30921 research outputs found
Sort by
Joint model of multilevel longitudinal and survival data: using the auxiliary mixed-effect Poisson model as the survival submodel
No full textJoint modeling (JM) of longitudinal and survival data is important in fields such as epidemiology and medicine, where understanding how repeated measures relate to the occurrence of events is essential. Cluster randomized controlled trials (CRCTs), which randomize groups rather than individuals, are increasingly used to assess interventions such as mental health programs. However, standard JMs struggle to accommodate the hierarchical structure of CRCTs, resulting in biased estimates. These challenges are exacerbated by the computational intensity of standard Bayesian methods, which typically rely on numerical integration or techniques like the “zero-trick” for modeling survival outcomes.
This dissertation addressed these methodological gaps by introducing a computationally efficient JM and extending it into a Bayesian multilevel joint model (MJM) specifically designed for CRCTs. Two studies were conducted to achieve these objectives.
The first study proposes a new JM leveraging a Poisson regression for survival submodel. Extensive simulations demonstrated that this Poisson-based approach maintains estimation accuracy while substantially reducing computational demands. Compared to existing approaches (e.g., JMbayes), the Poisson-based method provided comparable or superior accuracy and fewer convergence issues, even under moderate censoring rates and smaller sample sizes. This framework offers a more accessible and practical solution for standard JMs.
The second study extends this Poisson-based JM into a multilevel setting to account for nested structures of CRCTs. This proposed MJM incorporates a three-level structure—repeated measures nested within individuals and individuals nested within groups. It incorporates random effects, covariates and interactions across all levels, enabling a comprehensive evaluation of direct and indirect intervention effects. Simulations and empirical analyses reveal that ignoring clustering can underestimate intervention effects and inflate Type I error rates. The MJM more accurately recovers other fixed and random effects parameters, and the association between longitudinal and survival outcomes.
Overall, this thesis significantly advances JM methodology by offering a robust, flexible multilevel framework tailored to CRCTs. The findings highlight the necessity of incorporating multilevel structures into analyses of clustered data. This work provides valuable methodological guidance and tools for statisticians, epidemiologists, and policymakers, facilitating more accurate, nuanced, and policy-relevant evaluations of complex interventions within real-world clustered settings.October 202
Development of polymyxin and aminoglycoside-based outer membrane permeabilizing antibiotic potentiators
No full textThe prevalence of antimicrobial resistance (AMR) necessitates the development of alternative therapeutic options, particularly against critical priority Gram-negative pathogens. The utilization of antibiotic adjuvants or potentiators is an advantageous strategy that targets bacterial resistance mechanisms, thereby augmenting the activity of an antibiotic used in combination. β-lactamase inhibitors (BLIs) in particular, are widely used to preserve the efficacy of β-lactam antibiotics. Outer membrane (OM) permeabilizers are a promising class of adjuvants which disrupt the formidable OM of Gram-negative bacteria. Two structurally distinct antibiotics – polymyxins and aminoglycosides – have different modes of action but share the ability to interact with the bacterial OM, making them widely repurposed in the development of OM permeabilizers. The objective of the thesis is to contribute to the ongoing research of polymyxin and aminoglycoside-based OM permeabilizers by (1) synthesizing derivatives with reduced toxicity and enhanced OM permeabilizing capability and (2) determining suitable antibiotic partners to develop novel combinations active against Gram-negative bacteria. The various projects presented in this “sandwich” thesis include (1) guanidinylated polymyxins capable of potentiating rifampicin, erythromycin, ceftazidime and aztreonam against multidrug-resistant Gram-negative bacteria; (2) guanidinylated amphiphilic tobramycin derivatives which enhanced ceftazidime and aztreonam, and their combinations with avibactam against β-lactamase harboring Pseudomonas aeruginosa; and (3) a tobramycin ether derivative which sensitized Gram-negative bacteria to rifampicin, and restored susceptibility of MDR Escherichia coli to minocycline and doxycycline.Canadian Institutes of Research (162159)October 202
Leveraging multigenerational health data to enhance mental disorder risk prediction: a population-based cohort study
