Chapter 1. Background: Asthma is the most common chronic childhood disease and is characterized by recurrent airway obstruction, bronchial hyper-responsiveness, and airway inflammation. Asthma is the leading cause of childhood hospitalization and school absenteeism in the United States. The associations between adverse respiratory effects and exposure to indoor nitrogen dioxide (NO2) and other byproducts of combustion such as particulate matter (PM) in particular ultrafine particulates (UFP), Ozone (O3) and Sulfur Dioxide (SO2), have been the focus of many epidemiological studies in recent years. Indoor exposure to NO2 and other pollutants from combustion may increase the risk of acute and chronic respiratory disease, reduce lung function, initiate and exacerbate asthma in children. The levels of exposure to NO2 indoors are of public health concern because children spend nearly 70% of their time indoors at home. According to the 2010 US Census report, approximately 39% of US households use natural gas for cooking, and the primary source of residential NO2 is a gas-fuel cooking appliance. Indoor levels of NO2 where NO2 sources are present can be much higher than outdoors, where the primary source of NO2 is vehicular traffic. Epidemiological studies in developed countries suggest that gas stoves used for cooking and/or heat are associated with an increased risk of asthma and respiratory symptoms in children. While there are numerous, epidemiological studies supporting an association between increased NO2 levels and gas stoves and asthma symptom severity in children, there are other studies that have examined the relationship in homes that did not observe significant associations. A better understanding of how NO2 and other indoor environmental (e.g., environmental tobacco smoke (ETS), allergens) exposures contribute to asthma morbidity in inner city preschool children will allow interventions to more effectively designed and implemented. To date, there are conflicting results on the role of exposure to indoor NO2 and its association with new-onset asthma in young inner-city children. The recent studies assessing the effects of indoor NO2 on asthma morbidity were limited to inner-city children, largely older, who were diagnosed with asthma. A gap in knowledge remains regarding the role indoor NO2 plays on the development of asthma in children not previously diagnosed. The scientific and public health rationale for conducting this dissertation was to describe the association of exposure to indoor NO2 and primary sources with the initiation and exacerbation of asthma symptoms among pre-school children with and without diagnosed asthma. The data analyzed in the current research come from a larger study of Endotoxin, Obesity, and Asthma (EOA) in the New York City Head Start Program, funded in the summer of 2002. The primary research objective of that study was to identify modifiable risk factors associated with asthma and asthma persistence among preschool children from low-income families living in select New York City neighborhoods with high pediatric asthma hospitalization rates. We conducted a cross-sectional analysis of data collected from the study questionnaire and home visit sampling at study enrollment. The analyses were performed in two phases: the first phases used data collected at study enrollment and the second phase used data collected 12-months after study baseline. Henceforth, the dissertation will refer to the first analyses as the baseline study and the second as the follow-up study. The research evaluated the association of NO2 exposure with asthma status among New York City Head Start children with and without asthma at study enrollment and with respiratory symptoms among children with asthma at 12-month follow-up. Chapter 2. Baseline Study: We conducted a cross-sectional analysis of data collected from the study questionnaire and home visit sampling at study enrollment. Specifically, the research sought to evaluate the association of NO2 exposure with asthma status among New York City Head Start children with and without asthma at study enrollment and with respiratory symptoms among children with asthma at enrollment. A total of 503 children were included in the baseline study. A total of 105 children (20.9%) met the criteria for both asthma and allergy, and 67 (13.3%) met the criteria for asthma alone. Girls made up 51.7% and boys, 48.3% of the 503 study participants. Descriptive analyses suggested that asthma/allergy status was associated with: male gender, non-Mexican ethnicity/national origin, presence of a smoker in the child’s home, number of smokers in the child’s home, self-reported parental history of asthma, mother’s education level and sensitization to one or more of the four allergens. Logistic regression models were used to investigate the magnitude and direction (as well as trend) of the association between childhood asthma and indoor NO2 sources in the child’s home. Chapter 3. Follow-up Study: Our follow-up study involved the analysis of the 12-month follow-up data from the study of Endotoxin, Obesity, and Asthma in the New York City