Browsing by Author "Keller, Kayleigh, committee member"
Now showing 1 - 6 of 6
Results Per Page
Sort Options
Item Open Access Association between exposure to cadmium and lead during gestation and adverse birth outcomes in the household air pollution intervention network (HAPIN) trial(Colorado State University. Libraries, 2024) Alhassan, Mohamed Adnan, author; Peel, Jennifer, advisor; Clark, Maggie, committee member; Keller, Kayleigh, committee member; Neophytou, Andreas, committee memberLow- and middle-income countries (LMICs) are particularly vulnerable to the adverse effects of metal exposure. These countries' rapid industrialization coupled with population growth, result in substantial environmental exposures, which many governments have limited capacity to formally regulate. Even when regulations exist, many governments have a limited capacity to enforce those regulations. Additionally, LMICs bear a disproportionate burden of adverse birth outcomes, including low birth weight and preterm birth, which carry long-term health implications such as increased risk of chronic diseases, developmental delays, and mortality. Several studies have examined the association between metals and adverse birth outcomes such as low birth weight and preterm births. Specifically, despite the low number of studies, cadmium has been consistently linked to lower birth weights, smaller sizes for gestational age, and reduced head circumference. However, the association between lead exposure and birth outcomes shows inconsistent results. This inconsistency in findings, along with the low number of studies overall, especially in LMICs, regarding lead has prompted further investigation in our current study. Here we utilized data from the Household Air Pollution Intervention Network (HAPIN) trial, a randomized controlled trial conducted in rural areas of Guatemala, Peru, Rwanda, and India. The HAPIN trial evaluated the impact of replacing biomass stoves with liquefied petroleum gas stoves on various health outcomes, including infant birth weight among 3200 participants. The participants in the current analysis included pregnant women with a live singleton birth with exposure and birth data (n=2396). Maternal exposure to cadmium and lead were evaluated by analyzing dried blood spots using inductively coupled mass spectrometry. Blood spots were collected at baseline (9 - <20 weeks gestational age) and 32-36 weeks gestational age; we also evaluated the average of these two measurements. Birth weight was measured using a digital infant scale, with low birth weight defined as <2500 grams, and gestational age at birth was determined using screening data and ultrasonography, with preterm birth defined as <37 weeks. We utilized linear regression for birth weight and gestational age, logistic regression for dichotomous low birth weight, and Cox proportional hazards model for preterm birth. The models accounted for infant sex, maternal age, nulliparity, body mass index, maternal hemoglobin, mother's dietary diversity, food insecurity, tobacco smoking in the household, and study arm. We assessed effect modification by study location, sex, and study arm by including an interaction term. In sensitivity analyses, we included study location, household assets, maternal education in the models; replaced values below the limits of detection (LOD) with LOD/√2, and evaluated metal concentrations standardized by potassium levels. We also excluded maternal hemoglobin from the main model. The mean birth weight was 3,020 (standard deviation [SD]=445.5) grams, and 10.3% of all births were classified as low birth weight. The mean gestational age was 39.5 weeks (SD=1.7 weeks), and 5.2% of the births were preterm. The median lead concentration across the time points was 1.4 μg/dL (IQR: 0.9 – 2.2 μg/dL), and the median cadmium concentration was 1.0 ng/mL (IQR: 0.7 – 1.4 ng/mL), values comparable to those found in other studies. Overall, the results did not indicate a consistent or strong association between lead or cadmium and adverse birth outcomes. Baseline cadmium levels showed a modest increase in the odds ratio for low birth weight (OR per IQR increase: 1.2, 95% CI: 0.97 to 1.47). Sensitivity analyses closely aligned with the main findings. All the results for effect modification did not indicate differences in the strata. The study found a suggestive, but inconsistent evidence between exposure to cadmium and low birth weight. This study has some limitations. There is potential for non-differential measurement error due to the hematocrit effect, which alters the estimated spot volume based on participants' hematocrit levels. A sensitivity analysis using potassium standardized metal concentrations partially addressed this, but individual hematocrit variability can still bias the observed association towards the null, with a moderate magnitude. The probability of the bias is moderate. The chromatographic effect, which