Date of Award

Fall 10-1-2021

Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor



While there are several perspectives on marginalization, there are multiple marginalized individuals, social groups, and communities globally. This process of marginalization produces individuals, groups and communities which are refused complete privileges, rights, and power within the broader political and social framework. Social, cultural, biological, and economic factors can thus be used as yardsticks to marginalize individuals and communities. Marginalization can be based on gender, race and ethnicity, social class, and sexuality, among others. Clearly, marginalized communities face poorer health outcomes and these outcomes are sometimes linked to risky behaviors more prevalent in such demographics. Marginalization is associated with reduced health outcomes and can limit the agency of marginalized communities. However, even within sites of marginalization, affected communities make significant attempts to mitigate health risks and retain agency. For example, marginalized men who have sex with men in China face severe discrimination which affects their health outcomes. Even within such contexts, these men still encourage peers to receive sexually transmitted infection testing. I explore how marginalized communities reduce health risks likely produced by marginalization and retain agency through doing so. I explore sexually transmitted infection testing and related issues in Chinese men who have sex with men, the United States legal cannabis industry, and medication for those with opioid use disorder. In doing so, I will provide understanding on risk reduction of health behaviors in marginalized communities, building a knowledge base to aid overall health outcomes. In the first chapter, I detailed a range of cannabis-centric studies. First, I detailed cannabis usage preferences among United States cannabis users. I put forth that frequent cannabis use may increase risk of health harms and highlighted the need to minimize problematic use. I also explored sociodemographic indicators and their association with likelihood for cannabis-related emergency department admissions in New York City. Results suggested that cannabis use may further burden marginalized groups. I investigated large cannabis firms’ motivations for participating in the cannabis space. I put forth that policymakers be aware that non-profits and for-profits both seek to expand cannabis access and consider the groups as a unified whole. In the second chapter I explored concerns regarding sexually transmitted infection testing in the Chinese men who have sex with men environment. I first detailed factors associated with sexually transmitted infection testing. Results detailed the role of altruism in a sexually transmitted infection testing intervention. Expressions of altruism may promote contributions toward public health initiatives in marginalized communities. In the same vein, I detailed the association between men who have sex with men community-centric behaviors and contributions toward others’ sexually transmitted infection testing. I proposed that community-oriented behaviors may be related with a reduction in testing service costs. Then, I evaluated whether men who have sex with men selected a sexually transmitted infection test appropriate for their sexual behavior. I suggested that disclosing sexual identity to treatment providers can improve men who have sex with men sexually transmitted infection prevalence estimates. I also detailed the correlates of antisocial behavior on the world's largest gay dating app among Chinese men who have sex with men. I suggested that age, condom use, and number of social ties may be associated with antisocial behavior, with implications for the design of online sexual health interventions. Finally, I assessed if same-sex sexual behavior disclosure of Chinese men who have sex with men was related to number of HIV self-testing kits requested, and number of test results successfully uploaded by alters in a network-based HIV self-testing intervention. Findings had implications for the development of network-based interventions for key populations. In the final chapter I detailed that various forms of social network support may influence medication for opioid use disorder treatment outcomes. Failure to implement successful social network support programs within medication for opioid use disorder treatment settings may represent an important missed opportunity to engage patients at risk of treatment failure. While the topics here are broad, they all share similar thematic arcs. Low sexually transmitted infection testing uptake, opioid use disorder and cannabis use are issues often disproportionately faced by marginalized communities. Establishing marginalization as the causal factor behind these concerns is often complex, but there is significant work indicating that problematic patterns of drug use and poor sexual health outcomes are engendered by marginalization. Marginalization is associated with conditions inimical to health and well-being, creating a host of health risks. Such marginalization limits the agency of affected communities. However, even within these sites of marginalization, men who have sex with men seek testing and opioid use disorder patients seek medication, mitigating health risks borne from marginalization. I advance that marginalized communities are not completely helpless considering reduced health outcomes, indicating how agency is reclaimed. Finally, I indicated other cases where fostering agency in marginalized communities needs to be carefully considered.