95038 research outputs found
Sort by
Tailoring Reports for Patient-Reported Health Information within Digital Health Tools: Impacts on the Quality of Values Elicitation and Clarification for Complex Treatment Decision-Making in Older Adults with Advanced Cancers
Get PDFFor older adults (≥60) with advanced cancer, treatment decisions may vary substantially from those diagnosed at an earlier stage. Often these patients feel their values and concerns are not incorporated in treatment decisions. Effective methods are needed to elicit, discuss and incorporate patient values into treatment decisions. Best-worst scaling (BWS) is a theory driven approach used in healthcare to elicit values. Tailoring health information, including question prompt lists, can encourage patients to actively participate in consultations. This dissertation addresses a research gap in understanding how tailoring summary reports based on patients’ values elicited from a BWS instrument can prepare and encourage patients to discuss their values and engage in shared decision-making (SDM). This dissertation was conducted in four stages; engagement, development, pretesting, and pilot testing of a values-clarification tool called VOICE. During engagement and development, key stakeholders engaged in concept mapping activities, resulting in consensus on 7 treatment values and 2-3 question prompts per value that are important to older adults with advanced cancer. This aim emphasized participatory engagement among diverse stakeholders and incorporation of tailoring techniques to ensure VOICE was relevant to the intended patient population. During pretesting, participants perceived VOICE to be effective, acceptable, and useful for preparing older adults with advanced cancer to engage in values-based discussions with their clinicians. Patients’ desire to assess, reassess and discuss their values must be incorporated into consultations, and patients want to work with their clinician to align their care with what matters most to them as an individual. During pilot testing, VOICE was found to be more useful in preparing patients to discuss their values than an American Cancer Society flyer, and patients had a clear preference for values-based discussions. This dissertation provided evidence that VOICE resulted in higher quality values-based discussions while highlighting the need for values-clarification training for medical students and oncologists. This dissertation advances knowledge regarding values-clarification, yet further research is still needed to better understand the influencers and contextual factors that provide facilitation or barriers to improving engagement between patients and clinicians and ensures patients values are discussed, understood, and incorporated into treatment decisions.Doctor of Philosoph
Loop Polynomials For Knots on a 3-Sphere
Get PDFWe develop a method of calculating the first loop polynomials of a knot based on the -matrix Reshetikhin-Turaev formula for the colored Jones and HOMFLY-PT polynomials. The method is based on (1) replacing the finite-dimensional and representations with Verma modules of the same highest weight, (2) identifying each Verma module with the polynomial algebra, one generator per positive root, (3) presenting each -matrix as a composition of Burau-based linear transformation of polynomial variables and a series in with coefficients being differential operators acting on the polynomials and, finally, (4) computing the traces of products of these operators by using a combinatorial trick commonly used in deriving the Feynman diagram rules. We present some computational results based on this method.Doctor of Philosoph
Diverging Reproductive Futures: Exploring Biosocial Influences on Fertility Goals and Reproductive Preferences
Get PDFThe project presented sought to investigate the biosocial influences on fertility goals and family formation in the years prior to childbearing. Fertility desires and goals are thought to be important for shaping future demographic trends and understanding the underlying patterns of reproductive health surrounding the experience of reproduction. This work will utilize a biosocial framework to understand the “upstream” behavioral and anthropological components which shape reproductive outcomes across the lifetime. This interdisciplinary project seeks to elucidate important decision-making elements for meeting family formation goals towards or away from having children. Our first aim represents one of the first to compare self-described and supplied reasons for forgoing and/or delaying children, revealing key concerns over the contemporary affordability of children and the ardent desire to be financially and emotionally stable prior to the consideration of reproduction. We synthesize qualitative interview data with our quantitative findings to illustrate how concerns and desires are conceptualized by participants, indicating that childbearing decisions are nuanced, complex, and shrouded with uncertainty. For our second aim, we put forth evidence that subjective social status influences an individual’s predisposition towards positive or negative fertility aspirations. Results indicated that higher subjective social status was associated with an increased likelihood of positive inclination towards childbearing. Our findings highlight critical differences in fertility goals based on objective and subjective measures of social status, signaling a distinct role for two individuals perceived their status. With our third aim, we explore how perceptions of the future interrelate to fertility goals using the 5-item Dark Future Scale. We document a strong association between pessimistic views toward the future and an individual’s predisposition toward negative fertility aspirations. This finding highlights critical differences in fertility goals based on perceived conditions of the world, contributing to an increased understanding of how subjective notions and narratives of societal well-being may function as a key determinant of fertility goals. Together, these findings provide a valuable snapshot into the social processes behind fertility goals among existing non-parents.Doctor of Philosoph
