22689 research outputs found
Sort by
FUJITA-TYPE RESULT AND SELF-SIMILAR BLOW-UP BEHAVIOR IN A SEMILINEAR HEAT EQUATION WITH DEGENERATE COEFFICIENTS
Get PDFIn this thesis, we study two problems on a non-homogeneous parabolic equation with degenerate coefficients of the form u_t − Lωu = u^p, where L_ω = ω^{−1}div(ω∇).
The first problem is to establish the existence/non-existence of global-in-time mild solutions according to a critical exponent, known as the Fujita exponent. A similar topic for another degenerate semilinear heat equation with non-self-adjoint operator L_ω = div(ω∇) is treated in [10]. They considered two types of homogeneous weights: ω(x) = |x1|^a and ω(x) = |x|^b where a,b > 0. We extend the problem with self- adjoint operator to more general weights that meet certain restrictions, such as being in the Muckenhoupt class A_2, being non-decreasing, and having the limits α := lim_|x′|→∞(log ω(x))/(log |x′|) and β := lim_|x′|→0(log ω(x))/(log |x′|) exist, where x′ = (x1,...,xn) and 1 ≤ n ≤ N. The main result establishes that the Fujita exponent is given by p_F = 1 + 2/(N + α). This implies that the asymptotic behavior of the weight at infinity affects global existence of solutions and the one at the origin does not.
The second problem focuses on the pointwise characterization of the asymptotic behavior of local solutions near blow-up points as t approaches the blow-up time. This is examined under the condition 1<p≤p_S (i.e. 1<p≤ N+α+2 or N+α ≤ 2), where p_S + 1 aligns with the critical exponent for Sobolev embedding involving a weighted measure. Assuming the existence of a weak solution that meets the upper blow-up rate estimates and is singular at t = 0, we demonstrate that the scaled solution approaches (p − 1)−1/(p−1) near blow-up. The analysis leverages the scalar property of weights in the Muckenhoupt class A_2 and being homogeneous. Moreover, the methodology can be extended in a similar manner to the equation with non self-adjoint operator L_ω = div(ω∇).
Combining insights from both problems, we conclude on the existence of global-in-time mild solutions and the asymptotic behavior of local-in-time solutions near blow-up, contingent on the initial data and the critical exponents p_F and p_S
ADAPTATION AND IMPLEMENTATION OF A HYPERTENSION EXTENSION FOR COMMUNITY HEALTHCARE OUTCOMES PROGRAM FOR COMMUNITY HEALTH EXTENSION WORKERS WITHIN THE FEDERAL CAPITAL TERRITORY, NIGERIA
Get PDFIntroduction: Health systems strengthening and capacity building are critical to meet the needs of a growing and aging population in Nigeria. To inform scaling and sustainability of hypertension services within primary healthcare centers (PHCs), we adapted, implemented, and evaluated a hypertension education program for community health extension workers (CHEWs) in the Federal Capital Territory, Nigeria following the Extension for Community Healthcare Outcomes (ECHO) model.
Methods: This dissertation contains three manuscripts conducted under the ADAPT-ITT framework. The first assessed feasibility, acceptability, perceived needs, and contextual factors through qualitative research to guide implementation of the hypertension ECHO series. The second utilized convergent parallel mixed methods to evaluate Kirkpatrick and implementation outcomes. The third is a quasi-experimental evaluation of patient-level hypertension outcomes using concurrent data from 33 PHCs in the Hypertension Treatment in Nigeria Program, from which 12 were selected to participate in the ECHO series.
Results: We identified barriers and facilitators for the hypertension ECHO model and confirmed its suitability for CHEWs. We delivered a seven-part hypertension ECHO series between August 2022 and April 2023 to 1407 total live participants. Reaction to each session was positive and knowledge based on pre- and post-quizzes increased among participants (range: 2.3% to 10.5% relative to baseline). After each session, over 70% of healthcare workers reported applying learning to their clinical care. In follow-up focus groups discussions, healthcare workers identified that network connectivity and clinical demands were barriers to live participation and expressed preferences for blended training and asynchronous resources. Baseline treatment rates were similar between PHCs (ECHO: 94.0% and Non-ECHO: 94.7%) and increased at a higher rate (interaction p-value=0.045) in the ECHO cohort over time. After adjustment for baseline and within site variation, the difference was attenuated (interaction p-value=0.37). No adverse effects were observed on staffing or clinical capacity.