No full textBackground: Mental disorders are highly prevalent, and comorbidities between physical and mental health conditions are common. Physical comorbidities and family health histories may improve the accuracy of mental disorder risk prediction. We developed prediction models for mental disorder risk using comprehensive individual and family mental and physical health histories. Methods: We conducted a population-based cohort study using administrative Health data in Manitoba, Canada, and included adults between 1977 and 2020 with linkages to at least one parent and one grandparent. Mental disorders (mood and anxiety, substance use and psychotic disorders) for individuals, parents and grandparents were identified in inpatient and outpatient Health records. Predictors included demographics, family history of mental disorders and 130 health conditions in individuals, parents and grandparents. We used the Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression to build prediction models that sequentially included health conditions in individuals, parents and grandparents. Predictive performance was evaluated using the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, positive predictive value, negative predictive value and Brier score. Results: Of 125 070 individuals identified, 109 359 had no preexisting mental disorder. 52.9% were males and 52.8% were urban residents. 39 651 (36.3%) had a recorded diagnosis of mental disorders during follow-up. Predictive models incorporating Health histories of individuals, parents, and grandparents achieved the best predictive performance. Amongst all mental disorders, psychotic and substance use disorders had the highest AUCs of 0.78 (95% confidence interval (CI) 0.75–0.81) and 0.75 (95% CI 0.73–0.76), respectively. Key predictors included comorbid mental disorders, gastrointestinal conditions, female infertility and family history of dementia, gastrointestinal and metabolic conditions. Conclusions: Individual and family histories of physical and mental conditions improved mental disorder risk prediction, though accuracy was only moderate, highlighting the need for further refinement of risk prediction
Exploring game design and use in an elementary mathematics classroom
No full textIn recent years, game-based learning has gained significant attention for its potential to make learning more engaging and less intimidating (Clark et al., 2016). The integration of games with educational content can make learning more interactive, enjoyable, and less stressful (Huang et al., 2013). However, not all educational games are of equal value (Russo et al., 2018) and the way they are designed may influence students’ attitudes toward mathematics. This qualitative study examines the use of two games with different design qualities in a Grade 4 classroom. Specifically, it investigates students’ reported experiences with each game and their teacher’s perspective regarding the design quality and use of mathematics games in her classroom. I gathered data through classroom observations, student questionnaires, teacher initial and follow-up interviews, and video recordings of small-group gameplay sessions. I conducted a thematic analysis to identify recurring patterns and meaningful insights into the design and use of these mathematics games in this classroom. The findings provide insights into the need to find a balance between competition and collaboration; luck and skill; and familiarity and novelty as features of game design that may support student engagement and foster positive attitudes towards mathematics learning. This study also shed lights on three aspects of the use of mathematics games that teachers should consider as they design or select games for use in their mathematics classrooms: intended purpose(s), group composition, and adaptability. The study also offers practical insights for teachers who are designing or using games in their mathematics classroom.October 202
Three essays on household finance, entrepreneurship, and corporate inance
No full textThis dissertation consists of three essays. In the first essay, we examine how childhood gender socialization influences the gender gap in stock market participation. Using sibling fixed effects to exploit within-family variation, we find that individuals raised in more male-dominated households, where fathers have higher income or education than mothers, are more likely to invest in stocks. This relationship is stronger for men than for women, indicating that childhood socialization may constrain girls to traditionally feminine tasks, limiting their opportunities to learn about investing. We provide evidence that the primary mechanism driving this relationship is the internalization of masculine characteristics, rather than financial literacy or risk tolerance.
The second essay investigates how traditional household gender imbalances during childhood affect adult entrepreneurship. Analyzing data from the Panel Study of Income Dynamics (PSID), we find that individuals raised in more traditional gender-imbalanced households are more likely to own businesses and achieve better entrepreneurial performance in adulthood. This positive association is stronger for men than women, thereby contributing to the gender gap in entrepreneurship.
