Date of Award

Fall 2024

Document Type

Open Access Dissertation

Department

Epidemiology and Biostatistics

First Advisor

Monique Brown

Abstract

Introduction

Men who have sex with men (MSM) including gay, heterosexual, or bisexual men account for 3.8% to 6.4% (3.9 to 7.1 million) of all men in the US. Despite basic human rights, MSM is disproportionately impacted by adverse social, behavioral, and psychological challenges. These challenges include, but are not limited to homophobia, discrimination, stigma, lack of medical support services, fear of losing one’s job, confidentiality concerns, and fear of talking about sexual orientation and practices. According to the Centers for Disease Control and Prevention (CDC), these barriers may prevent MSM from seeking preventive and treatment services and intersect with a greater risk of disease incidence. MSM reports poor mental health, drug use, alcohol consumption, and smoking, as well as being at high risk for chronic, infectious, and sexually transmitted diseases. Adverse childhood experiences are also of high concern among this group, which has been linked with lower means of CD4/CD8 ratio (1.84 vs 2.47), Hepatitis C, and poor oral health outcomes such as tooth loss, dental caries, missing teeth and fillings or restorations, bleeding gums, orofacial pain, and untreated oral health care needs, denture use, and difficulty in chewing. Irrespective of these statistics, none of the previous studies have explored mediating pathways that may contribute to high morbidity. Thus, the current study aimed to 1) examine the mediation role of depression between adverse childhood experiences (ACEs) and CD4/CD8 ratio, 2) assess the mediation role of substance use between adverse childhood experiences (ACEs) and Hepatitis C, and 3) determine the causal effect of ACEs on oral manifestations and examine the mediation role of depression and substance use between ACEs and oral manifestations. The present study also aimed to assess the effect of measure modification by HIV status.

Methods

A retrospective analysis of a prospective Multicenter AIDS Cohort Study (MACS) included 2,083 MSM who responded to the childhood trauma questionnaire in meth and/or Healthy Aging sub-studies. We assessed the causal effect of ACEs (yes vs no), ACE types (no abuse, physical, emotional/psychological, sexual, and environmental abuse), and ACE score (0,1,2,3, 4, and above) on CD4/CD8 ratio at mid and final visit. For aim 1, depression, a mediator was operationalized as a continuous variable. For aim 2, substance use was used as a binary variable. For aim 3, depression and substance use were operationalized as binary variables. Age, race, employment, ethnicity, and income were included as covariates. For aim 1, AIDS manifestations and ART adherence were also included as covariates. For aim 1 and aim 2, the causal mediation analysis using a single mediator model and allowing for exposure-mediator interactions was used to calculate the direct and indirect effects. For aim 3, the causal mediation analysis using a multiple mediator model based on Lange’s method was used to calculate the direct and indirect effects at mid and final visits. For all three aims, we conducted a sensitivity analysis using multiple imputations method with fully conditional specification (FCS).

Results

The controlled direct effect (CDE) of ACEs on CD4/CD8 ratio was statistically significant among HIV-positive individuals (CDE=0.169; p-value=0.028). The total effect (TE) and CDE of childhood sexual abuse on CD4/CD8 ratio were also statistically significant (TE=0.128; p-value=0.060; CDE=-0.123; p-value=0.072) in the overall sample. However, we did not find depression mediating this relationship. Similarly, substance use does not seem to mediate the causal effect of abuse on Hepatitis C in complete case analysis. After data imputation, in the presence of substance use, participants who experienced overall ACEs (ORNIEmid: 2.72, ORNIEfinal: 2.72) had higher odds of Hepatitis C. Also, those who faced physical or psychological (ORNIEmid: 2.73, ORNIEfinal: 2.72), sexual (ORNIEmid: 2.72, ORNIEfinal: 2.72), environmental (ORNIEmid: 2.73, ORNIEfinal: 2.71), and/ or a greater number of ACEs (ORNIEmid: 2.73, ORNIEfinal: 2.72) had higher odds of Hepatitis C at both mid and final visits compared with individuals who did not face abuse. These results were statistically significant (p-value< 0.0001). In addition, after testing the interaction by HIV status, physical or psychological abuse was found to impact oral manifestations (OR=1.98; 95% CI: 1.345-2.918; p=0.0005) compared with those who did not experience abuse. The study also found that experiencing childhood sexual abuse was also linked with (OR=1.41; 95% CI: 1.072-1.849; p=0.014) oral manifestations. These findings remained statistically significant in the presence of interaction with HIV, where those who faced childhood sexual abuse had higher odds of oral manifestations (OR=1.78; 95% CI: 1.005-3.154; p=0.048). Likewise, a history of environmental abuse was found to impact the odds of oral manifestation (OR=2.10; 95% CI: 1.443-3.063; p=0.0001) in the presence of interaction with HIV at the final visit. We also found a dose-response relationship between abuse and oral manifestations, where experiencing a greater number of abuse events had slightly higher than one odds of oral manifestations in the overall sample (OR=1.05; 95% CI: 1.023-1.085; p=0.0005). The result remained consistent after assessing interaction with HIV (mid-visit: OR=1.07; 95% CI: 1.028-1.113; p=0.0008). In addition, in the presence of substance use (final-visit: OR=1.05; 95% CI: 1.009-1.105; p=0.017) and depression (final-visit: OR=1.06; 95% CI: 1.009-1.105; p=0.017), the odds of oral manifestations were also slightly greater than one among individuals who faced a greater number of abuse events. Also, a greater number of ACEs were found to have 1.98 higher odds of oral manifestations (95% CI: 1.345-2.918; p=0.0005) in the presence of HIV and absence of substance use and depression present at the final visit.

Conclusion

In conclusion, our study found mixed results, providing some evidence that depression and substance use may mediate the pathways from ACEs to a ratio of CD4/CD, Hepatitis C, and oral manifestation with HIV playing a crucial role. However, future research is needed to examine longitudinal effects with larger sample sizes. Our study suggests that it is crucial to understand the complex and multifaceted interplay between childhood traumatic experiences and morbidity. Targeted interventions, particularly in the context of HIV status may improve the overall quality of MSM.

Rights

© 2025, Amandeep Kaur

Download

Available for download on Thursday, December 31, 2026

Included in

Epidemiology Commons

Share

COinS