can cause variations in concentration due to the interaction between blood and the analyte with the filter paper, was also partially addressed using internal standards, blanks, calibration samples, quality controls, and reference materials. This potential bias is of low probability and magnitude, biasing the observed association toward the null. Confounding bias was considered a concern due to incomplete adjustment for covariates like seasonal variation, which can affect metal exposure and birth outcomes. Sensitivity analyses supported the main model findings, suggesting a low probability and magnitude of confounding bias, which could bias the observed association towards or away from the null. Despite residual confounding concerns linked to socio-economic indicators like assets and diet diversity, the sensitivity analyses did not deviate from the main model findings, indicating a small probability and magnitude of the bias, which would bias the observed association in either direction. The study had several strengths including a large sample size compared to previous studies, especially those in LMICs and it was conducted in three distinct rural LMIC settings, which, to the best of our knowledge, had not been done before. This study's strength lies in its large sample size of 2,152 participants with complete data, enhancing its statistical robustness and addressing the common issue of small sample sizes and missing data in prior LMIC research. Additionally, its unique examination across three distinct rural LMIC settings provides valuable insights into the regional variations affecting the outcomes studied. Future steps include using whole blood samples instead of dried blood spots (DBS) and measuring exposure at multiple time points, particularly at birth via the umbilical cord, could yield more accurate concentrations. It is also recommended that subsequent studies employ better socio-economic indicators to reduce residual confounding effects. Expanding the geographical scope of the study to include a broader range of urban areas within the HAPIN countries would improve the generalizability of the findings. Additionally, future research should consider analyzing the effects of metal mixtures to better replicate real-world environmental conditions and interactions. The results are generally consistent with existing limited data indicating no evidence of an association between lead and adverse birth outcomes and a potential association between higher cadmium exposure during pregnancy with increased risk of low birth weight.Item Embargo Bayesian tree based methods for longitudinally assessed environmental mixtures(Colorado State University. Libraries, 2024) Im, Seongwon, author; Wilson, Ander, advisor; Keller, Kayleigh, committee member; Koslovsky, Matt, committee member; Neophytou, Andreas, committee memberIn various fields, there is interest in estimating the lagged association between an exposure and an outcome. This is particularly common in environmental health studies, where exposure to an environmental chemical is measured repeatedly during gestation for the assessment of its lagged effects on a birth outcome. The relationship between longitudinally assessed environmental mixtures and a health outcome is also of greater interest. For a single exposure, a distributed lag model (DLM) is a widely used method that provides an appropriate temporal structure for estimating the time-varying effects. For mixture exposures, a distributed lag mixture model is used to address the main effect of each exposure and lagged interactions among exposures. The main inferential goals include estimating the lag-specific effects and identifying a window of susceptibility, during which a fetus is particularly vulnerable. In this dissertation, we propose novel statistical methods for estimating exposure effects of longitudinally assessed environmental mixtures in various scenarios. First, we propose a method that can estimate a linear exposure-time-response function between mixture exposures and a count outcome that may be zero-inflated and overdispersed. To achieve this, we employ a Bayesian Pólya-Gamma data augmentation with a treed distributed lag mixture model framework. We apply the method to estimate the relationship between weekly average fine particulate matter (PM2.5) and temperature and pregnancy loss with live-birth identified conception time series design with administrative data from Colorado. Second, we propose a tree triplet structure to allow for heterogeneity in exposure effects in an environmental mixture exposure setting. Our method accommodates modifier and exposure selection, which allows for personalized and subgroup-specific effect estimation and windows of susceptibility identification. We apply the method to Colorado administrative birth data to examine the heterogeneous relationship between PM2.5 and temperature and birth weight. Finally, we introduce an