the perception of threats to freedom, indicators of state reactance, and intentions to restore threatened freedoms. Research on psychological reactance in health communication has primarily focused on persuasive messages to explain how reactance leads to message rejection but has largely ignored risk messages mediated by daily news. Using West Nile Virus (WNV) as a case study, the purposes of our study are (1) to examine the moderating role of trait reactance and trait uncertainty on the processing of news coverage of health risks, (2) to examine what freedom restoration strategies are likely to be employed as a reaction to news of a health risk. A total of 559 undergraduate students participated in a (Reactance Proneness) x (Uncertainty Proneness) x (Presentation of Probabilistic Uncertainty: Percentage vs. Fraction) x (Framing of Probabilistic Uncertainty: Gain vs. Loss) between-subjects experiment. Preliminary results using Structural Equation Modeling have shown that state reactance to the threat of WNV did mediate the relationship between a perceived threat from WNV and intentions to modify behavior and take precautions to reduce risk. High trait uncertainty individuals who were exposed to loss-framed messages are more likely to perceive WNV as posing a threat to their freedom being outdoors, leading in turn to state reactance arousal. They then are more likely to take precautions to restore their freedom of being outdoors. Implications for the theory of psychological reactance and news coverage of health risks are discussed.

P.117 Risk Fatigue: The Dark Side of Risk Communication. Cummings CL*; North Carolina State University christopherlcummings@hotmail.com

Abstract: By title, risk communicators seek to provide adequate information for the public to make informed decisions about risk— however there is a dark side to Risk Communication. The proliferation of risk messages may inadvertently be putting the public at risk. Risk communication messages are only a fraction of the sum of messages today’s public receives. The competition for viewership spurs media and news purveyors to seize upon sensational and “newsworthy” risk communication messages, who often amplify the risk messages beyond original intentions—creating fear. Over time, risk messages accumulate causing people to feel overwhelmed and produce a reluctance to respond to risk messages in general. It may also cause people to pay heed to the “wrong” risk messages (i.e. parents worry more about a stranger abducting their child than acquaintances and family members who are more likely to abduct children). Risk fatigue has received little mention in “Risk Communication” literature, but this large-scale psychic dismissal caused by overexposure to risk messages is vital to recognize and control. A recent White Paper for the National Nanotechnology Coordination Office begins the conversation articulating the need to further understand Risk Fatigue and identify it as a dangerous problem. In coordination with the Public Communication of Science and Technology (PCOST) project at North Carolina State University, this paper details some of the possible mechanisms that can lead to Risk Fatigue, and hypothesizes some response strategies for the risk communicator in order to evolve future risk communication messages.

P.118 Frame construction and contestation over aided death: How journalists use physicians as sources. Teggatz JL*; University of Wisconsin-Madison jjanscha@wisc.edu

Abstract: Medical advances have enhanced recovery from disease, but also introduced ethical dilemmas about what to do when curative treatments are ineffectual or unwanted. When faced with dying, individuals or caregivers are often forced to weigh the ethical and physical risks and benefits associated with different end-of-life choices, such as do-not-resuscitate orders, living wills, pain medicines, removing nutrition or hydration, and taking prescription medication to end one’s life. The risks and benefits associated with these options include choices between physical pain, moral or religious values, time with and emotional well-being of loved ones, legality, and financial legacy. For example, someone who desires life-sustaining treatment in his living will, may also recognize that treatment could force his family to witness his slow decline and may leave his spouse or family in financial ruin. Alternatively, a patient with terminal cancer may desire death to escape the excruciating course of disease, but may also know her spouse or family is morally conflicted about assisted suicide. The study of frames, organizing principles that influence how a topic is covered in the media and perceived by audiences, are relevant to this issue. Journalistic frames of aided death can impact readers' views of end-of-life care, such as what options are ethically or legally acceptable, and which risks are most relevant in making end-of-life decisions. This study looks at journalists’ use of physician quotes in stories about aided death to explore the process of frame construction and contestation in print media. To do so, context and use of physician quotes in articles from six large newspapers were compared to medical community norms, as represented by common frames in professional medical publications during the same time period. We found that physicians were used as symbols of objectivity and to confirm others’ claims, but rarely addressed end-of-life options or risks.

P.119 Environmental Risk Indicators to Motivate Environmentally Responsible Behavior. Turaga RMR*, Borsuk ME; Dartmouth College rama.mohana.turaga@dartmouth.edu

Abstract: Indicators of environmental risk that simply describe the state of the environment are often criticized for failing to provide actionable information to the general public and decision makers. In this paper, we draw on theories of pro-environmental behavior from environmental psychology to develop hypotheses regarding the type of indicators that are likely to motivate environmentally responsible behavior. The norm-activation models of social psychology propose that pro-environmental behavior (PEB) is a function of a “sense of moral obligation” generated by the activation of personal moral norms. The sense of moral obligation is in turn a function of: (i) the strength of an individual’s altruistic values (specifically, values that reflect a concern toward other humans, the environment, and future generations – which we call “environmental sustainability values”) and (ii) their awareness of consequences (AC) of environmental conditions as well as a personal ascription of responsibility (AR) in alleviating those consequences. The implication is that information that influences AC and AR beliefs among people who hold strong altruistic values is likely to affect PEB. Accordingly, we hypothesize that the most effective indicators are those which: (1) expose adverse consequences to objects closely associated with people’s personal values and (2) highlight individual responsibility for alleviating those adverse consequences. We test our hypotheses in the context of indicators of mercury contamination, using a questionnaire survey of a random sample of respondents in New England.

