http://hdl.handle.net/1957/19679 2013-05-23T05:37:31Z http://hdl.handle.net/1957/35783 West Nile virus : forecasting models for a resurging vector-borne disease in Arizona, U.S.A. Roldan, Josiah Javier West Nile Virus (WNV), a vector-borne disease continues to be a serious threat to public health in the United States, particularly in the Southwest region. While all the states in the U.S. experienced a decreasing trend of WNV disease in 2010, the state of Arizona experienced a sharp increase from 20 in 2009 to 166 cases the following year. This dissertation endeavored to develop forecasting models to predict future cases of disease and identify counties with increased propensity for WNV. Furthermore, this study aimed to identify environmental and economic factors that contributed to the increase in WNV cases in Maricopa County, Arizona. A spatiotemporal stochastic regression model was developed using Bayesian principles and was successful in calculating the annual mean cases of disease from 2003 to 2011 for all counties. The model was also able to predict future cases of disease by fitting historical data. The model-based inference identified counties in the southern region of Arizona as having an elevated propensity for disease compared to counties in the northern region. A Seasonal Autoregressive Integrated Moving Average (SARIMA) model was developed and effectively forecasted monthly cases of human WNV in Maricopa County, Arizona. By fitting the SARIMA model to monthly historical disease data from 2005 to 2011, the temporal model presented a decreasing trend of monthly incidence of disease for 2012. The impact of home foreclosures, climate variability, and population growth on the resurgence of human WNV disease cases in Maricopa County during the 2010 epidemic was investigated. These factors were found to have contributed to the resurgence of the disease by creating the optimal environmental conditions that allowed the amplification of mosquito populations, thus increasing the risk of disease transmission to humans. As spatiotemporal disease data become readily available, forecasting models can be an important and viable risk assessment tool for public health practitioners. Forecasting models allow the mobilization and distribution of limited resources to areas with elevated propensity for disease, and the timely deployment of intervention programs to reduce the overall risk of disease. Graduation date: 2013 2012-12-03T00:00:00Z http://hdl.handle.net/1957/35506 Two epidemiological studies in central Haiti Bonnlander, Heinke P. Two epidemiological studies, designed to expand our knowledge of morbidity and mortality indicators concerning malaria and Bacillus Calmette-Guerin vaccination in the Hospital Albert Schweitzer health district located in Central Haiti's Artibonite Valley, are presented. The hospital serves a rural population of 190,000 in its 610 square mile district. A large proportion of the rural population still relies on traditional healers in times of illness. Consequently, accurate morbidity and mortality data from individuals and communities in the district are scarce. The first study investigated an outbreak of axillary lymphadenitis and abscesses after Bacillus Calmette-Guerin vaccination among rural Haitian children treated at Hospital Albert Schweitzer from January 1986 through March 1991. Seventy-seven cases of vaccine-related complications were identified, all among children immunized before the age of 1 year. The proportions of children with complications were 0.017% for 1986 through 1989, 0.91% for 1990, and 2.2% for January through March 1991. The probable explanation for the increase is a change in the BCG strain or in the reactogenicity of the Pasteur strain. Graduation date: 1994 1994-05-06T00:00:00Z http://hdl.handle.net/1957/35241 Drug and alcohol use by freshman at Siuslaw High School and their opinions regarding potentially effective drug and alcohol education programs Byrd, Marcia J. This study addressed the incidence of drug use by freshmen at Siuslaw High School in Florence, Oregon, and sought their opinions about effective drug and alcohol prevention programs. The purpose was to provide basic data which would ultimately assist in specific drug and alcohol programs for this population. In phase one, 83 students from four of seven freshmen Health Education randomly selected classes were given the Washington State Survey of Adolescent Health Behavior. This survey consisted of 81 questions about student demographics, students' drug use, and their opinions of drug prevention programs. During phase two, the students were asked open-ended questions about their suggestions for more effective drug prevention programs and curriculum. The survey revealed an average amount of drug use and that the students did not think the drug prevention program at Siuslaw was working. The survey also revealed that teenagers are influenced to drink and use drugs by the media, their friends, and by a lack of activities offered by the community. The study conclusions showed that Florence does have teenage drug use and the community should consider a variety of approaches. Data suggest that more youth support groups, more affordable counseling, and more recreational activities may help reduce drug use. The school needs to develop a more comprehensive drug prevention curriculum and implement it at the younger grade levels. The school combined with the community, should offer more drug-free alternatives and target those youths who are not using drugs and promote their positive behavior. Graduation date: 1995 1995-05-02T00:00:00Z http://hdl.handle.net/1957/34943 Factors influencing individuals attitudes toward voluntary active euthanasia and physician assisted suicide Champeau, Donna A. Issues of right to life, as well as death have surfaced as topics of hot debate. In particular, questions about when and if individuals have the right to end their own lives have emerged and gained considerable attention as health policy issues having the potential to affect all Americans.. The purpose of this study was to identify the factors that are most likely to influence an individual's decision to support or not support voluntary active euthanasia (VAE) and physician assisted suicide (PAS) in specific medical situations. This study also examined the differences in medical vignettes by various demographic and attitudinal factors. Data were collected from a sample of classified staff members at two institutions of higher learning in Oregon. A survey was used to collect all data. Paired sample T-tests, stepwise multiple regression analysis and repeated measures multiple analysis of variance (MANOVA) were used to analyze the data. Based on survey results, there were significant differences in attitudes toward PAS and VAE for each medical vignette. Religious beliefs, fear of dependency, and fear of death were the most powerful predictors of individual support for PAS in each medical situation. In the case of VAE, there were differences in support on each medical situation in terms of the most powerful predictors: fear of dependency and religious beliefs for the cancer vignette, fear of dependency, religious beliefs, and age for the ALS vignette, and religious beliefs and fear of dependency for the paralysis vignette. The repeated measures MANOVA revealed that in general, the older the individual was, the less likely they were to support PAS or VAE. However, women over age 66 in this study were more likely to support VAE than were the males age 66 and over. Males in the 51-65 year old category were more supportive of VAE than females in this age category. Also, those who were more fearful of death were more likely to have a higher level of support for VAE. In all three vignettes (Cancer, Amyotrophic lateral sclerosis (ALS), and paralysis) for both PAS and VAE, there was a significantly different level of support measured on a seven point Likert scale. Graduation date: 1995 1994-11-23T00:00:00Z