Head Start Program funded in the summer of 2002. We focused on assessing the magnitude and direction of the associations of exposure to indoor NO2 levels (based on baseline NO2 measurements) with children’s asthma status and with symptom severity among asthmatics at 1-year follow-up. For the follow-up study, we categorized children by whether their asthma status had changed since baseline. Descriptive analyses were performed looking at key characteristics by “change in asthma status.” Children’s asthma status at baseline and at follow-up, were based on responses to the questionnaire. We analyzed indoor NO2 level measurements at baseline in relation to asthma outcomes on follow-up. We did not have enough data on NO2 levels at follow-up to analyze them in relation to asthma status on follow-up. Unless the family had relocated since baseline and/or reported changes since baseline in the use of gas appliances or the number of smokers in the home, we assumed that baseline NO2 levels in the participating children’s homes were reasonable proxies for current exposures. We looked at the number of children who moved since baseline and whether the move (for example, looking at gas stove status, age of new building) may have impacted indoor NO2 levels. Of the 503 children who were included in the baseline analyses, 47.3% had data on asthma status on follow-up. A total of 238 children (111 male, 127 female) were grouped into the four mutually exclusive outcome categories: 122 (51.3%) did not have asthma at baseline or on follow-up, 34 (14.3%) had asthma on follow-up but not at baseline, 65 (27.3%) had asthma at baseline but not on follow-up, and 17 (7.1%) had asthma at baseline and on follow-up. The mean age at 1-year follow-up was 59.5 months (6.95), and neither age nor gender was associated with asthma. The distribution of ethnicity/national origin among the 238 children remained the same as at baseline; no one ethnicity group experienced disproportionate loss to follow-up, and asthma status remained associated with non-Mexican ethnicity/national origin, although 44.1% with new-onset asthma were of Mexican background. Asthma was also associated with self-reported parental history of asthma and allergy in children, but nearly 80% of children with new-onset asthma had no such parental history of asthma. More parents of children with new-onset (35.3%) or persistent asthma (23.5%) than of other children reported making efforts to reduce risk factors or triggers for asthma exacerbations in the past 12 months. Chapter 4. Dissertation Conclusion : The primary objective of the dissertation research was the examination of the relationship between asthma and asthma severity and exposure to gas cooking and residential NO2. In both our baseline and 12-month follow-up studies, exposure to indoor NO2 was represented by the baseline measurement of NO2 and the NO2 surrogate, gas stove. Asthma status of children was based on parental responses on the questionnaire regarding asthma symptoms and urgent care visits due to respiratory distress over the course of each 12-month period prior to the conducting study questionnaires. For both studies, we did not find an association between exposure to NO2 levels at baseline and asthma status or severity. Our findings contradict the results of most recent studies of both NO2 levels and residential sources of NO2 and their effects on asthma symptoms in very young children. However, it remains difficult to compare our results we those of previous published studies because those studies primarily focused on children who were diagnosed with asthma, whereas our research included preschool aged children with and without asthma. Based on our findings and the fact they conflict with other epidemiological studies, of which there were also conflicting results, we feel that the relationship between asthma symptoms and NO2 exposures remains ambiguous. The lack of consistent results of epidemiological research raises questions that should be the focus of future epidemiological studies. What are the roles of co-pollutants and co-risk factors? Does NO2 work alone or in concert with other indoor pollutants? There exists a real lack of understanding on the possible synergistic effects of exposure to NO2 and other combustion byproducts. Important to furthering our knowledge of the role of exposure to indoor NO2 and asthma is determining whether NO2 acts as a surrogate for co-pollutants that are considered risk factors for asthma and other respiratory conditions. Another focus of future indoor pollution studies should be the development of effective methods and technologies for measuring the constituents of the complex mixture of pollutants in indoor air; these methods and technologies can then be applied in personal monitoring of exposure to indoor pollutants in epidemiological studies that would help to determine with much more accuracy the effects of individual indoor pollutants on asthma and other respiratory symptoms. This knowledge would help in the development of more effective public health and environment policies towards reducing the burden of childhood asthma.

