The first chapter of this dissertation examines the effect of insurance mandates on infant immunization rates. Immunizations are one of the greatest public health achievements of the 20th century. While US infant immunization rates have been increasing in the last 20 years, the cost of fully immunizing a child with all recommended vaccines has almost tripled. This is partly due to new additions in the list of recommended vaccines, but also due to the use of new, safer, but more expensive technologies in vaccine production and distribution. In recent years, many states have mandated that recommended childhood vaccines be covered by private health insurance companies. Currently, there are 33 states with such a mandate. In this paper, I examine whether the introduction of mandates on private insurers affected immunization rates. Using state and time variation, I find that mandates increased the immunization rate for three vaccines — the diphtheria-tetanus-pertussis, polio and measles vaccines — by about 1.8 percentage points. I also find evidence that the mandates shifted some vaccinations from public to private sources. The second chapter of this dissertation studies the issue of whether concerns about autism affected vaccine takeup. In the wake of strong claims that there existed a link between autism and the measles-mumps-rubella MMR) vaccine, which was refuted by later research, I examine whether fewer parents immunized their children. This task becomes difficult as the timing of the controversy in the US coincided with expansions in medical access for children and other programs that affect childhood immunizations, as well as another controversy regarding mercury containing preservatives in childhood vaccines. Using a time trends analysis and a few differencing strategies that compare the take up of MMR to other vaccines, I find that the MMR-autism controversy led to a decline of about 2 percentage points in the take up of MMR and a negative spillover on other vaccines. I find some evidence that more educated mothers responded more to the controversy, which is consistent with more educated individuals absorbing health information more quickly. However, this disparity persisted even after new research and information about the lack of such link became widespread in the media. The third chapter of this dissertation analyzes the effectiveness of a peer counseling breastfeeding support program for low income women in Michigan who participate in the Women, Infants and Children WIC) program. Because there was excess demand for services provided by the program, many women who requested to participate were not subsequently contacted by the peer counselors. We compare the breastfeeding outcomes between the two groups and identify the effectiveness of the program based on the differences between the women who requested to participate and were enrolled relative to those who requested participation, but were not contacted due to lack of capacity. Our analysis uses survey data from the program as well as administrative data from Vital Records, Medicaid, and WIC from the state of Michigan. After providing evidence that our key assumption in identifying the effect if program is consistent with the data, we estimate that the program caused the breastfeeding initiation to increase by about 27 percentage points and the mean duration of breastfeeding to increase by more than 3 weeks. The support program we evaluated was very effective at increasing breastfeeding among low income women who participate in WIC, a population that nationally breastfeeds at rates well below the national average and below what is recommended by public health professionals.

In the last decade, governments, businesses, and households the United States have been encouraged to prepare for domestic hazards as part of the U.S. all-hazards preparedness goal. Of these groups, businesses and households have received the greatest outreach, but are still largely underprepared. To help these groups enhance their readiness, response, and recovery, investigators must understand what factors affect their preparedness behavior. This study theorized that preparedness behavior was a function of person and environment factors from the workplace and the employee household that jointly interacted to shape preparedness knowledge at work and action at home. In the literature, there are few, if any, social ecological studies or conceptual models that account for these person-environment factors or for nested environments when studying workplace and household preparedness. To address this absence, this study developed and tested a new social ecological, stage-based, and nested Model of Workplace-Household Preparedness, and two new all-hazards preparedness tools to better align the research with national readiness goals. To assess these models and scales, a cross-sectional, analytic survey design was applied and then supplemented by qualitative key informant interviews. Data were collected in 2009 from convenience samples of private, large, critical infrastructure companies in California and their employees. The final study sample included 16 companies, 18 emergency managers, and 2,571 full-time employees. Results showed that employee workplace and household preparedness was shaped through unique combinations of person and environment factors at different stages of behavior change i.e., hazard appraisal, decision-making, action). The nested interface of the workplace and household environments also directly shaped preparedness behavior, as did positive messages received from ones social network, particularly through work. The all-hazards preparedness scales also proved to be psychometrically-valid measures, and the research provided new benchmarks on the preparedness status of Californias private sector and its workforce at the start of the 21st century. This research adds to the literature empirical support for the role of multiple, nested environments in shaping individual preparedness behavior, and the utility of social ecological and stage-based models in explaining individual behavior change. These findings can be applied to help strengthen workplace and household preparedness.