P.121 Analytical and Statistical Approach for Effective Evaluation of Dioxins and Furans. Julias C*, Kirchner S, Luke N; Camp Dresser & McKee Inc., Edison, NJ luken@cdm.com

Abstract: Dioxin and furan analytical methods were compared for their potential effectiveness to yield useable data for risk analysis. The application of the methods for specific matrices, target compounds, quality control procedures, and reporting limits is evaluated and presented to add better understanding of analytical methods for data end users such as risk assessors. This information will assist in the selection of the appropriate method that can be applied to their study. Particular attention is paid to the method reporting limits and the analytical and sample matrix factors that affect data evaluation and application in risk assessments. Controlling the variables associated with method reporting limits is instrumental in enhancing the effectiveness of the data when applying derived results to the toxicity equivalents (TEQ) approach for dioxin and furan risk evaluation. TEQs for dioxin and furan are calculated by summing the products of concentrations and congener-specific toxicity equivalence factors (TEFs) for 17 individual congeners of dioxins and furans. TEFs are specified in EPA 2008 Framework for Application of the Toxicity Equivalence Methodology for Polychlorinated Dioxins, Furans, and Biphenyls in Ecological Risk Assessment. In some cases, congeners are reported as non-detect. To address non-detect concentrations rigorously, Kaplan-Meier (K-M) method as implemented in ProUCL Version 4.00.04 was used. This non-parametric method does not require assumptions about data distributions or use parameters such as mean or standard deviation. The advantage of the K-M method is that instead of assuming a distance below the reporting limit, one only needs to consider the relative order (rank) of the data. This is particularly important in dioxin and furan data sets since they contain multiple reporting limits for each congener and wide ranges of reporting limit for the 17 congeners that represent each sample. By selecting the most appropriate method and analyzing the non-detect values using K-M method, the dioxin and furan can be evaluated effectively.

P.122 PCB risks to northeastern Manitoba residents from consumption of caribou and fish . Naiman J*, Zeimer M, Rogers J, LaKind JS, Paul p, Ghosh U; The Park School; The Park School; The Park School; LaKind Associates, LLC; University of Maryland, Baltimore County; University of Maryland, Baltimore County lakindassoc@comcast.net

Abstract: Although western countries banned the manufacture and use of PCBs in the 1970s, because of the long half-lives of these compounds, PCBs are still found globally in the ecosystem and in humans. While PCBs were not used to any extent in the Arctic, PCBs have accumulated in the Arctic food chain through long-range transport. To our knowledge, an evaluation of health risks to the residents of the Churchill region from ingestion of PCBs via locally-caught foods has not been conducted. In this study, we will obtain caribou meat and liver from Northern Manitoba and Arctic Char from the Hudson Bay/Churchill River. Muscle and liver tissue from 3 caribou and 3-5 fish will be collected from cooperating local subsistence harvesters. Samples will be frozen at -20°C and shipped on dry ice from Northern Churchill Research Center to the University of Maryland, Baltimore County analytical laboratory. Samples extraction and cleanup will follow EPA methods SW-846 3550Bn, 3665A, and 3660B and modified EPA method 3620C (Florisil © cleanup). PCB congener analysis will be carried out using GC-ECD (Gas Chromatograph - Electron Capture Detector). 91 PCB congeners or congener groups will be identified and quantified. In order to determine frequency of consumption of caribou and Arctic Char, residents of Churchill - including First Nations people - will be interviewed. PCB concentrations in the samples and the food consumption interview data will be used to estimate human health risks associated with dietary exposure to PCBs.

P.125 Longitudinal Study of Exposure to Organophosphate Pesticides Comparing Farmworker and Nonfarmworker Populations. Vigoren EM*, Griffith WC, Coronado GD, Thompson B, Faustman EM; Univeristy of Washington, Fred Hutchinson Cancer Research Center, Seattle, WA gorst@u.washington.edu

Abstract: Organophosphate pesticides (OP) exposure in farmworkers and their children were studied in orchard workers in the Yakima Valley of Washington State with multiple samples in three seasons to understand how exposures varied between populations and across seasons. We studied 100 farmworker and 100 nonfarmworker households in three seasons—thinning season April to July, harvest season August to October and a non-spray season November to February—collecting urine in an adult and child (3-6 years of age) from each household three times in each season. The urine was analyzed for the six diakyl phosphate (DAP) metabolites produced by OPs. We have estimated seasonal trends in urinary DAP metabolites in the farmworkers and non-farmworkers, between and within person variability, and correlations between the adult and child metabolites in a household. The DAP metabolite with the highest concentration was dimethylthiophosphate (DMTP). Each of the dimethyl metabolites (DMP, DMTP, DMDTP) were higher than the diethyl metabolites (DEP,DETP, DEDTP) in the thinning and harvest seasons. The farmworkers were 10 times higher than the nonfarmworkers for DMTP in urine during the thinning season but dropped by 10 fold during the non-spray season so that the DAP metabolite levels were similar to the nonfarmworkers. Also farmworker children DMTP metabolite levels were 2 times higher than nonfarmworker children during the thinning season and these values also dropped by 2 fold during the off season to levels similar to nonfarmworker children. Diethyl metabolites were low in both populations during all three seasons we monitored. We also found that the within person variability during the thinning season was larger than the between person variability for farmworkers and nonfarmworkers adults and children, accounting for over 50% of the total variability. (Supported by grant 2-P01-ES009601 from NIEHS and RD-8370901-0, STAR RD-83273301-0 from US EPA. Contents are responsibility of the authors.)

P.126 Arsenic Bioavailability in Soil: Evidence from In Vitro and Microprobe Studies . LaVelle JM*; Camp Dresser & McKee Inc lavellejm@q.com

Abstract: Bioavailability of arsenic in soil is a continuing issue for appropriate human health risk assessment associated with arsenic exposure. Recent results of in vitro and microprobe analysis of arsenic at a former secondary lead smelter and two former mining/milling sites provide some additional perspective on this issue. In vitro analysis of arsenic bioavailability at a former lead smelter site was entirely consistent with low or zero assessability of arsenic in soils from non-contaminated areas, and increasing content of highly assessable arsenic in contaminated soils closer to the former location of the smelter stack. Results of the in vitro assay between to independent laboratories were consistent, demonstrating the reproducibility of the methods. Estimates for possible remediation goals for arsenic at this site for a given risk target were perhaps an order of magnitude higher than typical default values, when bioavailability information was taken into account. In vitro analysis of arsenic bioavailability at two former mine/mill sites suggested that, regardless of soil concentrations, the amount of available arsenic was small. Results ndicate that potential arsenic-related exposure may be only a fraction of exposure estimated using default assumptions. Electron microprobe analysis of soils indicated that arsenic was observed within the soil matrix and in forms typically non-assessable in the in vitro assay. Iron content of soils at these site was high (25,000 mg/kg or higher), and arsenic was likely associated with relatively insoluble iron species. The current results re-confirm that information on bioavailability of arsenic is critical to risk assessment at sites with arsenic soil contamination. Quantitative use of information on bioavailability of arsenic may provide the risk manager with addition options for remediation and/or mitigation.