Open burning of biomass and anthropogenic waste is a major source of aerosols at the biosphere-atmosphere interface, yet its impact on Earth’s climate and air quality is not fully understood due to the intricate feedbacks between the natural environment and human activities. Earth system models (ESMs) are a vital tool in the study of these aerosol-biosphere-atmosphere interactions. ESMs allow the estimation of radiative forcing and climate impacts in terms of changes to temperature and precipitation as well as the attribution to natural or anthropogenic drivers. To provide coherent results, however, ESMs require rigorous development and evaluation against observations. In my work I use the NASA-GISS ESM: ModelE. One of its strengths lie in its detailed aerosol schemes that include microphysics and thermodynamic partitioning, both necessary for the simulation of secondary inorganic aerosols. To overcome one of ModelE’s weaknesses, namely its lack of interactive biomass burning (BB) emissions, I developed pyre: ModelE’s interactive fire emissions module. pyrE is driven by flammability and cloud-to-ground lightning, both of which are calculated in ModelE, and anthropogenic ignition and regional suppression parameterizations, based on population density data. Notably, the interactive fire emissions are generated from the flaming phase in pyrE (fire count), rather than the scar left behind (burned area), which is commonly used in other interactive fire modules. The performance of pyrE was evaluated against MODIS satellite retrievals and GFED4s inventory, as well as simulations with prescribed emissions. Although the simulated fire count is bias-high compared to MODIS, simulated fire emissions are bias-low compared to GFED4s. However, the bias in total emissions does not propagate to atmospheric composition, as pyrE simulates aerosol optical depth just as well as a simulation with GFED4s prescribed emissions. Upon the development and evaluation of the fire-aerosol capabilities of ModelE, I have utilized it, with the EVA health model, to study the health impacts of outdoor smoke in 1950, 2015, and 2050. I find that chronic exposure to aerosols (PM2.5) is the main driver of premature deaths from smoke exposure, yet by 2050, acute exposure to ozone, formed downwind of BB smoke plumes, is projected to cause more premature deaths than exposure to PM2.5. I estimate the annual premature deaths from BB and waste burning (WB) smoke in 1950 to be ~41,000 and ~19,000, respectively, and in 2015 to be ~310,000 and ~840,000, respectively. By 2050 I project 390,000 and 1.5 million premature deaths from BB and WB respectively. In light of the growing impact of WB smoke exposure I identify the need to scale up viable waste management practices in regions of rapid population growth.

