Background <p>Antimicrobial resistance (AMR) is a serious global public health problem, that contributed to an estimated 4.95 million deaths in 2019 and with approximately 10 million annual deaths and up to US$100 trillion in cumulative economic losses projected by 2050. Its emergence and spread result from complex interactions between biological, ecological, and socioeconomic factors. Mathematical modelling has been recognized as a crucial tool for clarifying the dynamics of AMR emergence and transmission. However, the dominant literature is fragmented and characterized by notable methodological and contextual limitations. This scoping review aims to synthesize and analyse recent mathematical modelling studies on AMR to identify prevalent trends, methodological biases, and key research gaps.</p> Methods <p>We conducted a scoping review following the PRISMA-ScR statement. We systematically searched three databases (PubMed, Web of Science, and Scopus) from 2019 – 2024 for published papers that created or used dynamic mathematical models of AMR. After removing duplicates and screening, 36 studies were considered eligible for inclusion. Data were extracted via a structured form that was divided into three categories: model type and context, model construction and correlated parameters, and model outputs and validation. In each category, the information considered most relevant for further analysis was extracted.</p> Results <p>Our analysis demonstrated a predominance of deterministic models using ordinary differential equations (ODEs), which were mostly focused on bacterial pathogens such as <i>Pseudomonas aeruginosa</i>, <i>Escherichia coli</i>, and <i>Staphylococcus aureus</i>. The vast majority of models focused on the human host, with only one study adopting a One Health approach. The most commonly modelled resistance mechanisms are horizontal transfer by conjugation and mutation, and the rarely modelled mechanisms include transduction, transformation, host immunity, and spatial heterogeneity. Furthermore, only two have considered economic impact. There was apparent consistency in geographic inequality, with the vast majority of studies originating from high-income countries.</p> Conclusion <p>Mathematical modelling of AMR is an active field, but is characterized marked by low methodological diversity and is limited in scope to a few contexts. Given these limitations, there is a need to develop mathematical models of AMR that are capable of capturing the complex dynamics among hosts, environments, transmission, and intervention dynamics. The use of deterministic models based on ODEs contributes significantly to advancements in the study of AMR dynamics, but future work requires the integration of stochasticity, spatial structure, and ecological interactions to more realistically represent the complexity of the real world. Furthermore, the introduction of a One Health framework and the incorporation of economic and social variables will be essential for the development of models that not only explain the observed patterns but also guide effective global strategies to mitigate the impact of AMR.</p>