In 1990, the Clean Air Act (CAA) Amendments authorized the regulation of 188 hazardous air pollutants (HAP). Exposure to HAPs at sufficient concentrations and durations can increase both cancer and serious adverse non-carcinogenic effects. The purpose of this study was to conduct a human health risk assessment using data of 43 HAPs from five monitor sites in Portland, Oregon during July 1999-August 2000. HAP concentrations were compared to carcinogenic and non-carcinogenic (health) benchmark concentrations; and emission sources were determined for HAPs that exceeded health benchmark concentrations. Additionally, cancer risks were determined for subpopulations and compared to cancer risks generated for the general population. Results of this study indicate that 20 HAPs exceeded carcinogenic benchmark concentrations (corresponding to a risk level of 1 x 10������) in at least one location. Chromium compounds posed the highest cancer risk (3.5 x 10������). Seventeen HAPs exceeded carcinogenic benchmark concentrations at all five sites. Seventy-five percent (%) of the total cumulative cancer risk was contributed by chromium compounds, 1,3-butadiene, formaldehyde and 1,1,2,2-tetrachloroethane. Three HAPs, chromium compounds, acrolein, and formaldehyde, exceeded non-carcinogenic hazard ratios of 1.0. Releases from area sources accounted for the largest percentage of HAPs that exceeded health benchmark concentrations. With respect to subpopulations, asthmatics teenagers (age 11-16) and asthmatic adults (age 18-50), had slightly elevated cancer risks of 1.4 x 10������ and 1.2 x 10������. respectively, compared to the general population risk level of 1 x 10������. Results of this study indicate that several HAPs pose a potential human health concern in Portland and that efforts should be made to reduce their emissions. Additional studies are warranted to further assess potential human health risks and the extent of HAPs in Portland, Oregon. / Graduation date: 2003

Despite the fact that fishing is an inherently uncertain business, risk has rarely been formally recognized in fisheries science or management. Few fishery management plans include any form of risk assessment and those that do focus on minimizing risk caused by uncertainty associated with markets and environmental conditions. Fishermen's attitudes towards risk, whether they are risk-neutral, risk-averse, or risk-prone, have rarely been considered. Although fishermen's attitudes towards risk have been shown in theory to have an impact on fish populations, none of the previous investigations precisely identified whether fishermen are risk-neutral, risk-averse, or risk-prone. This research attempted to identify fishermen's attitudes towards risk from an analysis of their decisions about where to fish. The research applied risk-sensitive foraging theory to an analysis of data from the Oregon trawl fishery for 1991. The data were provided by the Oregon Department of Fisheries and Wildlife. One file contained tow-by-tow information for each fishing trip on landings by species, time spent fishing, type of gear, and fishing locations. A corresponding file contained trip-by-trip information on landings and price by species. The two data files were screened for inconsistencies and then classified into small homogeneous categories based on port, fishing gear, fishing area, and boat size. Various variance-discounting models were fitted to each category to determine fishermen's attitudes toward risk. The models describe the expected utility of fishing at a given distance from port as a linear function of the mean, variance, and third moment of the dollar value per hour of the retained catch. The unknown parameters were estimated from the data using logistic regression techniques. The results of the analysis indicated that in two of fifteen categories the fishermen were risk-averse, and in four categories they were risk-neutral. However, for the remaining nine categories the results were inconclusive and in some cases the fishermen's choice of fishing locations appeared illogical. Instead of preferring fishing grounds that generated higher profits, it appeared that fishermen actively avoided such grounds. The inconclusive and sometimes illogical results may have been due to inappropriate assumptions about the data and about the factors motivating fishermen's decisions. Additionally, there might have been some factors that could have affected the analysis which this research overlooked. For example, this research only accounted for monetary rewards, but fishermen may have preferences other than revenues and costs that influence their choice of fishing grounds. / Graduation date: 1995

This study examined lifestyle changes as related to the risk of coronary heart disease (CHD) in Chinese students at Oregon State University (OSU). The study population included male students or scholars from the People's Republic of China who were attending OSU during spring term 1992. Fifty subjects were interviewed using a structured questionnaire. The questionnaire included eight categories of information: (1) bodyweight and blood pressure, (2) diet, (3) alcohol consumption, (4) cigarette smoking, (5) physical activity, (6) psychological stress, (7) acculturation factors, and (8) demographic factors. Results indicated that for this group of Chinese students, bodyweight, consumption of dietary fat, dairy products, soft drinks, and psychological stress had increased significantly during their stay in the US. Meanwhile, the level of physical activity had decreased. These changes, especially if continued, may have the potential to increase their risk of developing CHD. On the other hand, there were no significant changes in blood pressure and alcohol consumption. For cigarette smokers, smoking had decreased. In their responses to the open-ended questions, the reasons given for bodyweight changes included diet, decreased physical activity, and increasing age. Diet changes were attributed to food availability, relative price, and convenience. For decreased cigarette smoking, lack of smoking environment was considered to be the most important factor. Automobile use, limited spare time, and no friend to play with were the reasons for decreased physical activity. Finally, pressure in school, financial difficulty, and worrying about future were considered to be the reasons for increased psychological stress. Multiple regression analysis indicated that the length of US stay and decreased physical activity were significant predictors for bodyweight gain. The length of US stay was also a significant predictor for changes in total dietary fat. Having financial aid from school was associated with decreased physical activity. Living as single was significantly associated with increased psychological stress. This study failed to identify any significant associations between acculturation factors and changes in the CHD risk factors. / Graduation date: 1993

