22689 research outputs found
Sort by
Investigating the Relationship between Age-related Hearing Loss and Loneliness among Community-dwelling Older Adults
No full textAddressing loneliness is a public health priority given mounting evidence of associated morbidities, including mortality and adverse health outcomes. One risk factor, hearing loss (HL), affects approximately 63% of all older adults in the U.S. and total case numbers are projected to increase by an estimated 70% over the next four decades. The proliferation of HL may exacerbate associated sequelae of loneliness. As a potentially modifiable risk factor, understanding HL’s relationship with loneliness may support developing effective response strategies.
Through a sequential explanatory mixed methods design, this dissertation aimed to propose an explanation for the relationship between HL and loneliness in an older population sample to inform future research. The quantitative study investigated the association between objectively measured HL (audiometric HL) and loneliness by examining the joint effects of audiometric HL with hearing handicap. The qualitative study collected theory-driven interviews following guidelines for naturalistic inquiry and descriptive thematic analyses. Integration/mixing evaluated qualitative themes along the respective communication statuses of informants.
Secondary analyses of a population dataset comprising older adults with HL and normal/typical hearing without hearing handicap as the referent (N = 2,527) observed that the joint effects models showed up to 95% higher odds of loneliness. Ten purposively sampled informants provided narrative data that produced themes organized across 13 theoretical domains, including six novel ones. A themes-by-statistics joint display revealed a pattern whereby descriptions and perceptions by older adults with combined audiometric HL and hearing handicap remarkably differed from their counterparts. Overall findings identified some shared experiences as people with HL, though older adults who endorsed hearing handicap described distinct experiences displaying functional/behavioral challenges within select settings as compared to counterparts with audiometric HL only.
This dissertation responds accordingly to recommendations published by the National Academies of Sciences, Engineering, and Medicine (2020) calling for further studies of loneliness in older populations to inform public health strategies. The integrated data identifies potential target points for future research. Subsequent confirmatory studies may consider building upon the reported findings through adapted study designs. Study purposes should prioritize supporting actionable next steps for promoting healthy aging and health equity in a diverse society
METAL BIOMONITORING USING LOW-MASS TOENAIL SAMPLES: APPLICATIONS IN ENVIRONMENTAL EPIDEMIOLOGY
Get PDFBackground: Understanding the relationship between metals and disease relies, in large part, on evidence from human epidemiologic studies and thus the accurate characterization of environmental metal exposures. Metal biomonitoring is critical to this effort as it characterizes internal dose integrated across multiple sources and routes of exposure. As technological advancements have made it easy to analyze a wide range of metals in a single biomarker matrix, new questions have been raised about the suitability of various matrices for such multi-elemental analysis. Toenail is a promising matrix, but further research is needed to standardize the analytical process and interpret metal concentrations measured in this matrix.
Objectives: The objectives of this dissertation are two-fold: 1) to overcome some of the limitations of existing metal biomonitoring strategies using a low-mass toenail metal biomarker; and 2) demonstrate potential applications and highlight limitations of this biomonitoring method in environmental epidemiology.
Methods: We used toenail samples, residential information, and questionnaire data from a sample of 413 non-smoking men across 4 US Gulf States from the Gulf Long-Term Follow-Up (GuLF) Study. In Aim 1, we demonstrate the suitability of a low-mass toenail sample for multi-elemental metals analysis. In Aim 2, we apply this biomonitoring method for industry emissions-related metal exposure assessment. In Aim 3, we use this method to assess metal exposure related neurobehavioral impairment in a cross-sectional neuroepidemiology study within a mixtures framework.
Results: In our low-mass toenail reliability study (Aim 1), we showed strong agreement among triplicate toenail sub-samples and demonstrated the suitability of a low-mass (roughly 1 – 2 clipping) toenail sample for ICP-MS metals analysis. In Aim 2, we showed that this biomonitoring method was effective in identifying ambient metal exposure trends in this population. In Aim 3, we showed that metals measured in the toenail were significantly associated with attention and memory performance in GuLF Study participants.
