Acute HIV Infection (AHI) occurs shortly after an individual is exposed to and infected with HIV. Although this period is believed to be short (at most 5 months in duration) compared to chronic infection, which can last for years or decades, those with AHI have been shown to be much more infectious. The contribution of this early stage to total incident infections among drug users is not known. Although previous studies have presented estimates for this value, as a proportion, for populations of serodiscordant couples and men who have sex with men, all have focused only on transmission from sexual exposures, and have not assessed incident infections arising from other modes, such as syringe sharing. To address this knowledge gap, this project will be the first to estimate the role of AHI in transmission of HIV among drug-using populations, including people who inject drugs (PWID). The role of AHI in overall transmission of HIV is crucial in understanding the limitations of ongoing prevention strategies, particularly those that require an individual to be diagnosed. One such prevention strategy is treatment as prevention (TasP), which typically involves frequent testing and early treatment of HIV positives with antiretrovirals to decrease their infectiousness. Since the vast majority of AHI cases go undiagnosed, prevention strategies should be informed by the role this early stage has in creating new infections, so that prevention modalities which are unable to avert transmission among undiagnosed individuals, such as TasP, are supplemented by those that can (e.g., needle and syringe programs, pre-exposure prophylaxis, opioid substitution therapy). In order to estimate the proportion of new infections resulting from AHI, this study will use an agent-based modeling approach that will track transmission events within a virtual population of agents, calibrated with previously collected empirical data. The model will be parameterized using surveillance and observational study data available for the New York City Metropolitan Statistical Area, which has some of the richest HIV empirical data in the world. The proposed agent-based model (ABM) will incorporate empirical data on the risk-behaviors and network structure of drug-using populations (both injection and non-injection) to estimate both the total number of new HIV infections in a given time period, and the proportion of those infections resulting from transmission during AHI. As a secondary aim, the ABM will also be used to estimate the effectiveness of TasP strategies under different sets of assumptions regarding characteristics of AHI (such as infectiousness and duration); this will allow researchers and policy makers to understand the potential limitations of this promising prevention modality.
Acute HIV infection (AHI) is a short but very infectious period of HIV disease in which many transmission events are thought to occur. This study will be the first to estimate the proportion of all new HIV infections that occur during AHI among drug users using a novel complex systems approach known as agent-based modeling. The results of this project will inform improved HIV prevention strategies for drug users, including the identification of AHI-specific interventions that best complement ongoing Treatment as Prevention programs.