This dissertation examines the way in which mosquito-borne malaria has been assessed over time, and in light of historical and contemporary settlement and agricultural patterns, specific to California’s Central Valley. The study includes observations in both northern and southern counties of the Central Valley, key informant interviews with epidemiologists, entomologists, local rice growers, and mosquito abatement personnel, and both archival and field research. In attempting to document evidence of malaria morbidity and mortality between 1830-2010, the study fused: historical landscape study (for nineteenth century North America), the ecological model (from biological sciences) and medical geography (itself an amalgam of botany, landscape and disease). Historic maps, and recent satellite data reveal the Central Valley’s topography and landscape features. Statistics, available as extrapolated estimates from the 1830s, and more systematically gathered beginning in the 1870s, relate historical disease patterns in a way that may be used to interpret the contemporary landscape. A multi-county study area permitted sub-regional comparisons, between such variables as effect of water projects, irrigation agriculture, and vector control on mosquito ecology. One sees both the theoretical premise of the ecological model, and the empirical evidence inherent in an infectious disease. However, to understand how not only human populations within the Great Central Valley adapted to malaria, but how anopheline mosquitoes adapted to human settlement, the conceptual research model is framed within the human-environmental relationship. In this study, socio-environmental systems are shown adapting to disease； the perspective of human geography, specifically the “man-land tradition,” integrates the work of medical, historical and cultural geographers. A second perspective, environmental history includes the work of historians who study disease within a human-environmental relationship. To some extent, a retrospective reality of malaria may still drive the current abatement methodologies, though any disease narrative ignores the past at its own peril. The California Central Valley manifests a complicated landscape, which is itself the result of history and the actions of those have settled it. In spite of incomplete and often erroneous health data, this study contributes to a greater understanding of the potential for re-emerging infectious disease in an agricultural region undergoing intense urbanization pressures.

Homelessness places people at risk of stressors that translate into stress and subsequently affect their health. Using Neuman Systems Model as a framework to identify modalities for nursing intervention among stressors and health problems of homeless people, this study compared stress levels among homeless people from three different homeless housing program types, investigated variables that predict the presence of stress among homeless, identified the degree of self-reported contact homeless people had with nurses, and measured to what extent nurses are preferred as health care providers by homeless people. This was accomplished through a cross-sectional, secondary data analysis of data from the evaluation study for the Chicago Plan to End Homelessness. Instruments utilized included the Perceived Stress Scale (PSS) and BPTSD-6. The sample size was 398 participants aged 18 years or older who were English speaking and clients of a homeless housing program in Chicago. The data was analyzed using ANOVA, multiple regression, odds ratios, and chi-square tests. The results of the study suggested that there were no differences in stress between participants of housing program types, and the variables prompting further assessment of stress in homeless patients included living with an adult child, availability of family and friends, psychiatric problem perception and burden, and PTSD. The results of the study also suggested that nurses were second to physicians in being seen as well as preferred by homeless participants.

This dissertation addresses issues of regional economics and natural resource economics. The dissertation has three main sections that can be taken as separate essays or chapters: I: Modeling U.S. Counties Innovation Capacity with a Focus on Natural Amenities. II: Modeling Regional Impacts from Rangeland Fires: An Application of a LP-SAM. III: Effects of Wildfires on Hospital Admissions for Respiratory Disease in Reno-Sparks, Nevada. Chapter one is a novel empirical analysis of innovation capacity for U.S. counties with a focus on natural amenities. This chapter analyzed the factors of innovation using three innovation measures: high technology employment share, number of patents and number of firms in Not-Elsewhere-Classified NEC) sectors as dependent variables. The explanatory variables are local economic conditions such as population, wage rate and unemployment rate； urban amenities such as crime rate, universities and museums； natural amenities such as climate, boat units, trailheads, camping ground and sites, and other outdoor activity services. An ordinary least square OLS) model and negative binomial NB) models were applied to the econometric analysis. Results showed that lower unemployment rates, more museums, more golf courses, warm winters and cool summers increased the innovation capacity for U.S. counties across three innovation measures. Therefore, the non-market attributes of natural amenities should be recognized in terms of their contribution to innovation capacity. Consequently, the competiveness of a county, in terms of innovation capacity, could be increased by providing more natural amenities to the public. Chapter two analyzed the regional economic impacts from rangeland fires in southeast Oregon. If a fire occurs on public grazing land, ranchers in the region are restricted access to public land for two years. This action would have substantial economic impacts in a regional economy because public grazing is a major source of forage for cattle and ranching in the region. This chapter developed procedures to simultaneously analyze the ranch and regional level impacts of rangeland fires on public land. This goal was attained by developing a model that links a linear programming LP) model and a Social Accounting Matrix SAM) model. Detailed modeling procedures are provided in this chapter. Results showed that the wildfire damage was far beyond simple wildfire suppression cost and direct impact on ranching business. The Cattle Sector, the Service Sector and the Manufacturing Sector are the most impacted sectors in the study area because southeast Oregon relies upon its ranching business. The distributional effect is the key finding from the LP-SAM model. Results showed that medium income households ranging from $25,000 to $75,000) are most impacted from rangeland fires. The distributive effects from wildfire for ranching business in Oregon are crucial to the policy makers who are dealing with rangeland fires and public land management. Chapter three analyzed the effects of wildfires on hospital admissions for respiratory disease in Reno-Sparks, Nevada. Wildfires, a common natural phenomenon in the western United States, produce a large amount of greenhouse gas and pollution particles, which might lead to health problems for nearby residents, and thereby increase local health care costs. This chapter identified 35 large wildfires that occurred near Reno-Sparks from 2005 to 2008 in order to explore the relationship between wildfires and health costs. The daily data of the major air pollutant PM2.5 from wildfires and number of respiratory patients admitted were employed in econometric equations to obtain the causal effect of wildfires on health. Results showed that wildfires that occurred away from Reno-Sparks within different distances increased the level of PM2.5 in the Reno-Sparks area. Furthermore, the regression on daily respiratory patients hospital admissions showed that the level of PM2.5 positively affected the number of patients admitted. The causal effect of number of patients admitted from wildfire varied by distances. The average damage of one acre burned by wildfires within a given distance category on hospital admissions in Reno-Sparks is $4.83. Wildfires that could affect the air quality in Reno-Sparks were identified with size over 500 acres. Therefore, if one average-size wildfire occurred near Reno-Sparks, the damage would cost approximately $156,096. This effect could be underestimated because historical data of wildfires showed that more than one wildfire occurred during the summer period, which means the damage could be much larger than estimated. Therefore, policy makers should pay more attention to wildfire management because wildfires do not only directly affecting the air quality. Overall, this dissertation describes the factors of innovation, regional economic impacts from rangeland fires and estimation of health effects from wildfires and offers suggestions to policy makers.