M. Tech. Environmental Health / Particulate Matter (PM) is a complex, heterogeneous mixture of smoke, soot, dust, salt, acids, and metals. Particulate Matter varies in concentration, size, chemical composition, surface area and sources of origin. Given the known ambient particulate pollution problem, the potential health risks posed by PM to the population of eMbalenhle are unknown. eMbalenhle (the study area) is a township located in Mpumalanga Province, about 12 km from Secunda. The area is surrounded by industries, power stations and mines, all of which are recognised emitters of PM. The main aim of this study was to assess the health risks posed by ambient PM10 exposure to the population of eMbalenhle.

Indiana University-Purdue University Indianapolis (IUPUI) / Objective: "The purpose of this research is: Study the use of Moderate Resolution Imaging Spectroradiometer (MODIS) data in retrieving the aerosol optical depth (AOD) over Indiana State at high resolution of 500 meters. Examine the potential of using the resulted AOD data as an indicator of particulate air pollution by comparing the satellite derived AOD data with the ground measurements (provided from the continuous air monitors available over the study area). If an association should be found, AOD data would be used to map particulate matter (PM) concentration. Assess current and future ambient concentrations of air pollutants in the State of Indiana using the AOD."

An energy transition can be defined as the adoption of a new primary energy system. As such, it is a structural change that implies a broad shift in technologies and behaviors in order to replace one source of energy with another. Energy transitions motivated by economic, climate, and/or health goals are taking place everywhere in the world. Air pollution, a consequence of fossil fuel-based and solid fuel-based energy use among other sources, is the largest environmental health risk accounting for 6.4 million premature deaths annually. Given the health implications of our energy systems and the social drivers of energy use, access, and burden, energy transitions have the potential to impact health outcomes and associated disparities in a context-dependent manner. The research presented in this dissertation has two objectives: 1) to evaluate the distribution of benefits from energy transitions in the United States (US) and in Ghana; 2) to identify and characterize health outcomes that are relevant to these transitions but are currently understudied. Chapters 2 and 3 are anchored in the US energy policy context. Chapter 2 focuses on the transition away from coal as the primary source of energy in the US and its implications for particulate matter pollution and preterm births. Leveraging a novel dispersion model, I assessed the association between coal PM2.5 and preterm birth rates along with effect modification by race/ethnicity. We observed a positive non-linear relationship between coal PM2.5 and preterm birth rate, which plateaued at higher levels of pollution. The findings of this study suggest that the transition away from coal may have reduced preterm birth rates in the US, but that the association was stronger among non-Hispanic White women compared to non-Hispanic Black women. In Chapter 3, I use a mixed-methods framework to evaluate the first pilot of gas-to-electric stove transition in low-income housing in the US. Through a collaboration with a community-based environmental justice group in New York City, we monitored indoor air quality in participants homes pre- and post-intervention, conducted controlled cooking tests, and carried out focus groups to characterize their experience. Post-intervention, daily NO2 concentrations were 46.3% (95% CI: -67.8%, -10.3%) lower in the intervention arm compared to the control arm. Participants were unanimously pleased with the transition, which had simultaneously improved their cooking experience and partially addressed energy insecurity concerns that plagued their building. Chapters 4 and 5 contribute to the characterization of the health implications associated with Ghana’s commitment to increase access to liquified petroleum gas (LPG) nationally. In Chapter 4, we aimed to describe the country-level incidence of severe cooking-related burns by fuel type in use and to identify effect modifiers. We conducted a nationally representative (n = 7,389) household energy use survey in all 16 regions of Ghana. The incidence rate (95% CI) of cooking-related burns per 1000 person-years among working age females was 8.3 (7.2; 12.0) per 1000 person-years, which was 8 times higher than that of working age males. Among adults, the odds of experiencing a cooking-related burn were doubled among solid fuel users compared to primary LPG users. In Chapter 5, we aimed to understand the effect of an LPG intervention on personal exposure to household air pollution in a peri-urban setting of Ghana and to characterize the distribution of benefits between male and female household members quantitatively and qualitatively. The difference-in-differences results suggested a reduction in exposure to carbon monoxide due to the intervention (-14.2%, 95% CI: -44.1%, 31.6%), which might have benefited male and female participants to the same extent. Importantly, focus group discussions suggested that the ability that LPG affords to cook “on demand” could negatively impact intra-household dynamics and expectations in a way that has not been previously documented. Taken together, our findings from two distinct socioeconomic contexts, highlight the non-uniform distribution of benefits from energy transitions, especially when benefits that are not mediated by air pollution are considered. These results contribute to the understanding that social drivers of inequities should be integrated in the design of energy policies and interventions aimed at generating equitable outcomes. Future directions include a detailed characterization of the context-dependent relative contributions of indoor and outdoor air pollution sources as well as a more systematic integration of quantitative and qualitative methods in policy evaluation.

Environmental exposure levels are often sufficiently disparate between populations such that there is little or no overlap, complicating our ability to ascertain the full dose-response curve and as such create informed regulatory policy.I reviewed the literature on methods available to address non- and partially-overlapping exposure distributions, drawing from both epidemiology as well as other relevant disciplines to describe the universe of proposed solutions. I also used the case study of maternal PCB-153 exposure and birthweight, utilizing real-world and simulated data to explore our ability to ascertain “true” dose-response curves from observational data given the limited cohort-specific exposure ranges. I investigated the importance of controlled and uncontrolled confounding as well as the impact of sample size on our ability to ascertain a “true” underlying dose-response curve. Pooling and meta-analysis were useful to increase the heterogeneity of exposure distributions despite imperfect confounding control and heterogenous confounding structures across cohorts. The analyses also serve as continued evidence of the challenges of making population-wide inferences from study samples with restricted exposure ranges as well as the danger of pooling multisite data without sufficiently accounting for heterogeneity in both exposure level and distribution of confounders. These results highlight the limitations of using both individual studies and systematic reviews of environmental chemicals, and emphasize the need for pooling and meta-analysis to widen exposure distributions that in turn permit us to accurately capture the negative effects of these environmental chemicals.