This study aims to explore the possibility of environmental justice as social consensus and an institutional framework to reduce socioeconomic differences in natural disaster vulnerability through a case study of flood risk management in Johnson Creek, Portland, Oregon. First, by analyzing institutions, policies, and currently ongoing flood mitigation projects, this study investigates how federal and local governments are addressing and responding to current flood problems. Second, through flood expert surveys and GIS spatial analysis, this study examines various factors that contribute to communities' susceptibility to flood risks, and whether there exist spatial differences between physically and socioeconomically vulnerable communities within the Johnson Creek area. Lastly, this study conducted comparative analysis of perceptions using Q-methodology to explore the diverse range of meanings and understandings that flood experts and urban practitioners construct in relation to the dilemmas of environmental justice in flood mitigation practice. The findings of this study indicate that institutional blind spots and barriers in natural disaster mitigation policy and planning can be generated by flood experts' and urban practitioners' different understandings of vulnerability, different interpretations of human rights, and different perspectives on the extent of institutional responsibility to assist socioeconomically vulnerable populations.

While many studies focus on the links between multiple risk factors and negative outcomes such as child maltreatment, less is known about the influence of protective factors in the face of risks. The theoretical base of this study was a social ecological model of interactive influences including individual parent, family, and neighborhood level factors to predict outcomes. Protective Factor Index (PFI) and Risk Factor Index (RFI) predictors were developed to explore potential multi-level protective factor buffering effects on key child development and parenting outcomes. Participants were first time mothers enrolled in a randomized controlled study of the Healthy Start/ Healthy Families Oregon (HS/HFO) home visitation program (treatment group) who completed a follow-up phone survey at the child’s 12 month birthday (n = 405). Families were offered HS/HFO services prenatally after meeting risk screening eligibility criteria on the New Baby Questionnaire (NBQ). Program mothers having received at least one home visit (n = 248) were included in the final analyses. Families had an average of 3.1 (SD = 1.2) NBQ risk factors at enrollment and 83% reported having trouble paying for basic needs. Families received an average of 16 home visits in the first 6 months of the program. Thirty-one percent of mothers were aged 19 or younger, 60% were White and Non-Hispanic, 31% were Hispanic, and 9% were another race/ethnicity. Hierarchical regression models with main effects (RFI, PFI, race) and an interaction term (RFI X PFI) were developed to predict eight outcomes. Interaction effects models were not significant. Five RFI main effects were significant: higher RFI scores were associated with greater likelihood of child welfare involvement, greater parenting stress, less favorable scores on child health and well-being, lower parent responsiveness and ii acceptance, and less supportive learning environments. One PFI main effect was significant: higher PFI scores predicted lower parenting stress. A trend level result showed higher PFI scores were associated with less child welfare involvement. Race was significant in two models: White/Non-Hispanic families were more likely to have a home visitor report child welfare involvement and had more frequent parent-child activities compared to other race/ethnicity families. Unpacking the results with separate single risk factor (12 items) and protective factor (10 items) regression models followed. Results showed parent’s prior family history of maltreatment and younger maternal age predicted child welfare involvement (home visitor report), while protection was seen for those with access to housing support. Social support and family functioning protectors were linked to lower parenting stress, while maternal depression showed the opposite finding. Better scores on a child health and well-being measure were seen with higher neighborhood cohesion and greater participation in HS/HFO; in contrast, neighborhood violence and frequent mobility were linked to worse scores. Developmentally supportive home environments were seen for families participating in additional parent support programs, in which the mother had greater knowledge of infant milestones and behavior, and if the family had access to housing supports. Unemployment proved to be associated with less enriched home environments. In summary, there was no support for the cumulative PFI in buffering risk for negative outcomes in this model. The RFI was also a more robust predictor of outcomes compared to the PFI in the main effects models. Overall, study findings provide some evidence for the utility of specific protective factors, as well as cumulative and specific single risk factors, for screening families for effectively targeting services and guiding the conceptual development of program and evaluation formats.