P.127 Health risk assessment of personal exposure to heavy medals in drinking water in one typical rural county in China. Duan X*, WANG ZS, Zhang W, Zhang JL; CRAES duanxiaoli0219@yahoo.com.cn

Abstract: Drinking water safey in rural China is among the great priorities in China. Miyang county, which is located in the junction of Yangtze river and Huai river basin, is one typical county which was concerned of its heavy metal pollution emmitted from nearby manufactories. The purpose of this research is to study health risks of personal exposure to heavy metals in drinking water in this county. Drinking water samples from 20 different sites of this county were collected for measuring 14 heavy metals by ICP-MS. About 2500 subjects with different ages and sex were selected to measure exposure factors. Time-activity of drinking water by ingestion and dermal contact of each individual subject during the last three days were kept in dairy in detail by questionnaires. Intake of drinking water from direct and indirect consumption of water and time duration of dermal contact to water in each individual subject were kept in record based on real time measurements.The results showed that cancer risks of people exposure to Arsenic were 9.3E-08 to 1.2E-07 from ingestion and 1.1E-07 to 1.3E-07 from dermal contact. Non-cancer risks of people exposure to Lead, copper and other 12 heavy metals were 2.6 E-07 to 1.7E-06 by ingestion and 1.5E-08 to 5.6E-08 t by dermal contact. The health risks of people exposed to heavy metals in drinking water in drinking water is acceptable. Sensitivity analysis based on Monte-carlo simulation showed that exposure factors (body weight, ingestion rate of drinking water, dermal contact time, surface area) contributes more than 60% of the health risk assessment results. It’s very important to use the accurate/proper exposure factors during health risk assessment.

P.130 Assessment of Dermal Hazards via the New NIOSH Skin Notation Strategy: Lessons Learned. Gadagbui B*, Maier A, Kroner O, Dotson G; TERA, TERA, TERA, CDC/NIOSH/EID gadagbui@tera.org

Abstract: The underlying decision criterion for assigning skin notations to a particular chemical is often not well documented. The National Institute for Occupational Safety and Health (NIOSH) has developed an enhanced strategy that provides the scientific rationale and basis for the assignment of multiple skin notations that address systemic toxicity, direct skin effects, and dermal sensitization. The current work presents experience in applying the NIOSH strategy to the assessment of the dermal hazards of over 80 chemicals and the development of appropriate skin notation assignments. Weight of evidence decisions to address conflicting or limited data sets are highlighted.

P.131 A systematic sensitivity analysis of a screening-level multipathway risk tool. Henning CC*, Shapiro AJ, Turley AT, Burch DF, Smith RL; ICF International chenning@icfi.com

Abstract: A conservative multipathway screening scenario was developed to provide estimates of lifetime cancer risks and hazard quotients to individuals consuming food grown on a hypothetical farm near an emission source. The model includes modules that estimate environmental fate and transport of chemicals emitted to air using U.S. EPA’s TRIM.FaTE model, uptake of the pollutant into a farm food chain, and ingestion of the farm food products by individuals living on the farm and fishing. Separate model scenarios have been designed for benzo[a]pyrene, 2,3,7,8-TCDD, cadmium, divalent mercury, and methyl mercury. To determine which model parameters are most influential and therefore may be of greatest interest with respect to uncertainty and variability in the risk and hazard quotient estimates, a systematic sensitivity analysis was performed by varying each input variable one at a time by a fixed percentage and finding the corresponding percent change in the risk or hazard for each pollutant. Both local and range sensitivity were analyzed by varying each variable up and down by 5% (local) and 50% (range). All user-specified numerical parameters used in farm food chain and ingestion modeling were included as well as TRIM.FaTE parameters known a priori to most strongly affect risk or hazard (to account for long model run time). Elasticity with respect to risk/HQ outputs was used to sort variables in order of decreasing influence for each pollutant. The results illustrate not only the individual variables that are influential on risk/hazard estimates but also the component of the modeling process (fate and transport, farm food chain, or ingestion exposure modeling) in which the influential parameters are used. The local and range analyses were examined separately but tended to give similar rankings and elasticities. These sensitivity analysis results can be used to inform configuration of a Monte Carlo analysis to quantify variability and uncertainty.

P.133 Modeling the spatial and demographic distribution of human exposure and risk from residential radon in the United States. Chahine T*, Subramanian SV, Zartarian V, Xue J, Schultz B, Levy JI; Harvard University tchahine@hsph.harvard.edu

Abstract: Previous studies have developed concentration-based radon zone maps and estimated total lung cancer cases from radon in the United States (U.S.), but do not address variability in radon risk or provide results and methods necessary for conducting community-scale risk assessments. In particular, available models have not incorporated the synergistic effects between radon and tobacco smoke, a key factor considering both are correlated with demographics and housing characteristics. Screening-level risk characterizations that do not require additional data collection are needed for communities across the U.S. This study assesses whether publicly-available information at the individual, community, and state level can characterize variability in residential radon risk. A multi-level regression model for predicting radon concentrations is developed and combined with a comparable model for smoking to determine the distribution of radon risk. This model is based on information from the U.S. Environmental Protection Agency (EPA) National Residential Radon Survey (NRRS) and U.S. Geologic Survey. It predicts radon concentrations as a function of surficial uranium as well as housing characteristics and population demographics, which were evaluated in the NRRS and are available through the U.S. Census and other national databases. Broad-based spatial trends in residential radon concentrations compare well to previous studies. The model also captures interactions that contribute to risk variability within and between communities. For example, many low-income populations have higher smoking rates but are exposed to lower radon concentrations, indicating that radon concentration patterns differ spatially from radon risk patterns. While the predictive power of this radon concentration model is low, given limited national-level data about local soil factors and other covariates, this risk model leverages available data to improve community-scale risk characterizations.

