The current study used a probabilistic Quantitative Microbial Risk Assessment (QMRA) framework to describe the change of E. coli O157:H7 concentration in lettuce through a foodborne pathway, to develop a predictive model for risk estimation for E. coli O157:H7 infection associated with lettuce. The model consisted of a series of pathogen-associated events including initial contamination, growth during cooling, cold storage and distribution, disinfection (chlorine, gaseous chlorine dioxide and gamma irradiation), and dose response after consumption. A modified Baranyi growth model was proposed which described the initial physiological state of E. coli O157:H7 as a function of the initial temperature. The modified Baranyi growth model was used to predict E. coli O157:H7 growth under realistic time-temperature profiles, accounting for the time dynamics of temperature fluctuation. The risk assessment model was constructed in an Excel spreadsheet and Monte Carlo uncertainty analysis was simulated using Crystal Ball. The results in the current study showed that temperature control was the key measure for minimizing the risk of E. coli O157:H7 infection associated with lettuce. Disinfecting contaminated lettuce using the hypothetical methods examined in the study had limited effectiveness in risk reduction. Temperature abuse occurring before or after the hypothetical disinfections significantly diminished the disinfection effect and contributed to increased risk. Of all simulated scenarios, the lowest risk was associated with adequate temperature control and irradiation (44 infections per 1000 consumptions [95%: 94 infection per 1,000 consumption; 5%: 5 infections per 1,000 consumption]). The model can be used to explore the public health impact of other potential strategies that can be adopted to minimize the risk of E. coli O157:H7, while taking into account the possible amplification of pathogen through the food chain.

Ingestion exposure estimates typically use a default bioavailability of 100%, thereby assuming that the entirety of an ingested dose is absorbed into systemic circulation. However, the actual bioavailability of ingested contaminants is oftentimes lower than 100%. The research described herein investigates the use of the Simulator of the Human Intestinal Microbial Ecosystem (SHIME) for the calculation of <i>in vitro</i> bioaccessibility (IVBA), a conservative predictor of bioavailability, of mercury (Hg) from traditional northern foods and arsenic (As) from soils. The primary objective of the research described herein is to address data-gaps which have hindered attempts to incorporate IVBA into risk assessment on more than a case-by-case basis. The hypotheses of this thesis are that (1) the bioaccessibility of contaminants is dependent upon concentration due to kinetic limitations on dissolution, (2) gastrointestinal (GI) microbes in the ileum and colon alter contaminant bioaccessibility and/or speciation, (3) the GI microbial effect on bioaccessibility is toxicologically relevant, and (4) metal bioaccessibility is predictable according to dissolution kinetics.<p> Mercury bioaccessibility from country food samples was independent of total Hg concentration (F=0.5726, P=0.578) whereas As bioaccessibility was inversely related to total As concentration for Nova Scotia mine tailings, synthesized ferrihydrite with adsorbed AsV, and synthesized amorphous scorodite (P=2 x 10-10). Isotherm analysis indicated that, at high soil As concentrations, saturation of simulated GI fluids limited As bioaccessibility under gastric conditions whereas kinetic limitations constrained As bioaccessibility under intestinal conditions. Additionally, we demonstrated that GI microbes may affect Hg bioaccessibility, either increasing or decreasing bioaccessibility depending upon the type of food. For example, the bioaccessibility of HgT decreased in the presence of GI microbial activity for caribou kidney, caribou tongue, seal blood, seal brain, seal liver, and walrus flesh. In contrast, HgT bioaccessibility from Arctic char and seal intestine increased in the presence of GI microbial activity. Similarly, colon microbial activity increased (Fishers Protected LSD, P<0.05) As bioaccessibility from synthesized amorphous scorodite (56 110%), Nova Scotia mine tailings (140 300%), an agricultural soil (53%) and an ironstone soil (350%) containing elevated arsenic concentrations. However, under small intestinal conditions, this microbial effect was transient and demonstrated a small effect size. The toxicological relevance of microbial effects upon As bioaccessibility was assessed using a juvenile swine model with co-administration of oral antibiotics (neomycin and metronidazole). This study research indicated that microbial effects on As bioaccessibility are not reflected in the juvenile swine model. For example, the microbial communities present in the pigs proximal colon clustered according to antibiotic treatment (e.g. microbial communities of antibiotic treated pigs differed from non-treated pigs). Despite this, the urinary arsenic excretion (and hence arsenic bioavailability) of antibiotic-treated juvenile swine orally exposed to soil-borne arsenic was equivalent (Holm-Sidak, P=0.930) to the urinary arsenic excretion of juvenile swine not treated with antibiotics. Therefore, in vitro GI models may not need to include a microbially active intestinal stage when measuring As IVBA.<p> Metal bioaccessibility from soils appears predictable according to fundamental chemical properties of the metal-of-concern. Specifically, metal bioaccessibility of 7 of the 13 metals (V, Ni, Zn, Cu, U, Cd, & Ba but not Tl, Pb, As, Se, Cr, and Hg) regulated according to Canadian Council of Ministers of the Environment Soil Quality Guidelines (CCME SQG) were strongly dependent (R2 = 0.7) on water exchange rate constants of metal cations (kH20) indicating that desorption kinetics may serve as the foundation of a predictive model of metal bioaccessibility.

