Methods. This was a systematized scoping review that queried 8 social science and clinical databases and yielded 903 citations which were then screened by two blind reviewers. Articles eligible for inclusion were published after 2000 (as RTM was developed in 2009 and there may have been some articles prior to its release), were available in English, and focused on a violent crime with an injury. A total of 29 articles met criteria. Data concerning the purpose of the RTM analysis, study location, place-based features and environmental risk factors, data sources, journal type, study design, crime(s) of focus, injury and publication year were extracted.

Results. No articles directly evaluated injury. Instead, RTM analyses for violent crimes indirectly reference injury due to the inherent naming of the crime (i.e., an assaultive injury is due to the crime of aggravated assault). Most publications leveraged local police department data (72.4%) and had a retrospective cohort study design (62%). Articles were published in criminal justice, geography, policing, social science, and violence-focused journals; no clinical or injury-focused journals. Assault (55.2%), homicide (44.8%), and gunshot wounds (20.7%) were the top three most common injuries inference from crimes of focus. Studies were largely conducted in the U.S. (86.2%) within a single city (86.2%) and utilized an average of 4 place features/risk-factors (e.g., bus stops, alcohol outlets, vacant lots) in their RTM models. The most common analytic purposes of RTM was to evaluate crime risk by place-based risk factors (45%), to compare RTM against other traditional geospatial analytic methods (21%), and to evaluate the effect of a psychosocial factor (e.g., seasonality, segregation, vacancy) on crime risk (17%).

Discussion. Leveraging RTM for injury risk prediction is possible due to its shared theoretical characteristics with crime. Both exhibit variation and clustering at different levels of geography over time and have place-based attractors and generators. This means that even injury unassociated with crime (e.g., unintentional motor vehicle crashes, falls) is suited for analysis by RTM. The ability to customize parameters, select place-based features important to both topic of interest and local context, and the application of weights to significant risk factors by relative risk values makes RTM capable of addressing questions beyond crime.

Conclusion. This work supports the use of RTM for expanded, multidisciplinary exploration of population-level injury risk. Such investigations can produce evidence to guide injury prevention initiatives, programming, and practice.