Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


45651.pdf (1.71 MB)

ETD Abstract Container

Abstract Header

Two Essays on Association Analysis for Discrete Outcome Variables with Applications to Well-being and Clinical Trial Studies

Fan, Zhaohu
http://rave.ohiolink.edu/etdc/view?acc_num=ucin1692299120046228

Abstract Details

2023, PhD, University of Cincinnati, Business: Business Administration.
Essay 1: In addition to clinical efficacy, safety is another important outcome to assess in randomized controlled trials. It focuses on the occurrence of adverse events, such as stroke, deaths, and other rare events. Because of the low or very low rates of observed adverse events, meta-analysis is often used to pool together evidence from dozens or even hundreds of similar clinical trials to strengthen inference. A well-known issue in rare-event meta-analysis is that part or even the majority of the available studies may observe zero events in both the treatment and control groups. The influence of these so-called double-zero studies has been researched in the literature, which nevertheless focuses on reaching a dichotomous conclusion -- whether or not double-zero studies should be included in the analysis. It has not been addressed when and how they contribute to inference, especially for the odds ratio. This work fills this gap using a comparative analysis of real and simulated data sets. We find that a double-zero study contributes to the odds ratio inference through its sample sizes in the two arms. When a double-zero study has an unequal allocation of patients in its two arms, it may contain non-ignorable information. Exclusion of these studies, if taking a significant proportion of the study cohort, may result in inflated type I error, deteriorated testing power, and increased estimation bias. Essay 2: Statistical methodology for categorical data analysis: quantifying and visualizing partial association between mixed variables. The outbreak of COVID-19 has lowered the well-being of college students across the world according to existing studies. However, these studies base their investigations solely on post-event surveys, and very few conducted the same surveys before the COVID-19 outbreak. Instead, college students were asked to recall their pre-COVID situations, which may likely result in recall bias and focusing illusion. To enable a counterfactual to study the COVID-19 impact, our research in this work analyzes survey data, collected in both pre-pandemic and early pandemic periods, from two independent cohorts of New Zealand first-year college students. In our association study for well-being and common psychological factors, we find that by controlling for age and gender, the other factors (physical healthiness, loneliness, and accommodation) show an increased moderation effect after the strike of COVID-19. Our empirical findings may deliver various insights to domain experts and lead to more specific studies to assist university policymakers and healthcare providers in decision-making. The statistical challenge in this empirical analysis lies in the lack of an appropriate assessment of partial association for mixed types of data. The core idea of this work is to map the residual randomness (after adjusting for covariates) to the same continuous scale, regardless of whether the outcome variable is binary, ordinal, or continuous. This is achieved by defining a unified residual that broadly applies to mixed outcomes and a wide range of regression models. Our methodology and theory substantially broaden the scope and applicability of the surrogate idea used in Liu et al. In particular, our new measure generalizes classical Kendall’s tau in the sense that it can size both partial and marginal associations. It plays an instrumental role in the discovery of an elevated moderation effect as a result of the COVID-19 impact.
Dungang Liu, Ph.D. (Committee Chair)
Nanhua Zhang, Ph.D. (Committee Member)
Yan Yu, Ph.D. (Committee Member)
Rashmi Adaval, Ph.D. (Committee Member)
62 p.
Business Administration
Mental health; moderation analysis; covariate adjustment; Adverse even; clinical trial; rare event

Recommended Citations

Citations

  • Fan, Z. (2023). Two Essays on Association Analysis for Discrete Outcome Variables with Applications to Well-being and Clinical Trial Studies [Doctoral dissertation, University of Cincinnati]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1692299120046228

    APA Style (7th edition)

  • Fan, Zhaohu. Two Essays on Association Analysis for Discrete Outcome Variables with Applications to Well-being and Clinical Trial Studies. 2023. University of Cincinnati, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=ucin1692299120046228.

    MLA Style (8th edition)

  • Fan, Zhaohu. "Two Essays on Association Analysis for Discrete Outcome Variables with Applications to Well-being and Clinical Trial Studies." Doctoral dissertation, University of Cincinnati, 2023. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1692299120046228

    Chicago Manual of Style (17th edition)

ucin1692299120046228
90
© 2023, all rights reserved.
This open access ETD is published by University of Cincinnati and OhioLINK.