Briefly summarize your background and interests, and how you came to the Institute.
I originally trained as a biomedical engineer and gained experience with in-host modeling of influenza during my Master’s degree at the University of Virginia. I was interested in continuing to work on infectious disease modeling, but wanted to move into the public health space. This led me to a job performing population-level math modeling of cervical cancer prevention and treatment interventions among women with HIV in South Africa with Dr. Ruanne Barnabas, who was then at the University of Washington. During my time in this role, I collaborated with international teams of modelers, many of whom were epidemiologists, and who inspired me to apply for PhD programs in epidemiology. During the application process, I met my advisor, Dr. Kim Powers, who, amongst other things, works on math modeling of HIV and other infectious diseases. Kim introduced me to Dr. Ann Dennis, with whom I’ve worked as a graduate research assistant on projects related to her HIV molecular epidemiology studies since coming to UNC in 2021. I’m now supported by the IGHID STI/HIV T32, and my training through UNC Epidemiology and IGHID has allowed me to expand on my interests in understanding sexual networks and modeling STI interventions.
What have you been working on with Dr. Dennis?
Ann works on neat projects evaluating different approaches for responding to growing HIV molecular clusters in North Carolina. Molecular clusters consist of people with similar HIV genetic sequences, indicating transmission. Cluster detection and response aim to interrupt the onward spread of HIV by prioritizing HIV prevention and care to groups experiencing rapid transmission and is a pillar of the Federal Ending the HIV Epidemic initiative. In order to design feasible and effective cluster-directed response strategies, we need good molecular data. Current practice in molecular cluster detection relies on sequences collected at entry to HIV clinical care, but reporting is incomplete, and the approach inherently misses people not linked to care.
To address this challenge, I worked on a project that leveraged a unique repository of HIV sequences collected at the time of HIV testing (versus later after linkage to care) as part of Ann’s PROMPT study (NIH R01) to understand how additional sequences could be used to identify data gaps and improve cluster detection in NC. (Shoutout to Dr. Shuntai Zhou and Nathan Long who performed the sequencing!) We found that incorporating sequences collected at the time of HIV testing improved molecular cluster detection in NC. While most people newly included in the cluster analysis likely had a sequence collected during clinical care but unreported, about a fifth were likely never linked to HIV care. People without a routinely collected sequence were diagnosed with HIV in earlier years and in more rural counties, were less engaged with public health and healthcare systems, or more transiently residing in NC. Our work emphasizes the importance of comprehensive HIV sequence reporting, either through increased laboratory reporting or sequencing remnant HIV test samples. Published here: (Broshkevitch, 2025, OFID) Sequencing HIV diagnostic samples to detect genetic clusters and assess sequence coverage gaps.
I’m also working to analyze the feasibility of the cluster-directed bridge counseling intervention evaluated in Ann’s PROMPT study, but this is still in progress!
What is next for you/ What are you doing for your dissertation?
For my dissertation I’m also thinking about sexual networks, but this time focusing on partner services networks and bacterial STIs (rather than molecular networks and HIV). Specifically, I’m working to quantify bacterial STI burden among men who have sex with men and transgender women in HIV and syphilis partner services networks. These people are identifiable and reachable, and could thus be a good group to prioritize for STI prevention and treatment interventions. I’m also aiming to model the impact of postexposure doxycycline (Doxy-PEP), taken after condomless sex to prevent bacterial STI acquisition, on bacterial STI incidence and detection in NC.” This research can hopefully support evolution of Doxy-PEP prescription guidelines and associated STI testing recommendations to ensure informed STI prevention implementation.
I’m excited about this project because it allows me to expand on my prior experience with modeling HPV/HIV interventions and analyzing DPH surveillance data, while learning more about bacterial STIs, network analysis, and network modeling.
Is there anything unique that you’ve learned or experienced? Or anything else you’d like to share.
I feel very fortunate for the opportunity to work with our DPH partners! I have learned so much about understanding STI/HIV surveillance data and how to think about the way forward for HIV and STI care in NC.