P.134 A Dynamic Data Fusion Based Modelling Tool for Collaborative Computational Toxicology and Health Risk Analysis. Mohapatra AK*; HEALTH CANADA asish_mohapatra@hc-sc.gc.ca

Abstract: Data Fusion (DF) models applicable to toxicological data processing were assessed under a collaborative risk analysis framework. Human health and ecological risk analysis projects use various types of data that are measured, estimated and predicted under various assumptions and scenarios. In this process, patterns, correlations, associations and causal inferences are made to draw conclusions and ultimately decisions to manage and communicate risks. High priority must be given to various dataset layers, their structural and functional relationships as well as inter-relationships arising from the fusion of those datasets. Data collection as well as data analysis process must be refined in order to improve collaborative risk analysis. DF frameworks and methodologies have shown promise and we propose that the modified DF model based on the Joint Director Laboratories (JDL) framework can be effectively used as a proactive and central tool in collaborative computational toxicology and health risk analysis. The objectives are (i) to have a front end or upstream approach towards an effective dynamic DF based computational toxicological analysis; (ii) proactive patterns detection in toxicological, human health and ecological data sets; (iii) facilitation of collaborative and integrated human health and ecosystem based risk analysis; and (iv) prevention of ecological and human health consequences. Application of DF based tools and data mash-up technologies facilitated by a semantic web informatics framework would significantly increase the efficiency of dynamic computational toxicological data integration, ecological health, and human health risk analysis.

P.135 Treatment of uncertainty from QSAR models in risk assessment. Oberg T*; UNIVERSITY OF KALMAR tomas.oberg@hik.se

Abstract: Quantitative structure-activity relationships (QSAR) and other in silico methods are anticipated to be of increasing importance in implementing the new European chemicals regulation REACH (Registration, Evaluation, Authorisation and Restriction of Chemical substances). A set of recommendations for validation of QSAR models were agreed upon by the Organisation for Economic Co-operation and Development (OECD) in 2004. Two of these principles, a defined domain of applicability and appropriate measures of goodness-of–fit, robustness and predictivity, are of particular importance in assessing the uncertainty in QSAR predictions. However, the quantitative representation of model uncertainty in risk assessment is not covered in these principles or in the REACH guidance documents. It is generally agreed that the uncertainty will become unacceptably high outside the applicability domain and this can easily be demonstrated also in practice. Many attempts have also been made to map the uncertainty in predictions within this domain, but these are largely unsuccessful. One of the work packages in the European research project CADASTER (CAse studies on the Development and Application of in-Silico Techniques for Environmental hazard and Risk assessment) will focus on the integration of QSARs with risk assessments. Guidance on using QSAR models for probabilistic risk assessments is a key deliverable for this subtask. In this poster we will discuss some of the problems involved in reporting and propagating prediction uncertainty and possible solutions. We like to invite the meeting participants to interact with us in the process of developing general approaches that can be applied independent of the specific QSAR modelling procedures. The aim is to advance the reliability assessment of the individual model predictions from a weight of evidence approach into quantitative hazard and uncertainty estimates.

P.136 Concepts and Measures of International Business Risk. Alday SS*; The University of Sydney s.seno-alday@econ.usyd.edu.au

Abstract: The field of international business research is relatively young, having been formally established only in 1958. The field, however, has grown in prominence and influence over the past 50 years and is now at the centre of debates surrounding the current global economic crisis. The current crisis and the central role that international business plays in the global economy call for a thorough evaluation of international business risk analysis methods and approaches. This evaluation is important, particularly in light of three major considerations. First, the world economy has exhibited increasing levels of volatility since the post-war period, with economic crises occurring more frequently and at shorter cycles in the past 20 years. Second, extant studies in the international business literature continue to disagree on the ultimate impact of internationalisation on firm performance and risk exposure. Third, in spite of the advent of a more volatile international business environment and the persistent disagreement in the literature on the effects of internationalisation, international business activity (as measured by levels of foreign direct investment and international trade) has continued to grow tremendously. The increasingly volatile international business environment signals the evolving nature of international business risk, which will have important implications on the impact of firm internationalisation on risk exposure and performance. These implications will not only signal a review of the fundamental assumptions of the established theories of international business, but will also dictate the pace at which international business will continue to grow and be managed in the next few years. The current study reviews the evolution of the concepts and measures of risk in the international business literature against the backdrop of an increasingly turbulent global economy, and attempts to make a preliminary assessment of the continued validity and usefulness of extant approaches to international business risk analysis.

P.137 Post audit of risk management measure for decabromodiphenyl ether in Japan. Tokai A*, Yamamoto Y, Watanabe S; Osaka University tokai@see.eng.osaka-u.ac.jp

Abstract: Risk assessment document of decabromodiphenyl ether – one of the representative flame retardant- was made public in Japan, 2008. The major results said that there was not recognized any concerned risk for decabromodiphenyl ether under the employed research framework based on the in-depth full scale of risk assessment. In this document, we conducted exposure analysis and hazard assessment for human health. However, ecological risk assessment was not discussed here. In this paper, we have done a post audit of risk management measures for this chemical from the viewpoint of manufacturer’s voluntary scheme and regulatory scheme. The risk management measures for this chemical applied to product chain were examined employing exposure and risk assessment modeling. The measures examined here were voluntary emission reduction of decabromodiphenyl ethers, substitution of decabromodiphenyl ether and introducing recycling system of Japanese waste electrical and electronic equipment. The cost effectiveness for these measures were examined and compared. Finally, further requirement for risk management of this chemical were discussed.

P.138 Stakeholder involvement in environmental risk analysis: Assessing impacts of genetically modified maize on South African biodiversity. Dana G*, Kapuscinski A.R., Donaldson J.S.; University of Minnesota; Dartmouth College; South African National Biodiversity Institute gvdana@gmail.com

Abstract: Environmental risk analysis (ERA) evaluates threats to species, natural communities and ecosystem processes, and is traditionally conducted by biological scientists with limited stakeholder involvement. Such an approach often excludes different knowledge types and epistemologies necessary for understanding the social, economic, and biological complexities inherent in biodiversity conservation. Our study describes the implementation and results of a highly participatory ERA process. Specifically, we address two key research questions: 1) how does one implement a scientifically rigorous ERA, using formal risk assessment tools, with stakeholder involvement at all steps of the process; and 2) how does the involvement of diverse stakeholders change the process, information, and conclusions of the ERA? We conducted a participatory ERA in two workshops at the South African National Biodiversity Institute (SANBI), to analyze potential impacts of genetically modified maize on biodiversity in South Africa’s Mpumalanga Province. The first workshop contained only four biological scientists, while the second workshop included 22 diverse participants, and we examined the ERA process and results between the two workshops. Our results indicate that the time consuming, delicate process of stakeholder involvement is enhanced by individual, face-to-face meetings to explain the ERA process and their contribution, use of technologically simple tools, and facilitation by a scientifically-literate neutral party. The addition of diverse stakeholders also results in a richer understanding of the complexities of ecological systems, and therefore a broader set of hazards for risk estimation.

