Makella S. Coudray

Emerging technologies—such as generative AI, predictive analytics, and network data collection—are reshaping social and behavioral research by offering new opportunities alongside complex ethical challenges. This moderated panel will examine oversight dilemmas through three illustrative case studies: predictive modeling to support harm reduction and reduce overdose risks; the use of spatial and social network data in HIV/STI research; and generative AI’s role in addressing drug-related health inquiries where misinformation can have life-or-death consequences. Drawing from real-world research examples, panelists will share practical strategies to navigate issues of equity, privacy, misinformation, and harm prevention, while strengthening public trust in research involving these cutting-edge technologies.

In the United States, the number of new cases of non-HIV sexually transmitted infections (STIs) such as chlamydia, gonorrhea, and syphilis continues to increase. STIs are commonly associated with short-term and long-term adverse sequelae which can ultimately impact an individual’s health. Additionally, STIs increase the odds of HIV acquisition and transmission. Given the high burden of STIs on some minoritized groups (e.g., people of color) and high costs associated with disease management and treatment, effective STI prevention techniques are essential. Historically STI prevention methods have focused on reducing sexual risk (at an individual level). However, effective STI prevention should have a multilevel approach. As such, integrating a sexual health paradigm into STI prevention may be a more successful approach.

As a researcher, Dr. Coudray is committed to advancing sexual health equity among marginalized groups. Using an intersectional lens, her research examines how the social determinants of health inequities, multiple marginalized identities, and discrimination contribute to HIV/STI testing, risk, and prevention among sexual and gender minority groups and racial and ethnic minority groups. Her talk will explore the (1) the relationship between bacterial vaginosis and STIs, (2) sexual health information sources and practices among emerging adults, and (3) STI testing, risk, and prevention among transgender women with and without neovaginas. She will discuss how she utilized techniques such as latent class analysis, mixed methods, and implementation science to further her work. Additionally, Dr. Coudray will give an overview of her future research, which will examine additional research areas such as sexual health research ethics, and bio-behavioral interventions for STI prevention among sexual gender minorities.

The exact etiology of bacterial vaginosis (BV) remains unknown, however, previous literature suggests an association between BV and vaginal douching. This study aims to investigate frequent and infrequent vaginal douching as a risk factor for BV among young women in US. Methods: Data were obtained from a recently completed randomized, open label trial of home screening and treatment of young women with asymptomatic BV who were also at high risk for sexually transmitted diseases. Baseline characteristics of the study sample, restricted to African American (AA) women, were analyzed. Descriptive analyses and logistic regression were performed using the statistical package SPSS. Results: The mean age among the 977 AA participants was 21 years (range 17-25 years). Most women (38.6%) attended college but did not receive a degree, 57.5% reported prior pregnancy and 54.1% were never treated for BV. The prevalence of chlamydia and gonorrhea was 15.7% and 5.3% respectively. While 12.2% of women had intermediate vaginal flora (Nugent score between 4-6), 87.8% of women had BV (Nugent score between 7-10). Women who douched daily had 10.49 odds of BV as compared to women who did not douche daily (95% CI 1.60-68.58). This association was statistically significant. Conclusion: Douching daily is a risk factor for BV among young AA women. Infrequent douching, measures of clinical history, sexual history and demographic data were not associated with BV.