Conclusion: We found positive results following implementation of the ECHO series at all four levels of the Kirkpatrick framework as well as acceptability, appropriateness, and feasibility among CHEWs in the Federal Capital Territory, Nigeria. These results can inform strategies for scaling hypertension services, and implementation of the ECHO model for CHEWs in other geographies and domains
Localized Professional Development: Creating Base-Wide Solutions to a Military-Wide Problem
No full textThis paper examines and synthesizes the literature surrounding professional development for Airmen in the United States Air Force utilizing Bronfenbrenner’s (1977) Ecological Systems Theory. Existing literature is categorized into Boshier’s (1991) seven motivations to pursue voluntary education and are explored at the societal, military, United States Air Force, and individual Airmen levels. The known perceptions of professional development within the United States Air Force are compared to the potential motivations to pursue professional development and culminate in a conceptual framework for motivations and perceptions of professional development at Freedom Air Force Base. A quantitative study utilizing Boshier’s (1991) Education Participation Scale was administered to Airmen at Freedom Air Force Base who attended a professional development seminar and showcased six of the seven known voluntary education participation motivations. All six motivations are not currently incorporated into the planning and facilitation of professional development programs within the United States Air Force. The results of this study were shared to military education practitioners and military leaders through publication with the Marine Corps University Press and presented during a military education forum. Additional research is recommended across multiple United States Air Force bases and with a larger sample size to validate the results
EXPLORING ANTISENSE OLIGONUCLEOTIDE THERAPY TARGETING PROTEIN TYROSINE PHOSPHATASE RECEPTOR TYPE S IN C9ORF72-ALS/FTD
Get PDFAmyotrophic Lateral Sclerosis (ALS) and Frontotemporal Dementia (FTD) are fetal neurodegenerative diseases that share clinical, pathological, and genetic characteristics. The hexanucleotide repeat (G4C2) expansion in the first intron of chromosome 9 open reading frame (C9ORF72) gene is the most common genetic cause of ALS and FTD. Translation of sense and antisense G4C2 repeat expansion produces neurotoxic dipeptide repeat (DPR) proteins. Among the various dipeptide repeat proteins, poly-GR and PR are relatively more toxic to neurons, positioning poly-GR and PR as a potential target to mitigate neurotoxicity and enhance neuron survival. Our previous CRISPRi screening in human cortical neurons (i3Neurons) identified protein tyrosine phosphatase receptor type S (PTPσ) as a potent modifier of poly-GR mediated toxicity in neurons, whose knockdown ameliorated glutamate-induced neuro excitotoxicity in C9ORF72-ALS/FTD patient-derived motor neurons (C9ORF72-ALS/FTD iMN). As antisense oligonucleotide (ASO) is an established method for gene targeting in neurodegenerative diseases, I aim to develop ASOs to knock down PTPσ and examine their rescue efficacy in GR-expressing i3Neurons and C9ORF72-ALS/FTD iMNs. I found that ASO candidates can reduce the neurotoxicity caused by poly-GR and rescue the glutamate-induced excitotoxicity in C9ORF72-ALS/FTD iMNs. The findings in this study provide insights on the development of future therapeutic strategies in ALS and FTD
Association Between Self-Reported Cardiovascular Disease and Exposure to Metals from E-Cigarette Use: A Longitudinal Analysis of the PATH Study
No full textBackground: The prevalence of e-cigarette use has been on constant rise, but limited studies have documented how metals exposure from e-cigarette use contributes to cardiovascular disease (CVD) outcomes.
Objectives: Our study aimed to assess the association between self-reported cardiovascular disease and exposure to metals from e-cigarette use in a national longitudinal study.
Methods: We used data from wave 1 to wave 5 from The Population Assessment of Tobacco and Health (PATH) study, a nationally representative cohort. A total of 4,824 participants were included to assess the association between e-cigarette use and CVD. Data from a subset of 1,124 participants with biomarkers were used to test the association between metal exposure and CVD. Metal concentration levels were compared among different tobacco use groups stratified by CVD outcomes. Each metal was divided into four groups based on the quartiles of concentration (Q1, Q2, Q3 and Q4). Cox models were used to estimate hazard ratio (HR).
Results: Adjusted model results indicated that exclusive cigarette use is associated with elevated level of combined CVD (HR=1.40, 95% confidence interval (CI): 1.21, 1.61), high blood pressure (HR=1.26, 95% CI: 1.02, 1.55) and high cholesterol (HR=1.32, 95% CI: 1.04, 1.68). E-cigarette users have higher urine lead (geomean=0.60, 95% CI: 0.42, 0.87) than cigarette users (geomean=0.44, 95% CI: 0.40, 0.47) and non-users (geomean= 0.34 ,95% CI: 0.31, 0.38). Among e-cigarette users, the cadmium geometric mean for combined CVD cases was about three times the geometric mean of those participants without CVD, and two times for uranium. Cadmium, cobalt, lead, and strontium were found to have no significant independent association with CVD after adjusting for confounders. Uranium was related to elevated risk of high-cholesterol (HR=1.36, 95% CI: 1.00, 1.85) in a dose-response manner. Thallium was associated with higher risk of high blood pressure (HR=2.53, 95% CI: 1.48, 4.31 for Q2 group), and lower risk of cholesterol (HR=0.38, 95% CI: 0.17, 0.90 for Q3 group).