Conclusions: This dissertation contributes information towards the standardization of toenail metals analysis and highlights considerations for biomarker selection and interpretation for multi-elemental analysis. Our findings add to the toenail validation literature and support the continued and intentional use of toenail biomarkers in environmental research
NEXT-GENERATION POLYMERIC PARTICLE DRUG DELIVERY SYSTEMS FOR CELLULAR AND IMMUNE MODULATION
No full textPolymeric nano and micro particles have demonstrated potential to improve cellular delivery of biologic therapeutics for a range of diseases including peripheral artery disease and cancer. However, few of these particle-based systems are approved for clinical use. Because of this, there is a need to improve polymeric delivery vehicles to create more effective strategies for tissue engineering and therapies for cancer. Some of the main challenges in the approval and effectiveness of existing particle-based drug and gene delivery systems are that they have poor bioavailability at the site of action, significant off target effects, and are difficult to manufacture and scale up. Potential avenues to improve upon existing polymeric particle technologies include the development and study of novel biomaterials and biomaterial-cell interactions, optimization of particle platforms for targeting specificity using active ligands, and reduction of manufacturing and scale up difficulty by expansion of the routes of particle administration and translation of current ex vivo therapies to in situ therapies.
The goal of this thesis is to advance the field of polymeric particle drug delivery systems by creating new material systems to target stem cells, immune cells, and cancer, exploring mechanical properties of particulate interactions with the immune system, and expanding the routes of delivery of PBAE polymeric particle drug delivery systems. Chapter one summarizes the specific aims of this thesis. Chapter two provides a comprehensive review of the literature in the field of polymeric drug delivery. Chapter 3 describes the development of a DNA delivery system for engineering primary human vasculature. Chapter 4 and 5 describe the development of a PEG particle platform for in situ expansion of CD8+ T cells. Chapter 6 describes the development of a PBAE mRNA delivery system for in situ genetic engineering of primary T cells. Finally, in Chapter 7 I detail my work in translating PBAE gene delivery systems from intravenous to oral delivery. Overall, I have developed both PBAE and PEG particle platforms that have enabled precise engineering of vascular structures, elucidated the effects of aAPC mechanical properties on T cell activation, and created a novel PBAE aAPC particle platform for gene delivery to T cells
Prostate Cancer Screening and Incidence in Men with HIV Compared to Men Without HIV
Get PDFBackground
It has been observed that prostate cancer (PCa) incidence in men with HIV is lower than in men without HIV. It is not known if this observation is attributable to differences in screening, detection, or underlying biology.
Methods
We described incidence of prostate-specific antigen (PSA) test and PCa in Medicaid beneficiaries with and without HIV enrolled in 14 US states in 2001-2015. We compared PSA testing by HIV status using Poisson regression. We also estimated models stratified by state and pooled estimates using random effects meta-analysis to account for heterogeneity in PSA testing by state. We estimated adjusted cause-specific and sub-distribution hazard ratios using Cox regression and Fine and Grey models, respectively, to compare PCa incidence by HIV status among Medicaid beneficiaries in 14 US states in 2001-2015. Models estimated among Medicaid beneficiaries were also stratified by age and race-ethnicity. Lastly, we evaluated factors for association with PSA test among men with HIV enrolled in the Johns Hopkins HIV Clinical Cohort (JHHCC) in 2000-2020 using Poisson regression.
Results
Men with HIV received PSA test more than men without HIV (pooled IRR=1.11, 95% CI: 0.97 1.27). Similar results were observed in models stratified by age and race-ethnicity. PCa incidence was lower in men with HIV compared to men without HIV (csHR=0.89; 95% CI: 0.80, 0.99), but this association varied by race-ethnicity, with PCa incidence in men with HIV lower than men without HIV among non-Hispanic Black (csHR=0.79, 95% CI: 0.69, 0.91) and Hispanic (csHR=0.85, 95% CI: 0.67, 1.09), but not non-Hispanic White men (csHR=1.17; 95% CI: 0.91, 1.50). Men with greater engagement in HIV and other healthcare were more likely to receive PSA test among JHHCC participants.