P.139 Meta-Analysis of Controlled Nitrogen Dioxide Exposure Studies Assessing Airway Hyper-Responsiveness in Asthmatics. Goodman JE*, Chandalia JK, Thakali S, Seeley M; Gradient, Cambridge, MA jgoodman@gradientcorp.com

Abstract: As part of their recent review of nitrogen dioxide (NO2) for updating the National Ambient Air Quality Standard (NAAQS), US EPA conducted a meta-analysis of airway hyper-responsiveness (AHR) to NO2 in controlled exposure studies based on a prior published meta-analysis. US EPA concluded that short term exposure to 0.1 ppm NO2 can cause small, but statistically significant, increases in response to a nonspecific airway challenge in asthmatics. Because US EPA did not evaluate the magnitude of AHR with NO2 exposure or conduct an exposure-response assessment, its analysis does not provide information regarding the clinical relevance of the observed increases or whether they are likely to be attributable to NO2. We conducted systematic meta-analyses and meta-regressions of 27 available studies with exposures ranging from 0.1 to 0.6 ppm NO2, considering the magnitude of effects at different NO2 exposure concentrations, to assess the association between NO2 and lung function in sensitive individuals. Several effect measures for AHR are statistically significant but of such a small magnitude that they are unlikely to be clinically relevant. These include a change in NO2 vs. air in (1) the provocative dose of a challenge agent necessary to cause a specified change in lung function (meta-EE = -27.04, 95% CI: -35.59, -18.49), (2) the change in FEV1 after an airway challenge (meta-EE = -1.75, 95% CI: -3.29, -0.21), and (3) the fraction of subjects with increased AHR (meta EE = 0.58, 95% CI: 0.52, 0.63). There are no exposure-response associations for any effect estimates based on linear meta-regressions or analyses of effect estimates for exposure groups (0.1 to < 0.2 ppm, 0.2 to < 0.3 ppm, etc.). The results of our analyses indicate that, to the extent the effects observed are associated with NO2 exposure, they are sufficiently small such that they do not provide evidence that NO2 has a significant adverse effect on AHR at concentrations up to 0.6 ppm.

P.140 Methodology of AFSSET Occupational Exposure Limit Committee to derive Exposure limits for non carcinogenic chemicals. El Yamani M*, Brunet D, Modelon H, Afsset OEL Expert Committee V; Franch Agency for Environmental and Occupational Health Safety (AFSSET) mounia.elyamani@afsset.fr

Abstract: Studying dose response, and developing dose response models, is central to determining safe levels for chemicals to which workers are exposed in the workplace. For non carcinogenic chemicals, OEL Committee principle is that there is a relationship between a toxic reaction (the response) and the amount of chemical received (the dose). An important assumption in this relationship is that there is almost always a dose below which no response occurs. Thus the Committee recommends only health based occupational exposure limits which are based solely on scientific data. The VLEPs (valeurs limites d’exposition professionnelle: occupational exposure limits values) for chemical substances recommended by AFSSET committee indicate the maximum concentration of an air contaminant below which a person exposed to that contaminant is regarded as protected from ill-health. In assessing a health-based level of a particular substances in the workplace, the procedure usually starts from an experimental database of animal or (preferably) human data (e.g., from epidemiological studies), giving a no-observed-adverse-effect level (NOAEL) or a lowest-observed-adverse-effect level (LOAEL) to derive an occupational exposure limit at a lower exposure value. An alternative is the benchmark dose method which has some obvious advantages over the NOAEL approach. Rather than focusing on a single test dose, it uses data from all dose levels employed. This method is strongly recommended by the committee. Although the scientific background for default assessment (uncertainty) factors in general remains unsatisfactory, the Committee recommends the inclusion of assessment factors in risk assessment, when justifiable. These factors are applied to take into account such uncertainties as the potentially higher sensitivity of humans to the toxicant as compared with animals, variation in sensitivity among individuals, different exposure conditions, etc. Toluene, dichloromethane, formaldehyde, 2-buthoxyethanol and its acetate will be presented as examples to illustrate the described methodology.

P.141 Exposure of different subpopulations to primary fine particulate matter (PM2.5) air pollution due to traffic and domestic wood combustion in Finland. Tainio M*, Ahtoniemi P, Karvosenoja N, Kupiainen K, Porvari P, Karppinen A, Kangas L, Kukkonen J, Tuomisto JT; National Institute for Health and Welfare (THL), Finland; Systems Research Institute, Polish Academy of Sciences; Finnish Environment Institute; Finnish Meteorological Institute marko.tainio@thl.fi

Abstract: Traffic and domestic wood combustion contributed approximately half of the all primary fine particulate matter (PM2.5) air pollution emissions in Finland in 2000. Both of these emission source categories have low emission height and therefore spatial variation of resulting PM2.5 concentrations is high. This might cause variation in exposure between different subpopulations. To study this variation we estimated different subpopulations exposure to primary PM2.5 air pollution due to traffic and domestic wood combustion in Finland. The exposure was estimated separately for different gender, age and education subpopulations. Primary PM2.5 emission strengths were based on Finnish Regional Emission Scenario (FRES) model. Dispersion of primary PM2.5 was calculated using the Urban Dispersion Modeling system of the Finnish Meteorological Institute (UDM-FMI). Emission, dispersion and exposure were all estimated with 1 km spatial resolution. The resulting population average exposure varied only slightly between gender and age. The variation in exposure between different education subpopulations was higher so that the people with higher education were exposed more to traffic related primary PM2.5 emissions. Education did not have any significant impact on exposure to domestic wood combustion related primary PM2.5 emissions. The higher educated people tend to live more in cities and in the South of Finland, and these might explain the difference in exposure between education subpopulation. This exposure difference might have impact on health effect calculations since education also modifies the background health status of population so that people with higher education have less background illnesses than average population.