Conclusion: Our results confirm elevated metal exposure from e-cigarette use, and in CVD cases, and also show potential association between CVD and exposure to metals
Supporting Public Health Decision-Making through an Engineering Perspective: From Pandemic Data Acquisition to Data-Driven Modeling
Get PDFThe COVID-19 pandemic has highlighted the vital need for reliable publicly available health data and data-driven models to support policymaking. This thesis presents an engineering perspective to address this need. The study starts with the introduction of the Johns Hopkins University CSSE COVID-19 dashboard, a crucial tool developed to track the pandemic’s progression and provide real-time information to researchers, policymakers, and the general public. This dashboard has been widely used worldwide, with over 228 billion feature layer requests by March 10, 2023. Building upon this foundation, the thesis employs statistical models to analyze vaccination rates and network tools to characterize online misinformation diffusion patterns. The analysis of COVID-19 vaccination rates at the U.S. county level identifies key socioeconomic, demographic, and political factors associated with vaccine uptake, with political affiliation, determined by the proportion of Trump supporters in the 2020 presidential election, having the strongest association with the percentage of the population not getting the COVID-19 vaccine. The thesis also examines changes in routine childhood immunization coverage after the COVID-19 pandemic, revealing a nationwide decline in measles-mumps-rubella (MMR) vaccination rates. We collected county-level historical MMR immunization data for kindergarten children, implemented statistical models, and identified factors such as higher education levels and the presence of non-medical exemption policies as positively associated with low MMR vaccination rates. Furthermore, the thesis explores the information diffusion dynamics of COVID-19 vaccination-related tweets based on their political ideological category and credibility on X (previously known as Twitter), finding that tweets from right-leaning influencers and low-credibility domains propagate faster. Using novel metrics we designed, we also find that accounts sharing low-credibility information tend to be less diverse in their sources and more likely to have unidirectional information flow from content creators to reposters. The data-driven efforts and models presented in this thesis contribute to a deeper understanding of public health needs, trustworthy data aggregation, vaccination behavior, and health information dynamics, carrying significant implications for mitigating the impact of public health emergencies and developing evidence-based public health policies and strategies
Computational Improvements To and Applications Of the Substitution Method for Site Percolation Threshold Bounds
Get PDFThe substitution method has proved to be a very successful tool for bounding site percolation thresholds of Archimedean lattices. Previous improvements to the substitution method algorithm had enabled more accurate, and thus tighter, bounds while reducing computational complexity. Despite the success of the algorithm improvements, severe constraints remained that restricted the size of feasible substitution regions. Chief among these constraints were round-off errors in the multiplication of large primes in intermediate steps, physical computation time, and memory capacity.
The constraints are relaxed through modifications of the algorithm that eliminate the need for identifiers in particular intermediate steps, decrease computation time, and improve the memory capacity. Specifically, revisions to the intermediate steps were made which leveraged group-theoretic concepts instead of identifiers and enabled parallel computing. Heuristically, larger substitution regions should yield tighter bounds since more upsets are being compared. Furthermore, there are simply more possible configurations to be considered in larger regions.
These assumptions are confirmed for the square lattice through the application of the modified algorithm. In particular, the upper bound for the site percolation threshold of the square lattice was decreased from to , providing the first improvement since 1995. This bound was then further improved to which is the current lowest upper bound. To achieve these improvements, the site percolation model on a self-matching lattice was compared to the square lattice site percolation model in a two-stage process
A RECIPE FOR PROMOTING HUMAN & PLANETARY HEALTH: EXPLORING PLANT-FORWARD FOOD PROCUREMENT IN UNIVERSITY DINING
No full textIntroduction: Transforming food systems to deliver healthy diets from sustainable food production is a critical strategy for concurrently mitigating climate change and improving public health. A growing number of institutions have committed to align their food procurement with the EAT-Lancet Planetary Health Diet to improve public health and fight climate change. The university sector is of particular interest given their significant purchasing power, increasing focus on advancing health and sustainability initiatives, and opportunities to integrate health and environmental sustainability across the campus foodscape more broadly.