Conclusions
PCa incidence was lower among men with HIV compared to men without HIV despite greater receipt of PSA test among men with HIV, suggesting differences in incidence are not due to differences in screening. Further research among men who have received screening is needed to identify where differences in detection or underlying biology of PCa may exist
MAKING DECISIONS ON THE FLY - UNRAVELLING THE COMPUTATIONAL PRINCIPLES GOVERNING CHOICE BEHAVIOR IN THE DROSOPHILA MELANOGASTER BRAIN
Get PDFForaging animals adopt decision-making strategies that successfully adapt to the dynamically changing availability of rewards in the environment, as well as account for the complexity of the sensory world around them. Understanding the neural principles that underlie these complex decision strategies has been an area of interest in neuroscience for a long time. While a lot of progress has been made on these fronts, a complete picture of these neural algorithms has eluded us for two reasons. In studies of larger vertebrates that have used tasks that account for richness of the natural world, the lack of manipulability of these complex brains has restricted our ability to map the underlying neural algorithms. On the other hand, studies in more accessible small brains have been limited to simple Pavlovian learning tasks for the most part. In this dissertation, my colleagues and I develop a novel foraging task for Drosophila melanogaster, the details of which are described in chapter 2, and leverage this task to provide insight into the neural algorithms underlying decision-making.
In chapter 3, we deal with dynamic probabilistic environments, where several animals are known to distribute their choices in proportion to the rewards received from available options - Herrnstein’s operant matching law. Theoretical work suggests an elegant mechanistic explanation for this ubiquitous behavior, as operant matching follows automatically from simple expectation-based synaptic plasticity rules acting within behaviorally relevant neural circuits. However, no past work has mapped operant matching onto plasticity mechanisms in the brain, leaving the biological relevance of the theory unclear. Here we discovered operant matching in Drosophila and using a combination of behavior, computational modeling and optogenetics showed that it requires reward expectation based synaptic plasticity that acts in the mushroom body. Our results reveal the first synapse-level mechanisms of operant matching in the fly brain.
In chapter 4, we provide the first biological test of Marr and Albus’ expansion layer theory about sensory discrimination. In particular, the ratio between sensory channels and expansion layer neurons and the number of sensory inputs that individual expansion layer neurons receive are theorized to be key parameters. Leveraging the development of tools that manipulate these parameters in the context of the fly mushroom body, we show that fly behavior agrees with many theoretical predictions. An increase in expansion layer neuron number improves discrimination of odors while an increase in input connectivity causes worsened discrimination. There are however some key differences, suggesting that theoretical models can learn from the experiments and be modified to better explain sensory discrimination.
In chapter 5, we move beyond just sensory discrimination and tackle the question of flexible decision-making that must depend on the available options. We find that Drosopihla melanogaster can successfully learn to both distinguish between very similar stimuli and generalize across cues. Rather than forming memories that strike a balance between specificity and generality, we find that flies flexibly categorize a given stimulus into different groups by performing a side-by-side comparison over time of the available options. Together the work in this thesis combines multiple important avenues of neuroscience, providing insight into the neural principles underlying decision-making in dynamic and sensory rich environments
APPLICATIONS OF TRADITIONAL EPIDEMIOLOGIC AND NOVEL INFODEMIOLOGIC APPROACHES TO MONITORING CANNABIS AND PATTERNS OF ITS USE IN THE US POPULATION: FINDINGS FROM THE NATIONAL HOUSEHOLD SURVEY ON DRUG USE AND HEALTH (NSDUH) AND REDDIT
No full textBackground
Rapid expansion of access to cannabis through Medical (MCL) and Recreational Cannabis Laws (RCL) in the United States has contributed to an emerging and diverse cannabis market. Due to logistical and methodological challenges in deploying large epidemiologic surveys to monitor changes in the population cannabis use behaviors amid this period of market expansion, social media data may serve as a complimentary source of insight into emergent behaviors. Therefore, this dissertation utilizes both traditional epidemiologic data and data generated on the social media site Reddit to explore trends in daily/near-daily use patterns in portions of the population who use cannabis.
Methods
Data from the National Survey on Drug Use and Health (NSDUH) and Reddit were used for analyses. A nationally representative sample of young emerging adults (18-20 years old) who reported using cannabis in the past 12-months from the 2013-2019 NSDUH were examined for changes in prevalence of current and daily/near-daily cannabis during the study period. Logistic regressions were used to estimate linear trends in cannabis use frequency over time in the overall population, as well as within and between subpopulations.
Publicly available submissions made by individuals between January 1, 2010 and December 31, 2019 were extracted from 18 cannabis-centric communities across. A lexicon approach was used to identify submissions containing mentions of specific cannabis product, mode, and frequency concepts. Prevalence of concepts within each category were estimated overall and by community. Logistic regression models were used to estimate linear trends in concept prevalence. In addition, qualitative coding and thematic analyses explored reported histories of daily cannabis use and negative effects among a subpopulation of Redditors.