R package dlmtree that integrates tree structured DLM methods into convenient software. We provide an overview of the embedded tree structured DLMs and use simulated data to demonstrate a model fitting process, statistical inference, and visualization.Item Open Access Integrated statistical models in ecology(Colorado State University. Libraries, 2023) Van Ee, Justin, author; Hooten, Mevin, advisor; Koslovsky, Matthew, advisor; Keller, Kayleigh, committee member; Kaplan, Andee, committee member; Bailey, Larissa, committee memberThe number of endangered and vulnerable species continues to grow globally as a result of habitat destruction, overharvesting, invasive species, and climate change. Understanding the drivers of population decline is pivotal for informing species conservation. Many datasets collected are restricted to a limited portion of the species range, may not include observations of other organisms in the community, or lack temporal breadth. When analyzed independently, these datasets often overlook drivers of population decline, muddle community responses to ecological threats, and poorly predict population trajectories. Over the last decade, thanks to efforts like The Long Term Ecological Research Network and National Ecological Observatory Network, citizen science surveys, and technological advances, ecological datasets that provide insights about collections of organisms or multiple characteristics of the same organism have become prevalent. The conglomerate of datasets has the potential to provide novel insights, improve predictive performance, and disentangle the contributions of confounded factors, but specifying joint models that assimilate all the available data sources is both intellectually daunting and computationally prohibitive. I develop methodology for specifying computationally efficient integrated models. I discuss datasets frequently collected in ecology, objectives common to many analyses, and the methodological challenges associated with specifying joint models in these contexts. I introduce a suite of model building and computational techniques I used to facilitate inference in three applied analyses of ecological data. In a case study of the joint mammalian response to the bark beetle epidemic in Colorado, I describe a restricted regression approach to deconfounding the effects of environmental factors and community structure on species distributions. I highlight that fitting certain joint species distribution models in a restricted parameterization improves sampling efficiency. To improve abundance estimates for a federally protected species, I specify an integrated model for analyzing independent aerial and ground surveys. I use a Markov melding approach to facilitate posterior inference and construct the joint distribution implied by the prior information, assumptions, and data expressed across a chain of submodels. I extend the integrated model by assimilating additional demographic surveys of the species that allow abundance estimates to be linked to annual variability in population vital rates. To reduce computation time, both models are fit using a multi-stage Markov chain Monte Carlo algorithm with parallelization. In each applied analysis, I uncover associations that would have been overlooked had the datasets been analyzed independently and improve predictive performance relative to models fit to individual datasets.Item Open Access Statistical models for COVID-19 infection fatality rates and diagnostic test data(Colorado State University. Libraries, 2023) Pugh, Sierra, author; Wilson, Ander, advisor; Fosdick, Bailey K., advisor; Keller, Kayleigh, committee member; Meyer, Mary, committee member; Gutilla, Molly, committee memberThe COVID-19 pandemic has had devastating impacts worldwide. Early in the pandemic, little was known about the emerging disease. To inform policy, it was essential to develop data science tools to inform public health policy and interventions. We developed methods to fill three gaps in the literature. A first key task for scientists at the start of the pandemic was to develop diagnostic tests to classify an individual's disease status as positive or negative and to estimate community prevalence. Researchers rapidly developed diagnostic tests, yet there was a lack of guidance on how to select a cutoff to classify positive and negative test results for COVID-19 antibody tests developed with limited numbers of controls with known disease status. We propose selecting a cutoff using extreme value theory and compared this method to existing methods through a data analysis and simulation study. Second, there lacked a cohesive method for estimating the infection fatality rate (IFR) of COVID-19 that fully accounted for uncertainty in the fatality data, seroprevalence study data, and antibody test characteristics. We developed a Bayesian model to jointly model these data to fully account for the many sources