In the third essay, we examine the relation between corporate pension liability and corporate social responsibility (CSR). On the one hand, firms that care about their employees tend to carry greater pension liabilities, and these firms are likely to be socially responsible firms, indicating a positive relationship between a firm’s pension liability and CSR score. On the other hand, resource-competing relation implies achieving and maintaining a fully funded pension plan often requires significant financial outlays, therefore may crowd out funding for other socially responsible activities, especially those not directly tied to employee satisfaction. Our results show that a firm’s pension liability positively affects its CSR score overall. Furthermore, by decomposing the CSR score into its employee satisfaction and non-employee components, we find that this positive effect works through CSR’s employee satisfaction component. Our results also show that pension funded status negatively moderates this positive effect: Funded status of pension competes with a firm’s CSR investments, given its limited resources and this resource-competition pattern is largely driven by CSR’s non-employee component. Granular analyses on CSR strengths and concerns present nuanced evidence for how pension liability and its funded status jointly affect CSR’s employee satisfaction and non-employee components. Using baby boomer as an instrument, we mitigate the endogeneity concerns and confirm our main results. Our work has real-world implications for human resource management in a broader CSR setting.October 202
Development and analysis of SARS-CoV-2 spike receptor binding domain-targeting aptamers
No full textThe severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic has highlighted the need for developing both diagnostic and therapeutic tools for emerging diseases. Development of aptamers, nucleic acid-based ligands that bind to target molecules, has become an increasingly popular field in the context of drug discovery and diagnostics. Due to its simplicity, low-cost, and rapid synthesis, aptamer-based diagnostic tools have shown to be on par with antibody-based lateral flow assays and RT-qPCR, which make aptamers a potentially powerful and flexible tool to respond to future outbreaks. The SARS-CoV-2 viral surface receptor Spike protein is a prime candidate target for both diagnostic and therapeutic purposes. Through the process of systematic evolution of ligands by exponential enrichment (SELEX), receptor binding site (S-RBD)-specific aptamers were screened and developed over multiple cycles of SELEX. Amplified binding sequences from each round of selection were collected and assessed using Bio-layer interferometry (BLI), electrophoretic mobility shift assay (EMSA), and qualitative polymerase chain reaction (qPCR) melt curve analysis. The final mature heterogeneous aptamer pool (10R) was then purified and sequenced on the Illumina next generation sequencing (NGS) platform. Samples from intermediate heterogeneous pools showed increasing target affinity as selection cycles progressed, indicating affinity maturation as stronger binding aptamers were retained and amplified while the weaker sequences were removed from the selection pool. After identifying the top candidates with NGS, they were evaluated with BLI assay, and the best four sequences were assessed with capture aptamer linked immobilized sorbent assay (ALISA). Phophorothioate-modified aptamers were able to retain sequence integrity after 48 hours of incubation in near-physiological environment. However, the phosphorothioate-modified aptamers were not able to neutralize Spike-presenting lentiviral pseudovirus in a surrogate neutralization assay. In conclusion, the study evaluated the viability of aptamer development as a response to the Coronavirus disease 2019 (COVID-19) pandemic. Throughout the process of developing and characterizing SARS-CoV-2 S-RBD aptamer binders, the study identified the benefits and pitfalls of the approach, and suggested that the aptamer platform could be a powerful therapeutic and diagnostic tool for rapid pandemic response once the major challenges have been considered and addressed.October 202
An educational leader's journey to cultivate reconciliation
No full textThe ability of an educational leader to look beyond what one views as normal is the first step to beginning a reconciliatory journey. The second step is to see the potential in cultivating a school culture that encompasses Indigenous knowledge and perspectives. Mamàhtawisiwin: The Wonder We Are Born With—An Indigenous Education Policy Framework, developed in collaboration with Manitoba Education and Early Childhood Learning (2022) includes four action-based strategies outlining the foundations of an Indigenous-inclusive education system. The curiosity of how educational leaders are currently cultivating a culture of reconciliation through the acquisition of Indigenous knowledge and perspectives directly relates to the documents vision. The purpose of this research was to explore how educational leaders have adapted their practice by utilizing the four action-based strategies. With use of the four action-based strategies outlined in the Mamàhtawisiwin as the theoretical framework, the overall aim of the research was to develop a set of wisdoms that would assist and support an educational leader’s journey to cultivate reconciliation.
A descriptive research study was the methodology utilized by conducting virtual interviews with non-Indigenous and Indigenous educational leaders. This descriptive research study design required listening to the stories, acquiring knowledge, acknowledging perspectives, and analyzing the dynamic relationships to identify attributes and strategies that could be utilized by other educational leaders to improve school cultures for the benefit of reconciliation. The objective of the descriptive research study was to explore how educational leaders nurtured school cultures by employing Indigenous knowledge and perspectives.October 202
Assessing the impact of day programs on individuals living with dementia and their family/friend caregivers (AIDA-DemCare): protocol of a prospective cohort study combined with a qualitative evaluation
No full textBackground: Persons with dementia, their family/friend caregivers, and health systems agree that the preferred place of dementia care is a person’s home. Health systems have struggled to provide sufficient community-based dementia care supports, negatively affecting the health and well-being of persons with dementia and their caregivers. Therefore, dementia care and support of dementia caregivers are critical public health priorities. Adult day programs aim to simultaneously support the health and well-being of persons with dementia and their caregivers, but research focusing on these day program outcomes is lacking. The primary objective of this study was to compare various outcomes of day program attendees with dementia and their caregivers over time to outcomes of non-attendees with dementia in the community (with care needs similar to those of attendees) and their caregivers.