Results
Trend analyses from NSDUH indicated there was no statistically significant increase in current or daily/near-daily cannabis use from 2013-2019 in the young EA population. However, among subpopulations NH Black young EA there was a statistically significant increase of 0.16 points (p=0.008) annually in the prevalence of current use.
In analysis of more than one million submission made on Reddit between 2010-2019, flower products were the most mentioned product, though non-THC cannabinoid mentions increased significantly and sharply starting in 2015. Daily use frequency was also increasing significantly over the period. Additionally, community level differences were found in how products, modes and daily use clustered together.
Lastly, the qualitative content analysis of three cannabis communities revealed that Redditors disclose important information about their patterns and histories of daily cannabis use, while also reporting negative effects associated with their use. However, the detailed disclosure of identifying information, patterns, and negative effects were notably different by community.
Conclusions
The work of this dissertation highlights the burden of daily cannabis use in the population, as well as important shifts in the patterns of cannabis use discussed on online forums. Furthermore, individuals who report using cannabis daily are also reporting negative effects associated with their daily use. These findings present opportunities to utilize social media data to improve applications of computational methods in the field of infodemiology, as well as integrate findings from social media to develop future survey measures to capture clinically meaningful difference in patterns of use. Lastly, this work has implications for future public health prevention programs including development of interventions for those already engaged in use, such as the use of a harm reduction framework for universal prevention education.
tion
Integrating numerical modeling and deep learning for stochastic microstructure reconstruction and multiscale mechanics
No full textDesigning materials with targeted properties requires efficient modeling of material behavior across different scales. Two key components to facilitating microstructure-resolved material modeling and design methods are (i) stochastic microstructure reconstruction and (ii) microstructure-resolved multiscale modeling for property prediction. However, conventional numerical approaches face significant challenges in the practical application of design, optimization, and uncertainty quantification. In this dissertation, an integrated machine learning (ML) and physics-based modeling framework is developed for 2D & 3D microstructure generation, micro-scale stress field prediction, and multiscale mechanics.
Conventional stochastic microstructure reconstruction approaches are prohibitively slow and limiting for complex microstructure systems. Therefore, we framed microstructure reconstruction as a gradient-based optimization problem. In this approach, statistical descriptors and feature maps from a pre-trained deep convolutional neural network (CNN), are combined into an overall differentiable loss function. This approach is applied for efficient 3D reconstruction of bi-phase porous ceramic and multi-phase polycrystalline materials.
Microstructural heterogeneity affects the macroscale behavior of materials. Conversely, loading at the macro-scale affects material behavior at the micro-scale. These up-scaling and down-scaling relations are often modeled using multiscale finite element (FE) approaches such as FE-squared (FE2). However, (FE2) requires numerous calculations at the micro-scale, which often renders this approach intractable. To address this, we developed an enormously faster ML-driven approach for multiscale modeling. This approach uses an ML-model, specifically a U-Net CNN, to predict stress in linear-elastic fiber reinforced composite materials. This ML-model is integrated in a discretization based multiscale approach to predict effective material properties for up-scaling and local stress tensor fields for subsequent down-scaling. Several numerical examples demonstrate a substantial reduction in computational cost using the proposed ML-driven approach when compared with the traditional multiscale modeling approaches such as full-scale FE analysis, and homogenization based (FE2) analysis
THE ELICITATION OF AND RESPONSE TO GUILT AND SHAME IN GENETIC COUNSELING COMMUNICATION
No full textGuilt and shame are emotions that may arise in genetic counseling sessions; however, little is known about the way genetic counselors elicit and respond to client disclosure of guilt and shame. The purpose of this mixed methods study was to qualitatively describe the context and communication dynamics of guilt and shame in simulated genetic counseling sessions and quantitatively assess session-level simulated client and genetic counselor communication patterns associated with the disclosure and depth of conversation surrounding guilt and shame. Finally, we qualitatively characterized dialogue preceding disclosure. This study was informed by the social cognitive processing model (SCPM), emotion theory and a sequential approach to studying patient-provider interactions. The data for this secondary analysis was drawn from genetic counseling session transcripts from the Genetic Counseling Video Project and Genetic Counseling Student Video Project. Coding and subsequent thematic analysis of the transcripts containing instances of guilt and shame were used to address the qualitative aims of the study. To address the quantitative aim of the study, linear and logistic regressions assessed the relationship between quantitative measures of communication and the disclosure of shame and guilt.