of uncertainty. A third challenge is providing information that can be used to compare seroprevalence and IFR across locations to best allocate resources and target public health interventions. It is particularly important to account for differences in age-distributions when comparing across locations as age is a well-established risk factor for COVID-19 mortality. There is a lack of methods for estimating the seroprevalence and IFR as continuous functions of age, while adequately accounting for uncertainty. We present a Bayesian hierarchical model that jointly estimates seroprevalence and IFR as continuous functions of age, sharing information across locations to improve identifiability. We use this model to estimate seroprevalence and IFR in 26 developing country locations.Item Open Access The association of lunar phases with pregnancy at first artificial insemination of dairy cows(Colorado State University. Libraries, 2024) Schatte, Margaret, author; Grandin, Temple, advisor; Pinedo, Pablo, advisor; Keller, Kayleigh, committee memberMyths and old farming legends have circulated the belief that the full moon affects livestock behavior and reproduction. To assess this association in dairy cattle, 13,558 records from 2019 to 2021 at an organic dairy farm in Colorado were analyzed. These records included lactation number, artificial insemination date (AI date), and pregnancy result. AI date was categorized into season and lunar phases. Lunar phases were separated into four equal categorizations: new moon, first quarter, full moon, and third quarter. The primary objective of this study was to identify any associations between the lunar phases and PAI1 (pregnancy rate at first AI). The secondary objective of this study was to use logistic regression to specify which phases had the lowest and greatest PAI1 while accounting for other effects on fertility. Logistic regression was used to complete this by comparing the pregnancy result of lunar phases while accounting for lactation number and season, which are known to affect pregnancy rate. The 4 lunar phase categorizations did have an association on PAI1 (p<0.05). The new moon phase resulted in the lowest PAI1 at 35.3% while the third quarter was the highest at 38.3%. Estimated marginal means were explored to identify the seasonal effect on pregnancy and found that winter had the highest probability of pregnancy and summer had the lowest. This analysis of 3 years of records provides evidence that the week of the new moon is the least probable week out of the month for pregnancy after first AI to occur, while the week before and week of the full moon are the most probable days for pregnancy after first AI to occur within the lunar cycle.Item Embargo The impact of control on national-scale livestock disease outbreaks in the United States(Colorado State University. Libraries, 2023) Smith, Samuel M., author; Webb, Colleen T., advisor; Beck-Johnson, Lindsay M., committee member; Keller, Kayleigh, committee memberOutbreaks of livestock diseases, like foot-and-mouth disease (FMD) and bovine tuberculosis (bTB), pose a significant economic threat to the United States livestock industry. Significant interest then lies in developing strategies to mitigate the impact of an outbreak should they occur. This thesis explores the effect of control interventions on outbreaks of FMD and bTB in the U.S. In chapter one, I weigh trade-offs associated with delaying the implementation of control on the economic impact of controlling an FMD outbreak in the U.S. This study aimed to understand whether control policies that adopt a conservative initial approach, but may be updated as an outbreak progresses, can reduce socioeconomic harm while achieving desired outbreak outcomes. I find that delaying the implementation of all available control interventions early on in an outbreak does not reduce the cost of small outbreaks and exacerbates the largest outbreaks, suggesting that the potential benefits of this type of adaptive response may be out weighted by the risk of allowing a large outbreak to become worse. Next, I investigate how the culling of infected cattle premises, diagnostic testing, and traceback investigations impact the size of bTB outbreaks. Results from this study show improvements to traceback investigations result in the largest decreases in bTB outbreak size, which suggests that improving the identification of premises via traceback investigations is more important than increasing antemortem diagnostic sensitivity. Although this thesis focuses on the control of livestock disease, we can abstract several broader principles that contribute to ecology and epidemiology's understanding of disease dynamics. Both chapters demonstrate the importance of a population's underlying demography to determining an outbreak's overall trajectory as well as minimizing the time until detection of an infection and the time until control is implemented.