P.142 Key Aspects of U.S. EPA’s External Review Draft Toxicological Review of Trichloroethylene. Chiu WA*, Bale AS, Brown RC, Caldwell JC, Cooper GS, Fox J, Jinot J, Keshava N, Makris S, Scott CS, Barone, Jr. S, Evans MV, Guyton KZ, Gwinn M, Lipscomb J, Schaum J; U.S. Environmental Protection Agency chiu.weihsueh@epa.gov

Abstract: Trichloroethylene (TCE) is a designated Hazardous Air Pollutant, a common groundwater contaminant, and present in at least 1,500 hazardous waste sites. The U.S. Environmental Protection Agency (EPA) recently released a draft Toxicological Review of TCE for public comment and external peer review. This assessment comprises a comprehensive, up-to-date review of over 3000 studies of TCE and its metabolites, and incorporates advice from previous reviews and consultations from the U.S. EPA Science Advisory Board, the National Research Council, and public comments. Included in the assessment are (1) an overview of sources of exposure to TCE; (2) reviews of the data on the toxicokinetics of TCE and its metabolites; (3) development of an updated PBPK model of TCE and metabolites; (4) characterization of the hazard posed by TCE exposure, including carcinogenicity and non-cancer health effects, based on the available scientific evidence; (5) reviews and analyses of the role of metabolites/metabolism in TCE toxicity and the hypothesized modes of action of TCE health effects; and (6) a quantitative risk assessment for TCE non-cancer and cancer effects. This presentation summarizes key aspects of the assessment, particularly focusing on the use of quantitative methods such as meta-analysis, physiologically based pharmacokinetic (PBPK) modeling, and probabilistic analyses to support EPA’s key draft conclusions and findings for both hazard and dose-response assessment.

P.143 Salts of the (rare) earth: Assessing oral toxicity of the lanthanides. Weinrich AJ*; U.S. EPA National Center for Environmental Assessment, Cincinnati weinrich.alan@epa.gov

Abstract: The term “lanthanides” refers to 15 metal elements with atomic numbers 57 through 71. Also known by the IUPAC name, lanthanoids, they are the major subgroup of elements commonly referred to as the \\"rare earths\\". This common name is a misnomer in that the lanthanides are abundant in nature and have prevalent commercial and some military applications, including carbon-arc lamps for movie projection, television picture tubes, permanent magnets, organic reagents, lasers, sunglass lenses, alloys, and contrast agents for medical imaging. Many early studies concluded very low or no oral toxicity of stable (nonradioactive) lanthanides, based on repeated and single dose oral toxicity data for the chloride and oxide salts. However, single-dose lethality studies using nitrate salts demonstrated oral LD50s at doses close to the subchronic daily doses considered NOAELs for several lanthanide chlorides. For salts other than chlorides and oxides of these lanthanides, the only repeated dose oral studies identified in the literature used the nitrate salt of samarium in drinking water. A recent subchronic study in rats demonstrated increased pancreas and lung weights, and increased liver malondialdehyde concentrations, indicative of oxidative stress, at samarium nitrate oral doses more than 2000 times lower than doses considered NOAELs in earlier studies with oral samarium chloride. Two other recent studies reported reproductive effects in male mice at oral samarium nitrate doses nearly as low. Available data suggest the nitrate anion, per se, does not cause the greater oral toxicity. These observations and the very limited absorption data indicate the lanthanide metals might not be of inherently low oral toxicity, but rather suggest that the low toxicity observed in studies of the chlorides and oxides resulted from poor gastrointestinal absorption.

P.144 Relationship of hurricane hazard proximity to risk perception and preparedness. Howe PD*; The Pennsylvania State University peter.d.howe@gmail.com

Abstract: Proximity is an important determinant of preparedness to natural hazards; it may influence risk perceptions by affecting the salience, immediacy, and availability of the risk. Several previous studies have assessed the effect of broad-scale spatial proximity to point-source anthropogenic and natural hazards such as terrorism targets, nuclear waste, earthquakes, and volcanoes. Using survey data of business owners in a coastal Florida county, this study explores the relationship of hazard proximity to risk perceptions and preparedness at a fine spatial scale and for a non-point-source hazard: hurricane storm surge flooding. Risk perceptions were found to relate positively with proximity to storm surge hazard zones, but an index of hurricane preparedness activities had an unexpected inverse relationship with exposure to flooding. With climate change projected to increase the magnitude and probability of coastal flooding, these results suggest that risk communication messages should provide place-specific information about flood risks and options for preparedness.

P.145 Determining the relative criticality of diverse exploration risks in NASA’s Human Research Program. Kundrot CE*, Edwards JM, Anton WQ, Robotham K; NASA craig.e.kundrot@nasa.gov

Abstract: The mission of NASA’s Human Research Program (HRP) is to understand and reduce the risk to crew health and performance in exploration missions. The HRP addresses 27 specific risks, primarily in the context of Continuous Risk Management. Each risk is evaluated in terms of two missions (a six month stay on the Moon and a thirty month round trip to Mars) and three types of consequences (in-mission crew health, in-mission performance, and post-mission crew health). The lack of a common metric between the three consequence scales, such as financial costs or quality adjusted life years lost, makes it difficult to compare the relative criticality of the risks. We are, therefore, exploring the use of a ternary scale of criticality based on the common metric of influencing an operational decision. The three levels correspond to the level of concern the risk generates for a “go/no-go” decision to launch a mission: 1) no-go; 2) go with significant reservations; 3) go. The criticality of each of the 27 risks is scored for the three types of consequence in both types of mission. The scores are combined to produce an overall criticality rating for each risk. The overall criticality rating can then be used to guide the prioritization of resources to affect the greatest amount of risk reduction.

P.146 Helping experts help themselves: RiskWiki - The Full-Spectrum Risk Knowledgebase. Jackson L (1), Neale D (1), McGill WL (2)*; (1) Office of National Capital Region Coordination, Federal Emergency Management Agency (FEMA); (2) College of Information Sciences and Technology, The Pennsylvania State University wmcgill@ist.psu.edu

Abstract: FEMA , in collaboration with The Pennsylvania State University, is leading the development of the RiskWiki, a secure online site designed to enhance risk-informed decision making. Most individuals tasked with advising and making decisions related to homeland security or emergency management are well aware of the risks they face. In fact, if you ask seasoned practitioners to list their top ten concerns, the raw, unfiltered response they will offer you in a matter of seconds, or minutes would likely rival that generated by days of deliberate risk analysis by those without first-hand knowledge of a particular jurisdiction or asset. The question, then, is how can we maximally leverage this raw intelligence to better serve those in a decision-making position? The RiskWiki supports federal, tribal, state, and local stakeholders by providing information and guidance aimed at helping them further develop and refine their own knowledge of all-hazard risks. The RiskWiki does not generate answers on its own; rather, it helps stakeholders formulate and answer key risk questions themselves based on information about how hazards evolve in time, what types of countermeasures may reduce risk, and how to apply different structured analysis methodologies to help reason through problems and develop answers to questions that are mindful about what is known, unknown, and assumed. As a wiki, the RiskWiki serves as a collaborative environment to build risk knowledge and is continuously growing and updated with new information provided by both academics and practitioners, such as case studies demonstrating the application of different structured analytic techniques, elaboration on cause and consequence in different contexts, varied perspectives on historical events, and articles on contemporary risk issues. This poster provides a summary overview of the RiskWiki, to include its present state and future direction.