The legitimacy of classifying female offenders in the correctional system has been disputed (especially the application of male-normed risk assessment tools), and yet, there is a need to accurately determine the risk of re-offending and the criminogenic needs of the offender along with general and specific issues (i.e., responsivity) that will encourage successful program delivery. The Level of Service Inventory Ontario Revision (LSI-OR; Andrews, Bonta & Wormith, 1995) is an assessment tool used throughout Ontarios probation services and provincial institutions. Although the first edition of the LSI was based primarily on a male sample, later revisions included norms for female offenders based on samples spanning three continents (Blanchette & Brown, 2006). Although its reliability and predictive validity has been demonstrated across many field settings and offender populations, few studies (e.g., Rettinger, 1998) have addressed the question of predictive validity on a sufficiently large sample of female offenders to convince the skeptics of the LSI-ORs applicability to women (Blanchette & Brown).<p> The current study examined internal consistency, the ability to discriminate recidivists from non-recidivists with t-tests, and the capacity to predict recidivism with correlation and receiver operating characteristic analysis. The sample consisted of 2831 female offenders who were either released from a provincial correctional facility, completed a conditional sentence in the community, or completed a sentence of probation in Ontario during a one year period (2002/2003). Special consideration was given to female offenders from different disposition groups, with different racial backgrounds, with mental health issues and with prior victimization. The LSI-OR had very strong internal consistency and was able to distinguish offenders who committed a re-offence from those who did not commit a re-offence; both when considering the scale as a whole and when considering individual subscales. The LSI OR was also found to predict recidivism for all female offenders. It also predicted recidivism for all subgroups with the exception of female offenders released on a conditional sentence and who had been previously victimized. While the use of the LSI-OR to assess provincial female offenders is supported, however, new risk levels are suggested to increase the predictive ability and reduce the potential for over-classification.

Current literature suggests that the Level of Service Inventory (LSI) and its derivatives (LSI-R, LS/CMI, LSI-OR) are capable of predicting violent recidivism, even though they were not initially designed for this function (Girard & Wormith, 2004; Mills & Kroner, 2006). The purpose of this study was to generate violence prediction scales, based on items or subscales from the LSI-OR, using five different statistical techniques. These analyses were completed on the full construction sample, then the males and the females separately to determine how the scales differ from each other and what, if any, benefits would accrue from utilizing a gender-specific scale.<p> A cohort of 27,027 offenders who were released from custody or entered into community supervision over a one year period was included in the study. There was an average followup time of 4.4 years. In this sample there was a general recidivism rate of 36.0% and a violent recidivism rate of 11.3%. Fifteen violence prediction scales were generated that ranged in predictive validity from r = .139 to r = .214. The scale with the highest predictive validity was the 11 item scale created from the full sample using the item linear regression technique. The scale contained items indicating that history of assault, lack of education and anger management issues were related to violent recidivism. Risk levels were developed for this new scale to classify offenders from very low to very high risk.<p> Although there was little difference in the predictive validity of the generated scales, the stepwise multiple linear regression technique was identified as the most successful method of creating a tool for predicting violent recidivism. There was no increase in predictive validity when using the scale that was developed for just the females in the sample, although fewer items were consistently generated for females than males. Therefore the full sample item linear regression scale is recommended for the prediction of violent recidivism of both male and female offenders in the jurisdiction from which the data were collected. Future research directions may replicate this study in other populations and further analyze the gender differences in violent recidivism.