Methods: This dissertation utilized quantitative and qualitative methods to evaluate alignment of university food purchases with the EAT-Lancet Planetary Health Diet and explore factors that may influence plant-forward procurement to guide more effective food policies, programs, and metrics. Annual procurement data from 19 universities in the Menus of Change University Research Collaborative (MCURC) was used to evaluate alignment of current food purchases with EAT-Lancet guidelines, and to model changes in food-related greenhouse gas emissions, Healthy Eating Index (HEI), and food costs associated with shifting food purchases to align with the Planetary Health Diet. Dining staff from these universities were surveyed to understand perceptions of which factors have the greatest influence on food purchasing decisions, and two-sample t-tests were used to compare procurement outcomes between universities with differing characteristics. In-depth interviews with key informants from university dining programs were used to generate a conceptual model representing the multidimensional system of factors influencing plant-forward food procurement in university settings.
Results: In this sample, university food purchases were generally not aligned with planetary health targets. Food costs and budgets were perceived to be the most important factors influencing food purchases. Additional factors may impact plant-forward food purchases including institutional nutrition and sustainability targets, dining staff support for plant-based foods, and student demand for meat. In-depth interviews revealed many interacting influences across dining program, institutional, student/customer, and supply chain domains that can promote or hinder plant-forward procurement.
Discussion: This study offers formative insights for promoting plant-forward procurement in universities that can support university stakeholders in designing more effective food policies and purchasing targets that integrate planetary health guidelines
AN INVESTIGATION OF ATTITUDES AND EMOTIONS OF COVID-RELATED CHATBOT USERS
Get PDFThis thesis evaluates the impact of a digital chatbot on public health communication during the COVID-19 pandemic. This chatbot aims to counter misinformation and shape public sentiment regarding vaccines. The study analyzes 45,422 user interactions using advanced analytic techniques, including thematic analysis via Latent Dirichlet Allocation (LDA) and sentiment analysis with the GPT-3.5 API. It examines sentiment trends over time using spline models and employs permutation tests to determine whether observed changes in sentiment are statistically significant or merely due to random variation.
The results indicate that regular interaction with the chatbot correlates with a positive shift in sentiment, suggesting that continued engagement enhances trust in the provided information. This study integrates analysis of sentiment dynamics with user engagement levels, exploring the relationship between repeated exposure and sentiment improvement. By assessing temporal patterns and user demographics, the research highlights how digital tools can effectively support public health strategies during crises. The findings demonstrate that the chatbot successfully fosters a more informed and positively engaged community, underscoring the potential of such technologies in health communication. This thesis not only highlights the role of artificial intelligence in health communication but also provides insights into public engagement and sentiment evolution in response to targeted health communication during a global health crisis
THE PREVALENCE OF ANTIBIOTICS PRESCRIPTIONS FOR PATIENTS DIAGNOSED WITH COMMON COLD IN THE EMIRATE OF ABU DHABI, UAE: AN EMPIRICAL STUDY
No full textStatement of the problem. The pattern of prescribing of antibiotics for outpatient visits in the Emirate of Abu Dhabi, United Arab Emirates (UAE), is of great concern. The study will help show the behavior of prescribing antibiotics, identify areas for improvements to enhance the quality of antibiotic prescribing and raise awareness to the government and healthcare leadership of Abu Dhabi.
Methods. Frequencies, bivariate and multivariable logistic analyses were used to analyze all 339,417 encounters with patient diagnoses of Common Colds. The source of data was the Department of Health claim data for the year 2018.
Summary of results. Multivariable logistic regression analysis results revealed substantial variations among patients, physicians, healthcare facilities, and comorbidity factors. Patients with comorbidity had a much higher chance of receiving antibiotics with an odds ratio about 19.11 times more likely than patients diagnosed with a Common Cold only (OR 19.11, 95% CI 17.11–21.36). Moreover, patients treated in the private sector were slightly more likely to receive antibiotics than patients treated in the public sector (OR 1.10, 95% CI 1.07–1.13). For facilities locations, antibiotics prescribing increased 25% in Al Ain and 77% in Al Dhafra compared to Abu Dhabi (Al Ain - OR 1.25, 95% CI 1.22–1.28), (Al Dhafra - OR 1.77, 95% CI 1.67–1.87). For female patients, antibiotics prescribing decreased slightly by 5% when compared to male patients (OR 0.95, 95% CI 0.93–0.98). Older patients were more likely to receive antibiotics prescriptions when compared to the reference age group of 18-30. Compared to General Practitioners, antibiotics were more likely to be prescribed by physicians in Internal Medicine, Otolaryngology, and Emergency Medicine. Out of the total 26,062 antibiotics drugs prescribed, 89.60% were broad spectrum. Co-amoxiclav and Azithromycin were the most frequent antibiotic drugs prescribed.
Conclusions. Antibiotic resistance increases morbidity and mortality for patients as well as posing a threat to public health. Interventions and antibiotics stewardship policies should be developed by regulatory bodies such as the Department of Health to influence and improve antibiotic prescribing in the Emirate of Abu Dhabi