P.147 Diagnosing foodborne illness: A behavioral analysis of test-promotion programs. Kaptan G*, Fischhoff B; Carnegie Mellon University gkaptan@andrew.cmu.edu

Abstract: Knowing the source and the disease-causing agent is key to reducing the prevalence of foodborne illness. It requires forensic work that begins with laboratory tests capable of characterizing pathogens that are mostly identified in stool samples. Currently, only a tiny fraction of cases is ever tested and even those tests often produce results too late to be of much immediate use. US FoodNet studies reveal great underreporting of acute diarrheal illness (ADI), raising the question of how to improve the flow of information needed to guide individuals’ decisions about seeking care. Our study applies behavioral decision research to understanding and improving that information. We analyze the cost-effectiveness of possible ways to increase the flow of stool samples, as well as the effects of those interventions on the information value of those samples. Our analysis creates a supply curve, ranking possible measures by the cost per sample produced from the target population: individuals with ADI. We look, specifically, at two strategies for encouraging individuals with gastrointestinal illnesses that might be ADI to seek medical care and one strategy for encouraging individuals to submit samples. All three programs offer ancillary benefits, in the form of opportunities to provide information to program participants, perhaps helping them to speed their recovery, avoid complications, and prevent future illness. Their public health benefit will depend on health authorities’ ability to act on results from the additional tests. Positive results might allow preventing additional infections (for accidental or malicious sources), as well as speeding the identification and treatment of individuals who are already infected. Negative results might allow relieving public concern about unusual pathogens and exonerating some potential sources. Whether any are realized will depend on health authorities’ resources, regulatory authority, and political backing.

P.148 Human Bocavirus: A risk assessment. Solon IS*, Nair C, Sunger N, Kundu A, Gurian PL; Drexel University iss23@drexel.edu

Abstract: Human Bocavirus (HBoV), named after additional members of the Parvoviridae family bovine parvovirus and canine minute virus, was discovered in 2005 in nasopharyngeal aspirates of young children. A quantitative microbial risk assessment was conducted in order to assess the risk of HBoV due to inhalation exposure in a pre-school. A Markov Chain approach was applied to estimate the number of HBoV plaque forming units (PFUs) that would be emitted into the room by an adult with a respiratory infection and model the resulting human inhalational exposure. Dose-response functions were based on two surrogates with similar viral properties, Adenovirus and Coxsackie virus. Final risk estimates were compared to data from an outbreak among kindergarteners infected with human parvovirus B19 (attack rate = 38%), the only other virus in the Parvoviridae family known to cause human infection . A Bayesian windowing approach was employed in which the upper limit of the window of the acceptable risk estimates was established as the 97.5% upper bound for the outbreak data. The lower limit of the window was set as zero since it is possible that none of the observed cases resulted from inhalation but from other pathways, such as fomite contact followed by ingestion. 10,000 trials were simulated for both surrogate viruses to identify the risk of children exposed to Bocavirus from an adult coughing over the course of several hours were simulated for both surrogate viruses. 98.6% of the risk estimates from the Coxsackie model fell within the 95% CI of the outbreak data, while only 4 of the 10,000 risk estimates for the Adenovirus were below the upper bound for the outbreak data. Results are found to be much more consistent with the Coxsackie virus dose response function than the adenovirus dose response function indicating that Coxsackie represents a better surrogate for future analysis of the risk that Bocavirus presents to young children.

P.149 Quantitative Risk Assessment of pathogenic V. parahaemolyticus in Thai pacific white prawns. Wanasen S*, Ross T, Tamplin M, Chokesajjawatee N, Mahakanchanakul W, Pornaem S, Rachtanapun C, Valyasevi R; University of Tasmania swanasen@utas.edu.au

Abstract: Thailand is one of the world’s biggest producers of farm-raised pacific white prawn and the world’s top exporters. The objective of this project is to quantitatively estimate the risk of developing illness from the consumption of pathogenic Vibrio parahaemolyticus in prawns, and to improve and ensure the quality and safety of the products. The prevalence and level of the total and pathogenic V. parahaemolyticus in prawns were measured from harvest through post-harvest handling, processing and storage to estimate human exposure from consumption of prawns and the subsequence illness. Prawn consumption per capita was estimated from the National Thai food consumption data. The impact of each step, from harvest through consumption, on the risk of developing an illness will be evaluated to determine the potential mitigation measures such as methods and conditions for harvesting and processing steps to ensure the safety of consumers. The existing guidance for the maximum numbers of V. parahaemolyticus per gram allowable in prawns will also be evaluated.

P.150 Risk Analysis at RIsk. Gray G*, Stautner J; Gray, G RIsk Science and Management, LLC; Stautner, J CEO ETSZone ggray@riskscienceandmanagement.com

Abstract: Risk analysis has become an important part of key public health decisions from food safety to environmental protection to homeland security. Policy makers, legislators, journalists and the public look to our field for sound information on ways to assess, manage and communicate risks. At the same time, the field is under attack. Detractors suggest analysis is just a means to delay action. Others suggest that the degree of scientific uncertainty we face for many risks makes analysis useless. Some fear that industry funded research will taint risk assessments while others decry a governmental risk assessment “black box” system with no opportunity for outside scientific interaction and feedback. The field of risk analysis is fundamental to making decisions. So why are we in this situation? We identify key questions including: Are public health professionals, including academics, NGO scientists, consultants, government scientists, private sector scientists and policy makers rigorously applying the tools of risk analysis?; Should not the demand for risk analysis and management expertise among all of these constituents be rapidly increasing, given the challenges we are facing? and; How does regulatory science drive the kinds of science that are done and how that science is interpreted and evaluated? We propose a framework for a broad range of constituents to answer these questions, share their ideas, give examples and propose approaches to bringing focus and attention on the importance of advancing risk science for environmental health decisions. Web technology can provide a place for risk and public health professionals to learn about and discuss the needs of and advances in the field. We recognize that risk professionals want a place for our ideas to be heard in a respected community of peers. We want our ideas to be improved through interaction and debate. We want ideas to have an impact on real-world decisions. We want risk analysis to be part of the solution.