Previous investigations of the Level of Service Inventory Ontario Revision (LSI-OR) have examined individual subgroups of offenders (e.g., women, Aboriginal offenders), which has made comparisons of its predictive validity between specific offender groups suspect. This study was conducted on a complete cohort of 26,450 offenders who were released from Ontario provincial correctional facilities, sentenced to a conditional sentence, or who began a term of probation in 2004. Participants were followed up for at least four years to collect recidivism information on numerous subgroups of offenders including males (81.7%), females (18.3%), Aboriginal (6.4%), Black (7.3%) and Caucasian offenders (59.2%). Analyses revealed that the LSI-OR scores are positively correlated with recidivism (r = .441, p < .001), and similar correlations were found for all offenders regardless of gender or race, (Aboriginal r = .377, p < .001; Black, r = .420, p < .001; Caucasian, r = .417, p < .001; Male, r = .439, p < .001; Female, r = .426, p < .001). LSI-OR scores are also correlated with severity of the recidivism offence, (r = .098, p <.001) indicating that higher LSI-OR scores are related to higher offence severity for all ethnicities, sentence types, and genders. These findings indicate that the LSI-OR is an effective risk assessment tool for use among different ethnicities, sentence types and genders for provincially sentenced offenders in Ontario.

<p>Nanomaterials are being increasingly produced and used across a myriad of applications while their novel properties are still in the midst of being designed and explored. Thus the full implications of introducing these materials into the environment cannot be understood, yet the need to assess potential risks is already upon us. This work discusses a comprehensive view of environmental impact with respect to material flows from across the value chain into all compartments of the environment, whereby interactions and potential hazardous effects become possible. A subset of this broad system is then chosen for evaluation; a model is derived to describe the fate of nanomaterials released to wastewater. </p><p>This analysis considers the wastewater treatment plant (WWTP) as a complete mixed reactor aerobic secondary clarifier, and predicts whether nanomaterials will associate with effluent or sludge to project potential concentrations in each. The concentration of nanomaterials reaching a WWTP is estimated based on a linear weighting of total production, and the fate of nanomaterials within the WWTP is based on a characteristic inherent to the material, partition coefficient, and on design parameters of the WWTP, such as retention times and suspended solids concentration. </p><p>Due to the uncertainty inherent to this problem, a probabilistic approach is employed. Monte Carlo simulation is used, sampling from probability distributions assigned to each of the input parameters to calculate a distribution for the predicted concentrations in sludge and effluent. Input parameter distributions are estimated from values reported in the literature where possible. Where data do not yet exist, studies are carried out to enable parameter estimation. In particular, nanomaterial production is investigated to provide a basis to estimate the magnitude of potential exposure. Nanomaterial partitioning behavior is also studied in this work, through laboratory experiments for several types of nano-silver. </p><p>The results presented here illustrate the use of nanomaterial inventory data in predicting environmentally relevant concentrations. Estimates of effluent and sludge concentrations for nano-silver with four different types coatings suggest that these surface treatments affect the removal efficiency; the same nanomaterial with different coatings may have different environmental fates. Effluent concentration estimates for C60 and nano-TiO2 suggest that these nanomaterials could already be present at problematic concentrations at current levels of annual production.</p><p>Estimates of environmentally relevant concentrations may aid in interpretation of nanotoxicology studies. These relative estimates are also useful in that they may help inform future decisions regarding where to dedicate resources for future research. Beyond attempting to estimate environmental concentrations of nanomaterials, this type of streamlined model allows the consideration of scenarios, focusing on what happens as various input parameters change. Production quantity and the fraction of this quantity that is released to wastewater are found to greatly influence the model estimates for wastewater effluent concentrations; in the case of wastewater sludge concentrations, the model is sensitive to those parameters in addition to solids retention time.</p> / Dissertation

