Unexpected findings that contradict well-established relationships between exposures and outcomes are often referred to as “paradoxes” in the epidemiologic literature. For example, the “birth weight paradox” refers to the observed protective association between smoking during pregnancy and infant mortality among low birth weight infants. A recent body of literature suggests that this and several other well-known epidemiologic paradoxes can be attributed to collider bias. Collider bias results from conditioning on a variable that is caused by the exposure or shares common cause with the exposure and is caused by the outcome or shares common causes with the outcome. Several recent epidemiology textbooks and methodological studies further suggest that collider bias is the graphical representation of selection bias, suggesting that these two biases are synonymous. This structural approach to bias is conceptually very useful for defining, describing, and identifying selection bias, but it introduces paradoxes of its own due to contradictory conclusions in the selection and collider bias methodologic literatures about their likely impact on study results in terms of magnitude, direction, and strata affected. Resolving these discrepancies is essential for our theoretical understanding of the relationship between selection and collider bias and has important practical implications for how we teach epidemiology, design studies, and evaluate and quantify the potential effects of bias on our results. For example, while patterns of collider bias coincide qualitatively with the birth weight paradox, the magnitude of collider bias would have to be substantial to reverse the sign of the association, contrary to prevailing beliefs that collider bias only minimally affects our results. To date, the plausibility of collider bias as an explanation for the birth weight paradox has not been empirically evaluated using data in which the paradox is observed.Taken together, these inconsistencies and contradictions suggest that our understanding of selection bias and collider bias remains incomplete. The overarching goal of this dissertation was to advance the theoretical and quantitative understanding of the impact of collider bias on study results to clarify the relationship between selection and collider bias. I began by systematically reviewing the methodologic literature on selection and collider bias. I found that selection bias and collider bias are increasingly treated as synonyms, but that conclusions about the magnitude and direction of selection and collider bias, the stratum affected, and the conditions under which the effects of each type of bias were evaluated are highly inconsistent. This suggested that divergent findings about the impact of selection and collider bias might be resolved by considering the impact of collider bias under a broader set of circumstances. I used microsimulations grounded in the sufficient component cause model to examine collider bias not under the null; interrogate why multiplicative interaction appeared central to the impact of collider bias; and clarify which stratum or strata are affected by collider bias. I identified clear patterns for the magnitude, direction, and strata affected by collider bias and successfully reconciled discrepancies with the selection bias literature. This work also enabled me to interrogate both the causal mechanisms and mathematical principles that underlie collider bias, which revealed how collider bias leads to non-exchangeability and when stratifying on a collider results in bias. Finally, I applied this deeper understanding of the mechanisms underlying collider bias to empirically evaluate the plausibility of collider bias as an explanation for the birth weight paradox. Using microsimulations parameterized with 2015 National Center for Health Statistics Cohort Linked Birth-Infant Mortality, I identified scenarios that successfully reproduced the paradox and all observed relationships between smoking during pregnancy, infant mortality, and low birth weight. These findings strengthen the evidence for the role of collider bias in producing the paradox and shed light on the potential magnitude of unmeasured confounding and direct effects of smoking and low birth weight on infant mortality that may be required for the observed magnitude of the paradox to arise. This work clarifies that almost all selection bias is collider bias; that the effects of collider bias vary in magnitude and direction; that selecting on a collider always leads to bias, but this bias may not occur in the stratum that coincides with our analytical sample; and that collider bias may resolve the birth weight paradox, but is unlikely to explain all epidemiologic paradoxes.

The purpose of this research project is to report on a comprehensive organizational audit of the Urgent Care Service of Desert Valley Medical Group in Victorville, California.