Exposure to environmental chemicals has been shown to affect health status throughout the life course. Quantifying the joint effect of environmental mixtures over time is crucial to determine optimal intervention timing. Establishing causal relationships from environmental mixture data can be challenging due to various factors, including multicollinearity, complex functional form of exposure-response relationships, and residual unmeasured confounding. These issues can lead to biased estimates of treatment effects and pose significant obstacles in accurately identifying the true relationship between the pollutants and outcome variables. Causal interpretation of longitudinal environmental mixture studies encounters challenges. This dissertation explores the use of causal inference in environmental mixture studies, with a particular emphasis on addressing three key challenges. First, there is currently no statistical approach that allows simultaneous consideration of time-varying confounding, flexible modeling, and variable selection when examining the effect of multiple, correlated, and time-varying exposures. Second, the violation of a critical assumption that underpins all causal inference methods - namely, the absence of unmeasured confounding - poses a significant problem, as models that incorporate multiple environmental exposures may exacerbate the degree of bias depending on the nature of unmeasured confounding. Finally, there is a lack of computational resources that facilitate the application of newly developed causal inference methods for analyzing environmental mixtures. In Chapter 2, we introduce a causal inference method, g-BKMR, which enables to estimate nonlinear, non-additive effects of time-varying exposures and time-varying confounders, while also allowing for variable selection. An extensive simulation study shows that g-BKMR outperforms approaches that rely on correct model specification or do not account for time-dependent confounding, especially when correlation across time-varying exposures is high or the exposure-outcome relationship is nonlinear. We apply g-BKMR to quantify the contribution of metal mixtures to blood glucose in the Strong Heart Study, a prospective cohort study of American Indians. Chapter 3, we address the issue of time-varying unmeasured confounding when estimating time-varying effects of exposure to environmental chemicals. We review the Bayesian g-formula under the assumption of no unmeasured confounding, and then introduce a Bayesian probabilistic sensitivity analysis approach that can account for multiple, potentially time-varying, unmeasured confounders and continuous exposures. Through a simulation study, we demonstrate that the proposed algorithm outperforms the naive method, which fails to consider the influence of confounding. Chapter 4, introduces causalbkmr, a novel R package and can be currently be accessed on Github. causalbkmr is designed to support the implementation of g-BKMR, BKMR Causal Mediation Analysis, and Multiple Imputation BKMR, thereby offering a user-friendly and effective platform for executing these state-of-the-art methods in practice in the context of complex mixtures analysis. While the package bkmr is available, the novel package causalbkmr expands upon bkmr by enabling its application specifically to environmental mixture data within a causal inference framework. The implementation of these novel methodologies within causalbkmr allows for the extraction of causal interpretations, thus enhancing the analytical capabilities provided by the package. Chapter 5 concludes with a discussion and outlines potential future directions for investigation.

Recent and past events in the South Durban Basin (SDB) have highlighted the possible connection between perceived air pollution concerns and deteriorating health. The aim of this study is to explore how the SDB community perceives air pollution and if this can be related to some of the health problems that are experienced in these communities. The ultimate aim is to determine whether the respondents connect air pollution to specific illnesses and also how they perceive the actions used to combat air pollution and its negative health impacts. This information is gathered through a questionnaire conducted in various study areas both within and outside the SDB, then compared with demographic variables. This is done to determine if perception is related to direct industrial activity or if there are other factors influencing results. Variables such as race and level of education had little impact on the results. The results showed that areas within the SDB experience a high neighborhood satisfaction despite the current general belief of air pollution being connected to ill health. The control areas outside of the SDB support theories of gender and race and its relationship to air pollution, currently presented by researchers in the perception field. Strong associations’ do exist between general environmental satisfaction and gender. Perception of current legal enforcement is exerting a strong effect on air pollution perception formation. Vitally important is the connection of daily concrete experiences of air pollution with the lack of transparency and communication between industry and communities. It is resulting in the current perception of illness being connected to tangible air pollution. The aim is to further future studies on establishing links between health and air quality. Gaining insight from the study of public risk perceptions based on local knowledge and experience in particular places, will help shape the role of environmental policy and management response systems. / Geography / M.A. (Geography)

published_or_final_version / Community Medicine / Master / Master of Philosophy

Radon is a naturally occurring radioactive element in the 238U decay series that is found in high concentrations in certain geological formations such as Caves. Exposure to high concentrations of radon has been positively linked to the incidence<br /> of lung cancer. This study used Electret ion chambers and the RAD7 continuous radon monitor to measure radon concentrations in the Cango Caves in the Western Cape Province, South Africa. Measurements were taken during summer i.e. February 2004 and March 2005. The results for the radon activity concentrations range from the minimum of<br /> about 800 Bq.m-3 to a maximum of 2600 Bq.m-3. The two techniques give very similar results, though the Electret ion chamber results appear to be consistently higher by a few percent where measurements were taken at the same locations. A<br /> mathematical model has been developed to investigate the radon concentrations in the Cave. Diffusion and ventilation have been considered as mechanisms for explaining the distribution of radon concentrations. The ventilation rate in the Cave has been estimated under certain assumptions, and it is found to be about 7 &times / 10&minus / 6 s&minus / 1 for the Van Zyl hall which is the first large chamber in the Cave. The radon concentration increases as one goes deeper into the Cave, but then becomes fairly constant for the deeper parts. The annual effective dose that the guides are exposed to in the Cave as a result of the radon concentrations, depends strongly on the time that they spend in the Cave and in which, halls they spend most of their time in the Cave. The initial results indicate an annual effective dose of 4-10 mSv, but this needs to be further investigated.