POLICE VIOLENCE EXPOSURE AND CARDIOMETABOLIC RISK IN BLACK WOMEN
Get PDFCardiometabolic diseases are the leading causes of death in women and disproportionately affect African American and Black women in the United States, particularly at younger ages (18-40 years) (American Heart Association, 2021). Traditional cardiometabolic risk (CMR) factors such as obesity and hypertension may not fully account for these disparities. Among African American and Black women, the intersection of race- and gender-related stress (e.g., discrimination, racism, sexism) may contribute to the disproportionate burden of CMR. While racism-related stressors such as perceived discrimination and experiences of racism have been related to higher blood pressure and sleep disturbances, few studies have evaluated how police violence may impact the health of African American and Black women and CMR in particular. Yet, African American and Black women are more likely to experience or witness police violence and worry about a family member becoming a victim of police violence than non-Hispanic White women (Alang et al., 2017). Thus, the overall goal of this dissertation is to examine associations between police violence and CMR in young African American and Black women. To address this overall goal, we first completed a scoping review to map out the definitions and conceptualization of police violence in the health literature (Aim 1, Chapter 2). Next, conducted a cross-sectional study with 59 African American and Black women (aged 18-40 years) from North Carolina to investigate the relationships between exposure to police violence and two key health aspects: cardiometabolic risk (Aim 2, Chapter 3) and pro- and anti-inflammatory markers (Aim 3, Chapter 4). Findings from this dissertation could significantly influence research and policy regarding police violence and its connection to cardiometabolic health.Doctor of Philosoph
KNOWLEDGE-DRIVEN ANTIVIRAL DISCOVERY
Get PDFOur unpreparedness for the recent SARS-CoV-2 pandemic highlighted the need for continuous research and financial investments into antiviral research prior to likely unavoidable novel viral emergence. Broad-spectrum antiviral (BSA) drugs are a promising strategy to protect against emergent viruses; however, the development of such drugs has been challenging indicating a prime opportunity for using computational technologies to accelerate discovery. In this dissertation, we emphasize the “three R’s” of modern discovery, whereby data science enables the revision, reduction and (partial) replacement of wet-lab experimental antiviral research. We take an approach that spans from the population down to the molecular level, underscoring both the interconnectedness and importance of each level in antiviral discovery. We identify key viruses with high pandemic potential, regions of the world where the next outbreaks are more likely to occur, follow-up on viral cases that have occurred since our initial study, and summarize developments that can help us prepare for such outbreaks. We hypothesize that conserved binding sites in key coronavirus proteins can be explored for the development of BSA compounds, identified such conserved binding site residues across coronaviruses and validated our hypotheses with existing experimental data. Over the course of this thesis project, we have built a curated, annotated, and publicly available database of compounds tested in both phenotypic and target-based assays against high-threat viruses, and developed a knowledge-based computational hit discovery and experimental nomination strategy. This strategy was used to identify compounds with BSA activity and we report the results of this experimental effort herein. We also applied this multi-faceted cheminformatics mining approach to build a database of helicase inhibitors and select viral helicase inhibitors that underwent experimental testing the results of which are reported here. The knowledge-based curation and database generation also enabled us to build a high quality predictive Quantitative Structure Activity Relationship (QSAR) model which we used for virtual screening of compounds, some of which we nominated as inhibitors of Marburg Virus and report the initial experimental result here. The experimental nomination strategies in this dissertation aim to revise and reduce the time and cost of wet lab antiviral research, providing a cost-effective strategy to combat the lack of funding and interest in viral diseases after the initial emergence event.Doctor of Philosoph
ENGINEERING OF A DIRECT ELECTRON TRANSFER TYPE FAD-DEPENDENT GLUCOSE DEHYDROGENASE FOR ITS APPLICATION IN ELECTROCHEMICAL GLUCOSE SENSORS
Get PDFThe development of continuous glucose monitors (CGMs) are the pinnacle achievement in wearable sensors and the gold standard for diabetes management. The advancement of currently available CGM devices which are absent from hydrogen peroxide or toxic, artificial mediators is the future of CGM technologies. Critically important to this work is the development and improvement of an enzyme to use in a 3rd generation sensor, direct electron transfer (DET)-type enzyme, Burkholderia cepacia derived glucose dehydrogenase (BcGDH). The information contained in this thesis therein details the efforts to make this enzyme suitable for its future application to construct an electrochemical glucose sensor for continuous glucose monitoring, by focusing on its beneficial and inferior features. To complete these tasks, the BcGDH enzyme must undergo improvements—through engineering strategies—to make it suitable for a 3rd generation sensor to be used in CGMs. This BcGDH enzyme was engineered to increase its specific activity towards glucose by minimizing its catalytic activity towards galactose, maltose, and xylose, alike. This enzyme was used to determine the causality of electrochemical inactivation set forth under continuous applied potential using chronoamperometry and mitigate this inactivation by engineering strategies. Enzyme engineering strategies and knowledge of electrochemical inactivation were used to create a novel biological reference electrode, a Bioreference, to enhance the stability of long term, continuous operation of a glucose sensor. This iterative improvement towards a 3rd generation sensor for CGMs showcases the ability to move towards a future sensing modality with extended sensor lifetimes and is devoid of hydrogen peroxide or toxic, artificial mediators.Doctor of Philosoph