P.151 Using alerts and models to assess mutagenicity. Myatt GJ*, Bower D, Cross K, Quigley D; Leadscope gmyatt@leadscope.com

Abstract: A database of around 9,500 chemical structures and 150,000 genetox tests has been produced by integrating information from numerous public sources including the US Food and Drug Administration’s files. Using the ToxML standard, information is integrated from various sources and approximately 20 endpoints calculated. This database has been analyzed based on structural alerts described in the literature with known mechanisms. The alerts were refined and new alerts identified. Computational models have been developed using the Leadscope software. This poster outlines how this database was put together using the ToxML controlled vocabulary, how the endpoints for modeling were determined, and how models and alerts were developed and can be used to assess mutagenicity.

P.152 The Influence of Numerical Presentation on the Processing of Risk Evidence. Skubisz C*; University of Maryland skubisz@umd.edu

Abstract: People have difficulty understanding their personal risk. In fact, years of research on risk perception has concluded that actual risk and perceived risk rarely match. Therefore, communicating about risk presents a number of challenges. The biggest challenge is due to the fact that risk information is communicated within a context of uncertainty, making evidence a critical part of a risk message. However, there are some ways of presenting evidence that are easier to understand than others. Risk information presented as numbers has many advantages: numbers convey the magnitude of risks and benefits more clearly than words, numerical statements of risk lead to less variation in risk perception, people often prefer to receive numerical information, numerical information is trusted, and in some cases numerical information is better understood than verbal information. Research has compared various numerical evidence types, but this line of work is largely atheoretical. The purpose of this research is to explore when and why quantitative information affects information processing and risk perception. Two theoretical perspectives that can explain the findings in the literature are being tested and compared. The first is the evolutionary perspective that argues that some numerical formats are more accessible in the human mind, making them more effective. The attitude accessibility framework will be applied to test the predictions of this theory. The second theoretical perspective is the affective processing paradigm, represented by cognitive-experiential theory and the affect heuristic. This alternative perspective suggests that some numerical formats produce more affect laden imagery. Thus, some numerical formats elicit more meaning for receivers, making them more effective. In addition, previous research suggests that numeracy, the ability to use and manipulate numbers, may affect processing of quantitative information. The moderating effects of this variable are also being explored.

P.153 Consumer Product Design Safety Assessment Certification Scheme. Rider G, Inman M, Nielson E, Reilly J, Vincenten J, Hakkinen P, Reiss R, Easton R, Chen X*; 1,9 Intertek, U.S.A.; 2 RABQSA International, Inc., U.S.A.; 3 Consultant (previously Health Canada), Canada; 4 DuPont Hospital for Children, U.S.A.; 5 European Child Safety Alliance, Netherlands; 6 National Institutes of Health (NIH), U.S.A.; 7 Exponent, U.S.A./Society for Risk Analysis (SRA); 8 Federation of Australian Scientific & Technological Societies (FASTS), Australia xiao.chen@intertek.com

Abstract: Consumer product safety is vital in our daily lives. Driven by recent recalls, regulatory authorities around the globe are strengthening its requirements and enforcement in order to regain confidence among consumers. Historically, approximately 70% of recalls were the result of design defects. Robust design appraisal as a remedy for the root causes of product recalls has received significant attention. Nonetheless, there has been a lack of governance and standardization for design appraisal. As a result, RABQSA International, a recognized leader in designing, developing, and delivering accredited personnel and training certification services, has moved forward with an initiative to develop a competency-based personnel certification program for Consumer Product Design Safety Assessment (CPDSA). An international multistakeholder steering committee was formed to develop a personnel certification scheme for design appraisers of consumer products through incorporating the best techniques and practices available. Six elements of competency were identified: 1) Understand mandatory and voluntary requirements for products to enter the market place. Has in-depth knowledge of product testing/inspection or other suitable approaches to assure product compliance; 2) Identify appropriate data sources that preferably reveal causation rather than correlation, of incidents/recalls through data analysis; 3) Understand the drive for optimal stimulation for people at different developmental stages in order to predict foreseeable use of products; 4) Understand how engineering/chemical/(micro)biological analysis may assist with hazard identification and characterization; 5) Understand the physical interaction of a consumer and a product, by considering the human anatomy and physiology to identify potential hazards posed by consumer products; 6) Understand the hazard types and severity associated with the product. A skill examination was developed based on these six elements to assess the practical competency of applicants. This initiative paves the way for future development of a certification scheme for consumer product risk analysts.

P.154 Testing the Effectiveness of the New Hazard Category Language for the ATSDR’s Public Health Assessment Products. Allen-Lewis* S, Furphy* L, Telfer* J, Walker** S, Chervin** D, Dantzker** H*, Laurenson** J; *Agency for Toxic Substances and Disease Registry, **ICF International jlaurenson@icfi.com

Abstract: The Agency for Toxic Substances and Disease Registry (ATSDR) provides health information to help prevent harmful exposures to toxic substances. ATSDR conducts public health assessments and consultations by reviewing available information about hazardous substances at a site and evaluating whether exposure to them might cause harm to people. Using these data, sites are classified into one of five categories. Concern about the meaning, use, and intent of the categories had been raised by ATSDR’s programs and partners. ATSDR reexamined the hazard categories to test and refine the messages. The goal was to improve the clarity and understandability of the categories among the lay public. The proposed hazard category language was reviewed by an expert panel of risk communication scientists; stakeholders and partners; and members of the lay public using focus groups and cognitive testing. The categories were revised incorporating the use of plain language and information mapping. The new language reflects clearer explanations of the exposure pathways, time periods, and effects on people’s health and thus should result in fewer unintended consequences and less misunderstanding. Interim guidance recently was released, and ATSDR has begun testing the new language as it is applied for the first time at actual sites to assess understanding of the health messaging among communities receiving the messages and to understand whether ATSDR health assessors and partner groups are adopting the new language and category descriptions.

Society For Risk Analysis Annual Meeting 2009

Risk Analysis: The Evolution of a Science

Session Schedule & Abstracts

Poster Session

Room: 6:00-8:00 pm