BACKGROUND: Dynamic models of HIV transmission have proven valuable tools for informing HIV prevention strategies. Including structural determinants in models is crucial to estimate their population-level impacts on HIV transmission and inform efforts towards HIV elimination. However, this is challenging due to a lack of coherent conceptual frameworks, limited understanding of their specific causal pathways, and few empirical estimates of their impacts on downstream mediators. METHODS: With the overarching aim to improve models, we conducted a scoping review of studies that used dynamic HIV transmission models to evaluate the impact of structural determinants. From included studies, we extracted information on the types of structural determinants and methods used to model their impacts on HIV transmission. We appraised studies on how they conceptualized structural exposures and represented their causal relationships over time within models. RESULTS: We identified 9 dynamic transmission modelling studies that incorporated structural determinants of HIV, including violence (N=3), incarceration (N=2), stigma (N=2), housing instability (N=2), migration (N=1), and education (N=1). Only one study modelled multiple determinants simultaneously. In most models, structural determinants were conceptualized using current, recent, non-recent and/or lifetime exposure categories. Modelled structural determinants largely impacted HIV transmission through mediated effects on one or more proximate risk factors, including sharing injection equipment, condom use, number of partners, and access to treatment. However, causal pathways were simplistic, with few mediators and/or lack of clear empirical justification. To measure impact, most studies simply assumed the elimination of structural determinants in counterfactual comparison scenarios. Few models included long-term and/or delayed effects of past, recurrent, or acute exposure, potentially overestimating impacts of determinants. CONCLUSIONS: Despite the importance of structural determinants for HIV prevention, methods for including them in dynamic HIV transmission models remain insufficient. Few studies have attempted to incorporate structural determinants in HIV models, and methods vary considerably. To improve inferences, models should adopt precise exposure definitions, deconstruct and estimate their complex causal pathways, and translate them into their mechanistic components. The need for development of coherent frameworks to conceptualize the synergistic interplay between strengthened empirical data analysis and the inclusion of structural determinants in dynamic models is pressing.