IMPLEMENTING THE PHQ-(A) DEPRESSION SCREENING IN POST-CONCUSSION VISITS TO IDENTIFY EARLY SIGNS OF DEPRESSION IN ADOLESCENTS
Get PDFPurpose: The purpose of this quality improvement project was to implement a depression screening, the PHQ-(A), in all post-concussion visits for adolescents aged 12 to 18. After a thorough literature review, it was determined that adolescents who have experienced one or more concussions are more likely to develop depressive symptoms. They are also at an increased risk of suicidal thoughts and suicidal acts. Current guidelines do not require providers to conduct any depression screenings on post-concussion visits. The goal of this project, through early screening, is to identify and track any depressive symptoms that may develop after a patient experiences a concussion. Methods: This quality improvement project was implemented at Clinton Medical Clinic, in rural southeastern North Carolina. The project began by providing education on the project to health care providers within the practice. An overview of the PHQ-(A) depression screening in all post-concussion visits in adolescents aged 12 to 18 was given and providers were asked to record the score within the electronic medical record. The provider then determined treatment, including follow-up, medication, referral, etc. To determine effectiveness, I looked at the following measures: percentage of screenings conducted in post-concussion visits, percentage of positive screens identified, the percentage of patients who had some sort of treatment recommendation, and provider feedback on the project. Results: 16 participants with 21 post-concussion visits were included in the project. Some participants were seen multiple times for follow-up appointments. After 12 weeks, 66.67% of these visits conducted a PHQ-(A) screen. Of these screens, there was 1 positive screen identified (7.14%). Of the 14 screens completed, the following treatment recommendations were documented within the EMR: follow-up appointment (3 patients), referral to concussion clinic (1 patient), and started on an anti-depressant medication (1 patient). Providers found the project to be beneficial to their current practice, finding the screening to be efficient and sustainable. Conclusions: Based on the overall research available about depression in post-concussion adolescents, it is important to take depression into consideration when conducting post-concussion visits. The PHQ-(A) is an easy and effective way of identifying early symptoms of depression and tracking the progression of one’s depression. Depression screening should become a requirement in all post-concussion visits to help identify worsening depression symptoms and start early interventions to help alleviate the patient’s current symptoms.Doctor of Nursing Practic
Exploring “Navigational Hubs” To Improve Population Health, A Role of Governmental Public Health
Get PDFThis dissertation examines navigational hubs as a strategic approach to strengthening local public health departments (LHDs) and enhancing their ability to serve their communities effectively. The Navigational Hub model positions public health agencies as key players in care coordination, bridging critical gaps between medical services and social support systems. By adopting this role, LHDs improve access to essential health resources and establish themselves as trusted conveners—organizations capable of leading broader public health initiatives and driving systemic change.To explore this model, a mixed-methods study focused on LHDs in Texas was conducted. A survey distributed to 170 public health entities yielded 51 responses from LHDs, offering insight into their involvement in care coordination. The results revealed opportunities and challenges: 78% of respondents identified care coordination as a departmental priority, and 84% reported inadequate funding to sustain these efforts. Key informant interviews with leaders from eight health departments representing a range of geographic and demographic contexts were done to further the analysis. These interviews identified catalysts that drive care coordination efforts and the barriers that hurt progress. Departments cited disasters, community needs, strong leadership, organizational culture, innovation, and staff longevity as primary motivators for embracing care coordination initiatives. Despite this promising work, challenges remain, including the lack of sustainable funding, workforce shortages, and fragmented data systems.The findings of this research highlight the transformative potential of the navigational hub model in modernizing public health practice. By integrating care coordination into their core functions, LHDs can expand their reach, strengthen partnerships, and become leaders in addressing the social determinants of health. For this model to be successful, policymakers, healthcare leaders, and community stakeholders must recognize and actively support the evolving role of public health agencies. Investments in funding, workforce development, and data infrastructure will be crucial in ensuring that LHDs can realize their potential as navigational hubs, ultimately leading to healthier and more resilient communities.Doctor of Public Healt