Simulated clients disclosed guilt or shame in 14 of the GCSVP sessions and 15 of the GCVP sessions, totaling 29 out of the 311 (9.32%) analyzed sessions. Qualitative analyses showed a variety of approaches taken by genetic counselors to elicit and respond to guilt and shame. In many cases, there were indicators of therapeutic relationship building prior to the disclosure. Genetic counselors and genetic counseling (GC) students' responses to the disclosure varied, with some genetic counselors providing little to no direct response and others attempting multiple interventions to address the simulated client's guilt. Quantitative analyses suggested that verbal dominance and the patient centeredness were significantly associated with the elicitation of guilt or shame and the length of discussion around these emotions in a session. This study is a first step towards identifying communication practices that facilitate disclosure and exploration of feelings of shame and guilt in simulated genetic counseling sessions, which may ultimately help to inform guidelines for guilt and shame counseling for genetic counselors and students
Staging Labor: On Theatricality as Mode of Production
No full textThis dissertation identifies an anti-theatrical prejudice in theoretical attempts to conceptualize production shared by otherwise divergent schools of thought. It demonstrates how this ‘horror of theatricality’ has structured a wide array of philosophical traditions in their thinking through the question of what it means to have effects on the world; traditions which would otherwise appear to have few metaphysical presuppositions in common. From Plato’s poet to Rousseau’s professional actor; from Hegel’s pre-rational man of picture-thinking to Nietzsche’s imitative slave; from Marx’s ideologue to Freud’s melodramatic hypochondriac: to be productive has always meant, first and foremost, to not be an actor. The premise of this dissertation is that, if theatricality has consistently been conjured up only to be dismissed as a perversion of the productive impulse, then the very procedures of its exclusion might offer clues as to the alternative, theatrical understanding of production. Rather than a substantialist-affirmative conception of production (production as the incessant creation of the new) or a negative one (production as negating the given), this dissertation proposes an understanding of labor as a staged phenomenon; one that produces effects only to the extent that it tacitly presupposes an already dramatized world in which hammers and sickles always start off as masks and costumes
Evaluation and implementation science methods for behavioral health interventions
Get PDFBackground: Many behavioral health interventions are implemented in practice without evidence supporting their effectiveness in the receiving target population. The objective of this dissertation is to facilitate the use of statistical methods not commonly used by behavioral health services researchers that can be applied to learn about different aspects of intervention effectiveness. I apply each method to study a behavioral health intervention that is being used widely without supporting evidence or is evidence-based but not yet being used widely.
Methods: In Chapter 2, I apply a difference-in-differences method designed for situations where there is staggered adoption of the intervention to estimate the average effects of state opioid prescribing cap laws on opioid prescribing after surgery. In Chapter 3, I transport the effect of the ACHIEVE weight-loss intervention for people with serious mental illness estimated in a trial to the Maryland Medicaid population, focusing on practical issues applied researchers encounter when conducting such transportability studies. In Chapter 4, I describe a meta-regression method researchers can use to study the association between implementation variables and intervention effectiveness across sites. Then, I apply the method to identify which dimensions of integrated care are associated with behavioral health homes’ effects on outpatient mental health utilization in Maryland.
Results: On average, state opioid prescribing cap laws evaluated in Chapter 2 did not affect or only minimally affected the prescribing of opioids to manage postsurgical pain. In Chapter 3, ACHIEVE was predicted to be as effective among psychiatric rehabilitation program utilizers and people with schizophrenia spectrum disorders as it was in the trial. It was projected to be less effective in a broader population of people with serious mental illness, but the validity of that projection relies on stronger assumptions. In Chapter 4, indicators for having co-located mental health providers and having regular collaboration with primary care providers were positively associated with health homes’ effects on outpatient mental health utilization.
Conclusions: Methods for estimating, transporting, and examining heterogeneity in effects of behavioral health interventions are available to researchers. The careful application of these methods can yield complementary insights into the effectiveness of such interventions in practice