Methods: In each of four Canadian health regions (York Region, Ontario; Interior Health, British Columbia; Calgary, Alberta; Winnipeg, Manitoba), this study will recruit 250 day program attendees with dementia and their caregivers, and 500 community-based non-attendees with dementia and their caregivers. Applying criteria used by the health system to determine day program eligibility, we will identify non-attendees with care needs similar to those of attendees. We will combine participants’ longitudinal health administrative data with repeated (baseline, 1-year, and 2-year follow-up) surveys to include variables not routinely collected by healthcare systems. Primary study outcomes are quality of life of the person with dementia and their caregiver. Secondary study outcomes include mental health (older adults, caregivers), cognitive and physical decline (older adults), time to admission to congregate care (older adults), and emergency room, hospital, and primary care use (older adults, caregivers). Using a day program survey, we will also assess day program characteristics. Using covariate-adjusted general estimating equations and time-to-event models, we will compare these outcomes between groups of day program exposure (no, low, medium, high).
Discussion: Persons living with dementia, their caregivers, and health systems urgently need solutions to support living at home with a good quality of life. This study will generate provide evidence on the potential of day programs to address these critical public health needs.
Trial registration: ClinicalTrials.gov: NCT06496945
Woven Roots: exploring immigrant identity, cultural heritage, and social change through symbolic art
No full textIn my thesis, "Woven Roots: Exploring Immigrant Identity, Cultural Heritage and Social Change Through Symbolic Art," over time, the focus of my artwork has become more defined. My pieces now reflect cultural identity through the use of symbols.
Relocating from Bangladesh to Canada profoundly deepened and personalized my understanding of cultural identity. This geographical and emotional distance heightened my awareness of what I had left behind, the textures, rituals, and visual language that shaped my upbringing. In response, I began to incorporate these cultural symbols into my artwork, using them to reconnect and reflect. They became a visual vocabulary through which I could express the layered emotions that come with displacement, longing, and identity in flux.
While researching the topic, I became conscious of the colonial history of my culture and how it reshaped the culture of Bangladesh. This curiosity led me to explore the status of women in the post-colonial era, a subject that felt inherently significant to me as a woman. To examine this idea, I have integrated symbols from traditional Bengali scroll paintings alongside my own symbolic language, illustrating how my culture has transformed over time and how the lasting impact of colonization continues to shape my country.October 202
The role of metabolic enzymes in virulence and immunopathogenesis of Leishmania infection
No full textLeishmaniasis is a neglected tropical disease affecting approximately a million people yearly. The disease is caused by intracellular protozoan parasites that belong to the genus Leishmania. Different Leishmania species cause different clinical forms of the disease, which can present as self-healing cutaneous lesions or often fatal visceral disease that affects the liver, spleen and bone marrow. Unfortunately, there is yet no approved vaccine against human leishmaniasis. Although some drugs are effective, treatment often results in serious drug-mediated toxicity and side effects, and there is increasing evidence of drug resistance. However, vaccination is possible because people who recover from the disease are immune to reinfection. Knowledge of the antigens that induce protective immunity in infected hosts is critical for designing effective vaccines and therapeutic strategies against the disease. Dihydrolipoyl dehydrogenase (DLD) and Phosphoenolpyruvate carboxykinase (PEPCK) are key metabolic enzymes in numerous eukaryotic cells, including Leishmania and play critical roles in metabolism. This present study aims to assess the impact of DLD and PEPCK deficiency on parasite virulence and host immune response. Using the CRISPR-Cas 9 gene editing tool DLD and PEPCK gene in L. major and L. donovani, respectively, were deleted. The impact of deficiency of these enzymes on parasite survival, infectivity and host immune response were assessed. The gene-deficient parasites demonstrated reduced proliferation in axenic cultures and inside infected macrophages in vitro.
Mice infected with DLD-deficient parasites had no observable lesions and harboured significantly lower parasite burden compared to their WT or complementary add-back control-infected animals. Interestingly, DLD deficient L. major infected mice mounted blunted immune responses compared to their WT and complementary addback infected control counterpart mice. Vaccination with DLD deficient parasites induced strong protective responses upon virulent L. major rechallenge, and this was associated with strong IFN- response.
The ability for PEPCK-deficient L. donovani to differentiate into their infective forms was not compromised in vitro. Although their infectivity in macrophages were compromised in vitro, these parasites maintained their virulence and pathology in vivo. In mice infected with PEPCK deficient L. donovani, immune responses in the liver and spleens were relatively unaltered apart from increased TNF production in the liver. Interestingly, the infectivity of PEPCK KO parasites in Kupffer cells in vitro was compromised compared to their WT or AB parasite controls, suggesting that the in vivo microenvironments may predispose PEPCK-deficient parasites to a compensatory mechanism that enables their proliferation.
Collectively, these findings show that DLD is a critical metabolic enzyme for intracellular survival of L. major and targeting this molecule could be a viable option for controlling the disease. Although PEPCK-deficient L. donovani is attenuated in vitro but not in vivo, they may not be a suitable live attenuated vaccine candidate against visceral leishmaniasis.February 202