The avian influenza virus can be transmitted from sick birds to other birds by fecal and oral secretion. Therefore, when birds with avian influenza virus enter domestic bird farms and make contacts, it is possible to carry the virus in and transmit it to domestic birds. Since the outbreak of the avian influenza in Taiwan, most bird farms had set up the screen. Yet, the actual effectiveness of screen coverage rate has not been researched. This study will discuss further regarding the results of different screen coverage rate in preventing wild birds¡¦ entering to poultry farms in Kaohsiung area. Field observations were carried out at the wetlands in Kaohsiung county, including Old Railroad Bridge artificial wetland, Fongshan reservoir and 20 poultry farms which were less than five kilometers away from wetlands from September 2008 to October 2009. Chicken and duck farms were analyzed respectively, and the screen coverage rate was categorized into full, partial and no coverage, respectively. The aims of the study were to investigate the effect of screen coverage on the entrance of wild birds to the farms, and found out the potential bridge species near poultry farms. By observing different levels of screen coverage in poultry farms, migratory birds¡¦ activity, and the contact situation between wild and domestic birds, we can analyze the contact level of high risk species and speculate possible transmission path of the avian influenza, and understand the effectiveness of the screens. Also, showing the correlation between the density of domestic poultry and the number of wild bird enterance. The result showed, in terms of screen with different coverage rates, the number of bird species within the duck farms had significant different in full and no coverage. The number of total birds within the duck and chicken farms with full coverage was significant less than partial coverage. Both in chicken and farms, screen had a great effectively in preventing small sized birds entering, yet not on big sized birds. The density of domestic poultry and the number of wild bird enterance showed negative correlation only in chicken farms with partial coverage. In terms of the wild land birds entering poultry farms, Passeer montanus is the largest species, followed by Hirundo rustica and Streptopelia tranquebarica. As for aquatic birds, it is mainly of Ardeidae, Scolopacidae, Charadriidae and Rallidae. Among all, Passeer montanus, Hirundo rustica, Streptopelia tranquebarica, and Bubulcus ibis are not migratory birds and highly pathogenic avian influenza virus species which move between poultry farms and wetlands. Therefore, they were considered bridge species. Due to the bridge species makes direct contacts with poultry, it might be the spreader of the virus. Decreasing the possibility of wild birds¡¦ contact with poultry is an important mechanism for avian influenza control. Also, based on the model of European Food Safety Authority (ESFA), risk assessments will be conducted on Kaohsiung County wetland¡¦s migratory birds¡¦ highly pathogenic avian influenza virus H5N1 transmission to poultry farms.

Complex chemical mixtures may be released into the environment from a variety of sources including hazardous waste sites. Components of chemical mixtures and their metabolites may be genotoxic leading to cancer and heritable gene mutations. Chemical analysis alone does not always provide the most accurate information from which to estimate the risk of adverse effects associated with exposure to mixtures. Current methods to estimate the human health risk for complex mixtures assume additive effects of the components. Although it is assumed that this approach is protective of human and ecological health, it is also recognized that chemical mixtures may induce a variety of interactions including potentiation, synergism, and antagonism. A combined testing protocol, using chemical analysis coupled with a battery of in vitro, in vivo, and in situ bioassays, provides the most accurate information from which to estimate risk. Such a combined testing protocol provides information to describe the major organic and inorganic constituents, as well as the pharmacokinetics and potential interactions of chemical mixtures. This research was conducted to investigate the potential genotoxic effects of complex chemical mixtures of polycyclic aromatic hydrocarbons (PAHs) and polychlorinated aromatics (PCA) using microbial bioassays (Salmonella/microsome assay and the E. coli prophage induction assay), the 32P-postlabeling assay in mice, and in situ measurements of genotoxicity using flow cytometry. Samples of environmental media and wildlife tissues were collected from four National Priority List Superfund sites within the United States. In general, chemical analysis was not always predictive of mixture toxicity. Although biodegradation reduced the concentration of total and carcinogenic PAHs in soils and groundwater, the genotoxicity of extracts from environmental media did not display a corresponding reduction. Mixtures of polychlorinated biphenyls (PCBs) extracted from sediments were found to inhibit the genotoxicity of PAH mixtures when administered dermally to rodents. This inhibition exhibited a dose-response relationship, with the adduct frequency reduced at increasing doses of sediment extract. Finally, PAH concentrations in environmental media and tissues were found to correlate with DNA damage in wildlife receptors. An integrated approach, combining in vitro and in vivo methods to characterize genotoxicity provides more accurate information from which to estimate uptake and risk associated with exposure to complex mixtures and should be considered in both the human and ecological risk assessment process.

