Topic Modeling Trends in First Responder Research

Despite decades of inquiry into first responder mental health, a critical gap remains in understanding how research priorities in this field have evolved. While the prevalence of burnout and PTSD is well-documented, theoretical frameworks often fail to account for how shifting occupational demands. Mass casualty events and pandemic responses are examples of occupational demands that have reshaped the academic discourse. The purpose of this study is to map the topical structure of first responder literature. Utilizing a specialized corpus of peer-reviewed article abstracts retrieved from the Web of Science Core Collection, this study employs Latent Dirichlet Allocation (LDA) via the Orange Data Mining Toolbox to uncover latent thematic structures. The study poses the following research question: (1) What is the topic structure present in first responder literature? To answer this, the study will determine the optimal number of topics (k) using log perplexity and topic coherence scores. This project aims to provide a data-driven taxonomy of the field and highlight critical gaps between current research focus and the evolving needs of the first responder workforce.