Getting to Zero (GTZ) and other HIV elimination initiatives have gained momentum in the United States following the UNAIDS strategic plan GTZ in 2010. These initiatives have typically focused on HIV elimination within local jurisdictions and been driven by a plateauing domestic HIV incidence combined with growing use of biomedical prevention modalities such as antiretroviral treatment (ART) initiation for treatment-as-prevention, and pre-exposure prophylaxis (PrEP). While overall HIV incidence in Illinois (and across the U.S.) has plateaued over the past decade, these declines have not been experienced equally by sub-populations. Younger (18-34 years) Black gay, bisexual and other MSM (YBMSM) have experienced relatively stable incidence rates. These observations support development of interventions designed to reduce HIV incidence among YBMSM. The scale-up of ART and PrEP among BMSM is constrained in part because of the many barriers that prevent their wider use among YBMSM. Addiction, mental health conditions, criminal justice involvement, and unemployment upon release from custody are some of the barriers that disproportionately impact YBMSM. Substance use is one such barrier, and it has been associated with suboptimal ART adherence and missed PrEP doses among MSM. Substance use interventions have been found to reduce behaviors that elevate risk for HIV acquisition, engagement in the continuums of ART care, and PrEP retention. Combination biomedical substance use interventions may therefore provide a pathway to decrease HIV incidence that may be more implementable than the current scenarios considered. It is difficult, however, to design evidence-based interventions that will optimally implement GTZ goals. Computational models can provide an in-silico laboratory to examine how the impact of such combination interventions can be maximized. Rigorous examination of how the impact of such combination interventions can be optimized should be used to inform best practices prior to intervention roll-out. Agent-based network models (ABNMs) provide a computationally rigorous, yet flexible, platform that allow for inclusion of a variety of micro-level individual behaviors, meso-level network factors, and their feedbacks with macro-level population structures, which allow for rigorous examination of processes and their parameterization at multiple scales. ABNMs can therefore promote development of multilevel interventions prior to their roll-out. In the proposed study, we will: (1) develop a comprehensive suite of parameters that describe the impact of stimulant use on ART and PrEP care continuums among YBMSM in Illinois; (2) recalibrate an existing ABNM of HIV transmission among YBMSM using the parameters collected above to develop modules that account for the impact of stimulant use on engagement in the ART and PrEP continuums; (3) estimate the potential efficacy of combination biomedical substance use interventions.
Younger (18-34 years) Black gay, bisexual and other MSM (BMSM) have experienced relatively stable incidence rates over the past decade even as overall HIV incidence in the U.S has declined. ART and PrEP use among BMSM need to be scaled up to reduce HIV-related disparities, but such scale-up is constrained because of addiction and other barriers that prevent wider ART and PrEP use. Combination biomedical substance use interventions may help increase ART and PrEP use among BMSM; here, we will develop computational agent-based network models (ABNMs) that provide an in-silico laboratory to rigorously examine how the impact of such combination interventions can be maximized.