Novel Methodologies in Network Analysis, Adversarial Learning, and Bandit Problems
Get PDFStatistics and machine learning are extensively utilized and widely favored across numerous domains for their ability to offer valuable insights into relationships between objects of interest and to make accurate predictions. This dissertation focuses on three key areas within statistics and machine learning: network analysis, adversarial learning, and bandit problems, exploring fundamental challenges and proposing novel solutions in each domain. Firstly, we introduce NetOTC (Network Optimal Transition Coupling), a transport-based method for the comparison and alignment of two networks. Given two networks and a cost function relating their vertices, NetOTC finds a transition coupling of their associated random walks having minimum expected cost. The minimizing cost quantifies the difference between the networks, while the optimal transport plan itself provides alignments of both the vertices and the edges of the two networks. We investigate a number of theoretical properties of NetOTC and present experiments establishing its empirical performance. The second part of this dissertation focuses on adversarial learning, exploring the geometry of the data manifold. While adversarial training has proven effective in improving the robustness of deep neural networks against adversarial attacks, it often leads to a substantial drop in accuracy on clean data. To address this trade-off, we propose a novel approach called Tangent Direction Guided Adversarial Training (TART). We argue that training with adversarial examples with large normal components significantly distorts the decision boundary, resulting in reduced accuracy. TART overcomes this challenge by estimating the tangent direction of adversarial examples and adaptively adjusting the perturbation limit based on the norm of their tangential component. Lastly, bandit problems are a fundamental setting in online learning, where a learner makes sequential decisions to maximize cumulative rewards. This dissertation explores two previously unaddressed bandit problems. The first is the dueling bandit problem with stochastic delayed feedback. Real-world applications frequently face unavoidable stochastic delays in feedback, presenting significant challenges for existing dueling bandit algorithms. To address this, we propose two delay-resilient algorithms. The second is the Quantum Lipschitz Bandit problem, where the expected reward function satisfies a Lipschitz condition over an arm metric space. Motivated by advances in quantum computing and the success of quantum Monte Carlo methods, we introduce the first quantum Lipschitz bandit algorithms. Extensive experiments validate our theoretical improvements, showcasing the superior performance of our methods.Doctor of Philosoph
Essays on Econometric Modeling of Forecasting, Uncertainty, and Price Impact
Get PDFThis dissertation develops new econometric methods and empirical applications in macroeconomicforecasting, economic uncertainty measurement, and decentralized finance (DeFi) marketmicrostructure.In the first chapter, we introduce a heterogeneous coefficients model within the sparse-groupLASSO MIDAS framework to address cross-sectional heterogeneity in mixed-frequency paneldata regression. By allowing coefficient similarity within predefined economic groups whilemaintaining flexibility where needed, our model provides a structured compromise between fullypooled and fully heterogeneous specifications. We apply this methodology to macroeconomic panelforecasting, demonstrating its advantages in capturing meaningful heterogeneous relationships whilemitigating biases introduced by overly restrictive pooling assumptions. Our results highlight thetrade-off between model flexibility and generalizability in high-dimensional economic forecastingapplications.In the second chapter, we develop novel measures of economic uncertainty at the state andextended metropolitan area (EMA) levels, leveraging forecast errors from GDP growth models.Using a sparse-group LASSO MIDAS regression framework, we estimate state-level uncertaintyand construct an EMA-level uncertainty measure to capture economic volatility at different regionalscales. Our empirical analysis identifies key macroeconomic drivers of uncertainty, including policyinstability, industry composition, and financial conditions. We further validate our EMA uncertaintymeasure by comparing it with established uncertainty indices, demonstrating its effectivenessin capturing crisis-driven fluctuations. These findings contribute to the broader literature onmacroeconomic uncertainty and regional economic dynamics.In the third chapter, we examine the price impact of trades on decentralized exchanges (DEXs)using Uniswap as a case study. Despite the deterministic nature of automated market makers,realized price impact remains uncertain due to block inclusion and transaction ordering mechanisms.We quantify the information advantage of Miner Extractable Value (MEV) traders, who exploitpending orders for arbitrage. To address limitations in traditional vector autoregression (VAR)models of price impact, we propose a novel empirical framework that accounts for differential priceeffects based on trader type. Our findings provide strong evidence that MEV traders play a dominantrole in price discovery on DEX markets, contributing to the literature on market microstructure indigital asset markets.This dissertation advances econometric methodology for high-dimensional panel regressions,provides new insights into economic uncertainty at regional levels, and offers empirical evidence onprice formation in DeFi markets. The findings have implications for macroeconomic forecasting,regional economic policy, and digital asset market analysis.Doctor of Philosoph