Today the web has become a major conduit for information. As the World Wide Web?s popularity continues to increase, information security on the web has become an increasing concern. Web information security is related to availability, confidentiality, and data integrity. According to the reports from http://www.securityfocus.com in May 2006, operating systems account for 9% vulnerability, web-based software systems account for 61% vulnerability, and other applications account for 30% vulnerability. In this dissertation, I present a security analysis model using the Markov Process Model. Risk analysis is conducted using fuzzy logic method and information entropy theory. In a web-based application system, security risk is most related to the current states in software systems and hardware systems, and independent of web application system states in the past. Therefore, the web-based applications can be approximately modeled by the Markov Process Model. The web-based applications can be conceptually expressed in the discrete states of (web_client_good; web_server_good, web_server_vulnerable, web_server_attacked, web_server_security_failed; database_server_good, database_server_vulnerable, database_server_attacked, database_server_security_failed) as state space in the Markov Chain. The vulnerable behavior and system response in the web-based applications are analyzed in this dissertation. The analyses focus on functional availability-related aspects: the probability of reaching a particular security failed state and the mean time to the security failure of a system. Vulnerability risk index is classified in three levels as an indicator of the level of security (low level, high level, and failed level). An illustrative application example is provided. As the second objective of this dissertation, I propose a security improvement model for the web-based applications using the GeoIP services in the formal methods. In the security improvement model, web access is authenticated in role-based access control using user logins, remote IP addresses, and physical locations as subject credentials to combine with the requested objects and privilege modes. Access control algorithms are developed for subjects, objects, and access privileges. A secure implementation architecture is presented. In summary, the dissertation has developed security analysis and improvement model for the web-based application. Future work will address Markov Process Model validation when security data collection becomes easy. Security improvement model will be evaluated in performance aspect.

Risk-based corrective action (RBCA) is rapidly becoming an accepted approach for the remediation of contaminated sites. Under a RBCA approach, the risks to human health and the environment associated with a contaminated site are evaluated and appropriate corrective measures are taken as needed to reduce risk to acceptable levels. A series of standard guides of RBCA have been developed by American Society for Testing and Materials (ASTM). The major tasks of this study were to (1) perform the sensitiveness analysis to evaluate the effectiveness of each input parameter on the calculated risks, (2) application of Monte Carlo simulation using a statistic software (@RISK) to analyze the distribution probability of inhalation risk, (3) conduct a risk evaluation and risk calculation at a chlorinated-compound contaminated site. Results from the sensitiveness analysis show that the major factors, which play important roles in the risk evaluation including sources of air pollution, vapor transportation rate, pollutant volatilization rate, length and direction of wind, distance of pollutant transport, width of pollution source, and groundwater flow velocity. Results from the Monte Carlo simulation show that the carcinogenic risk is about 15¡Ñ10-6 when the accumulation rate is 95% via inhalation. Moreover, the carcinogenic risk is about 31¡Ñ10-6 when the accumulation rate is 95% via ingestion. The calculated risk levels are higher than the requirement for minimum target risk level (cancer risk of 1x10-6) described in Taiwan¡¦s ¡§Soil and Groundwater Remediation Act¡¨. Results also show that the hazard index of non-carcinogenic risk is about 0.5 via the route of inhalation, which is higher than the minimum target risk level of 1. Moreover, the hazard index of non-carcinogenic risk is about 1.3 via the route of ingestion, which is lower than the acceptable level of 1. Results from the case study show that the major pollutant exposure routes at this chlorinated-compound contaminant site include inhalation of contaminant vapor and groundwater ingestion. Therefore, the input parameters affect the calculated risks include daily intake of drinking water, groundwater infiltration, groundwater flow velocity, aquifer depth, and distance of pollutant transport. Based on the results of risk assessment, it is very important for the decision makers to incorporate remedial activities including institutional controls, engineering controls, and remediation programs from RBCA results. This study provides a streamlined process and guidelines of developing the risk-based decision-making strategy for contaminated sites in Taiwan.