This thesis consists of three chapters that focus on the problem of measuring population health. Chapter One: Distribution-Sensitive Measurement of Population Health. It starts from the observation that standard measures of population health aggregate individuals well-being without taking account of any inequality in the distribution of health in the population. It tries to fill this gap in the literature by exploring two approaches to constructing a distribution-sensitive measure of population health. The first approach, based on Sens 1976) poverty index, provides an operationally-simple method for comparing the health of two populations when these populations vary with respect to a number of different health attributes. The second approach, drawing on work by Davidson and Duclos 2000), demonstrates that stochastic dominance can be used to compare the health of populations, both when a cardinal measure of individual well-being is available, and when individual health states are described only by a multi-dimensional vector of health attributes. Chapter Two: Robust distributionally-sensitive comparisons of population health. Drawing on the literature on the measurement of poverty and income inequality, it considers the use of stochastic dominance in measuring population health and compares this with a new measure developed in this thesis, the Mean Inequality Measure. It shows that stochastic dominance may not be the most appropriate approach to the measurement of population health, largely because of its sensitivity to individuals in very poor health. These two approaches are contrasted empirically, by applying them to the evaluation of health data from the US and Canada. Chapter Three: Measurement of Population Health Over Time. It extends the study of distribution sensitive measures of population health to take account of the specific conceptual challenges posed by the fact that individual health outcomes change over time. The strategy to tackle this issue in this paper is to first establish a number of axioms which should be satisfied by any useful measure of population health over time, and second to develop a measure which satisfies these properties. Subsequently, this measure is compared to a number of other possible approaches.

This study surveyed 20 practicing health care professionals [pediatricians, pediatric residents and medical students] who are currently treating children for medical problems in a local community medical clinic. This study examined the personal thoughts of the currently practicing physicians to better understand their approach to the care of overweight and obese youths and how they perceive social workers assisting them in controlling the current epidemic of overweight children. The clinic serves a population of children who come from low-income households. The study design was a qualitative, emergent design. This study highlighted gaps in health care treatment and service procedures that hinder overweight children from properly receiving the needed interventions for control of their weight. This study clarified further, the role that both parents and the community at large can accept in helping overweight children live healthier lifestyles. Lastly and most importantly the study focused on the importance of education of children and parents about how overweight exacerbates long term medical risk and how social workers can become active participants in the epidemic of childhood obesity.