This study explored the effectiveness of a student question creation process engaging students actively in self, peer, and instructor interaction in development of affective, cognitive, and meta-cognitive skills. Employing a mixed-methods sequential explanatory design assigning both treatment and control activities sequentially in an alternating pattern over a six week period, students' performance on exams as well as their perceptions of various aspects of the student question creation process were used to evaluate the effectiveness of student-created questions (SCQs) activities as a cognitive strategy and to identify factors contributing to the effectiveness of question creation activities on students' learning. Subjects of this study were high performing and highly motivated graduate students in an 8-week online biomedical statistics course, part of a specialized master's program designed for medical school preparation. Survey findings and focus groups strongly supported the student question creation process as a facilitator of higher order thinking. However, the relatively short study duration, comparison of student question creation with another competing method for facilitating learning (discussion board) and not a pure control group, and availability of a common study guide course with student-created questions on all course topics may have muted assessment of the full impact of the strategy on learning. Although practically difficult in an education environment, further research to assess fully the impact of the student question creation strategy is desirable especially if these confounding factors can be greatly minimized, if not eliminated.

Indiana University-Purdue University Indianapolis (IUPUI) / Healthcare data visualization presents challenges due to its non-standard organizational structure and disparate record formats. Epidemiologists and clinicians currently lack the tools to discern patterns in large-scale data that would reveal valuable healthcare information at the granular level of individual patients and populations. Integrating geospatial and temporal healthcare data within a common visual context provides a twofold benefit: it allows clinicians to synthesize large-scale healthcare data to provide a context for local patient care decisions, and it better informs epidemiologists in making public health recommendations. Advanced implementations of the Scalable Vector Graphic (SVG), HyperText Markup Language version 5 (HTML5), and Cascading Style Sheets version 3 (CSS3) specifications in the latest versions of most major Web browsers brought hardware-accelerated graphics to the Web and opened the door for more intricate and interactive visualization techniques than have previously been possible. We developed a series of new geotemporal visualization techniques under a general healthcare data visualization framework in order to provide a real-time dashboard for analysis and exploration of complex healthcare data. This visualization framework, HealthTerrain, is a concept space constructed using text and data mining techniques, extracted concepts, and attributes associated with geographical locations. HealthTerrain's association graph serves two purposes. First, it is a powerful interactive visualization of the relationships among concept terms, allowing users to explore the concept space, discover correlations, and generate novel hypotheses. Second, it functions as a user interface, allowing selection of concept terms for further visual analysis. In addition to the association graph, concept terms can be compared across time and location using several new visualization techniques. A spatial-temporal choropleth map projection embeds rich textures to generate an integrated, two-dimensional visualization. Its key feature is a new offset contour method to visualize multidimensional and time-series data associated with different geographical regions. Additionally, a ring graph reveals patterns at the fine granularity of patient occurrences using a new radial coordinate-based time-series visualization technique.

Indiana University-Purdue University Indianapolis (IUPUI) / In medical research, data analysis often requires complex statistical methods where no closed-form solutions are available. Under such circumstances, Monte Carlo (MC) methods have found many applications. In this dissertation, we proposed several novel statistical models where MC methods are utilized. For the first part, we focused on semicompeting risks data in which a non-terminal event was subject to dependent censoring by a terminal event. Based on an illness-death multistate survival model, we proposed flexible random effects models. Further, we extended our model to the setting of joint modeling where both semicompeting risks data and repeated marker data are simultaneously analyzed. Since the proposed methods involve high-dimensional integrations, Bayesian Monte Carlo Markov Chain (MCMC) methods were utilized for estimation. The use of Bayesian methods also facilitates the prediction of individual patient outcomes. The proposed methods were demonstrated in both simulation and case studies. For the second part, we focused on re-randomization test, which is a nonparametric method that makes inferences solely based on the randomization procedure used in clinical trials. With this type of inference, Monte Carlo method is often used for generating null distributions on the treatment difference. However, an issue was recently discovered when subjects in a clinical trial were randomized with unbalanced treatment allocation to two treatments according to the minimization algorithm, a randomization procedure frequently used in practice. The null distribution of the re-randomization test statistics was found not to be centered at zero, which comprised power of the test. In this dissertation, we investigated the property of the re-randomization test and proposed a weighted re-randomization method to overcome this issue. The proposed method was demonstrated through extensive simulation studies.