This dissertation explores the effect of three different types of crises -conflict, natural disasters and macroeconomic crisis- on child health and nutrition. Chapter 1 seeks to generalize previous results on the effect of armed conflict on child malnutrition. Consistent with existing single-country studies but now using country data from a sample of developing countries, results show a sharp increase in the prevalence of chronic child malnutrition with increases in conflict intensity. Further, on average, both boys and girls seem to be deeply but equally affected by the shock. Moreover, I provide initial evidence of a potential malnutrition-reducing role of aid flows specific to post-conflict settings, supporting donors post-cold war pattern of aid, in which flows increase post-conflict instead of during conflict periods. Chapter 2 shows that even in a country classified as upper middle-income, with safety nets targeting the poor such as South Africa, children are not exempt from the short-term negative consequences of climatic shocks. With data from the province of Kwazulu-Natal and using a difference-in-difference approach, I exploit the exogenous intensity and spatial variation in a severe drought to find that children in affected areas had a higher probability of being ill and lower weight-for-height z-scores. Quantitative and qualitative evidence shows that these effects are potentially transmitted through channels other than losses in household income, namely through lack of access to water and/or food prices/shortages. Further, findings suggest that results are not driven by differences in HIV/AIDS or access to social grants between affected and non-affected groups, both relevant factors in the context of South Africa. Finally, Chapter 3 focuses on the decline in living standards from the Indonesian financial crisis and its effect on child height-for-age z-scores. I propose an individual fixed-effects methodology using, for the first time, both medium and long-term data and a continuous shock measure in the form of district-level changes in poverty rates. Where most of the existing research focuses on the short or medium terms and reports no average effect of the crisis on child nutrition, I find that, while indeed a generalized effect is not observed, the poorest and youngest children, particularly those in rural areas, show worse malnutrition levels in the medium-term, in some cases persisting almost a decade later. Results highlight the relevance of going beyond national averages, as important negative consequences of shocks for some groups might remain undetected.

Previous studies only partially explain the variance in condom use behavior among high-risk groups. In low income countries, social-structural factors rather than individual factors alone might better explain the variance in safer sex behaviors among sex workers. This study seeks to determine whether social and structural influence factors, moderated by individual characteristics, are important for female entertainers working in night clubs, spas/saunas, and karaoke bars with regard to HIV risk behavior in the Philippines. To examine social and structural influences, this study draws upon the theoretical constructs of social cognitive theory and the theories of reasoned action and planned behavior that may explain interpersonal and individual determinants of risk behavior, institutional theory that addresses organizational factors, and the theory of gender and power to explain social dynamics. An assumption underlying this study is that the entertainers sexual risk behaviors are influenced by peers, managers, and organizational policies of the workplace. Interpersonal and organizational influences interact to influence risk behavior of entertainers, and may be modified by depression, substance use, and social support. Four-hundred ninety-eight entertainers from 54 establishments and managers in 47 establishments in Quezon City, Metro Manila, the Philippines, were interviewed for this dissertation study, using structured surveys, in addition to focus groups and key informant interviews conducted prior to survey administration. The study utilized three types of multivariate regressions: linear regression for condom attitudes and consistent condom use), logistic regression for condom use negotiation), and multilevel modeling to control for individuals nested within establishments. Multiple imputation methods were conducted for missing variables and compared with non-imputed models. Findings indicate better condone attitudes among entertainers with social support vs. manager support alone, and poorer condom attitudes for those using drugs. Entertainers used condoms more consistently with casual partners and when they had more peer support for condoms. Condom negotiation was higher with casual partners than with regular partners, in establishments with a rule supporting condom use, and with less frequent drug use. Those who had both casual and regular partner types negotiated more when they were not depressed and when they drank alcohol less frequently. They negotiated more with their causal partners than their regular partners, and with increased peer and manager support. Entertainers with the condom attitude that “condom use depended on men in their culture” tended to negotiate condoms less with their partners. In a sub-analysis, predictors of depression included histories of sexual and emotional abuse and alcohol use. Although the managers held attitudes that supported condom use and their workers health, few managers in the study discussed condom use with their workers. Also, Kappa statistics revealed that agreement between the managers and workers reports about condom use rules and condom availability in their establishments was not higher than would be expected from chance alone. Results suggest a need for multilevel interventions among entertainment workers and their managers that incorporate workplace and city health department policy and address the impact of depression, drug use, sexual and emotional abuse, social support, partner type, and gender related societal norms for condom use.

The Centers for Disease Control and Prevention predicted that the incidence of HIV/AIDS among the American population aged 50 years and over would account for 50% of all HIV/AIDS cases in the United States by 2015. Yet there is little research available about the reason for the high prevalence or the availability of interventions available for this population. The purpose of this study was to explore factors related to HIV/AIDS awareness in a group of older recipients of health care and their health care providers. This study was intended to help fill the information gaps in this area by using a conceptual framework based on systems theory and social network theory. A mixed-method design was used to collect data； semi-structured interviews were conducted with a sample of 17 health care providers and analyzed using thematic content analysis. Structured exploratory surveys were distributed to a sample of 99 HIV/AIDS-positive adults, aged 50 and over, to understand common factors likely to have contributed to their infection and the results were compared to common factors already identified in the general population in the literature. Results of this study indicate possible communication barriers that exist between care providers and their older clients when discussing topics related to behavioral factors contributing to HIV/AIDS. This study contributes to social change by adding support to the importance of education programs and awareness campaigns about factors related to HIV/AIDS to curtail the spread of this disease among older adults.