Non-fatal firearm violence across 17 US cities using police dashboard data, 2020–2025 Authors: Maureen R. Benjamins, Cameron De La Huerta, Nazia S. Saiyed, Madison Offstein, Melissa Gutierrez-Kapheim, Ashley Wolf Published: 2026 ABSTRACT ObjectiveNon-fatal firearm violence is a substantial, but often overlooked, public health burden in the United States (US); however, local data infrastructure remains limited. We identified and summarized city-level public dashboards reporting non-fatal firearm violence.MethodsIn this descriptive epidemiologic study, we searched for US cities with publicly available dashboards reporting local police-reported non-fatal firearm violence. We then compared violence counts, population-based rates, and trends from 2020 to 2025.ResultsOnly 17 cities, varying by size and geography, met the inclusion criteria. Rates of non-fatal firearm violence varied nearly eight-fold across cities in 2025, ranging from 13.3 per 100,000 (Seattle, Washington) to 106.0 (Portsmouth, Virginia). Annual non-fatal shootings ranged from 83 (Pittsburgh, Pennsylvania) to 1515 (Chicago, Illinois). Most cities saw a spike in non-fatal firearm violence during the peak of the COVID-19 pandemic (2021), followed by a decline through 2025.ConclusionsOnly 17 US cities had publicly available dashboards reporting police-reported non-fatal firearm violence. Given the substantial variation in rates across cities, publicly available local dashboards can strengthen firearm injury surveillance, enable city-level comparisons, and inform violence prevention policies and programs. Introduction Firearms play a large role in the United States (US), where over half of residents (54%) report that they or a family member have experienced a firearm-related incident (Schumacher et al., 2023). While most attention is paid to fatal outcomes, it is critical to also understand firearm-related injuries to fully capture the burden of firearm violence in the US. Notably, for every firearm death, an estimated two individuals survive with injuries (Kaufman et al., 2021). There were approximately 85,000 visits to emergency departments for firearm-related injuries each year in the US between 2009 and 2017 (Kaufman et al., 2021). These injuries range from minor wounds to severe multisystem trauma (Kaufman et al., 2021). They often lead to lasting disabilities impairing the ability to work and study, plus the psychological sequelae of trauma (Song et al., 2022; Zatzick et al., 2008). Additionally, family members and witnesses face elevated behavioral health risks (Song et al., 2022), including substance use, psychiatric disorders, depression, anxiety, and aggression, which can perpetuate cycles of violence and negatively impact entire neighborhoods (Kaufman & Delgado, 2022). Consequently, firearm violence imposes a substantial (and increasing) economic burden, at the national and local level. The estimated overall societal cost of non-fatal injuries was $9.3 billion in 2020 (Miller et al., 2024). Despite this, non-fatal firearm violence remains understudied, particularly at the local level, where there is not a standardized collection system for data (Barber et al., 2022; Braga, 2025; Kaufman & Delgado, 2022). It is critical to look specifically at cities because they are key political jurisdictions where many policies and programs are based. However, existing local data sources vary in accuracy. Hospital records often miss on-scene fatalities and only include data on their patients, while police data captures more shootings but may miss unreported injuries (Anchan et al., 2025; Barber et al., 2022; Kaufman et al., 2019). However, because the goal is to capture information at the city level, police data is the most relevant. Concerns about systematic undercounting have led to several data aggregation initiatives. The Gun Violence Archive collects data from over 7500 sources daily on fatal and non-fatal firearm injuries (Gun Violence Archive, 2026). Additionally, The Trace pulls Gun Violence Archive data into interactive maps and data (The Trace, 2026). While valuable, the data's ability to correctly identify cases varies widely by city (50.2% - 99.5% in 2020) (Gobaud et al., 2023). It also combines all gun-related deaths, including suicide, and lacks details about the incidents (e.g., victim demographics) and local geographic areas (e.g., neighborhoods, police districts). There are multiple studies of firearm injuries within individual cities, including studies using police (Kaufman et al., 2019; Semenza & Stansfield, 2021) and trauma registry data in Philadelphia (Kaufman et al., 2019), police and media data in Indianapolis (Magee et al., 2024), police data in New York City (Adams et al., 2025), and urban hospitals data in Chicago (Fitzpatrick et al., 2018; Wolf et al., 2026) and New Orleans (Silver et al., 2023; Winter et al., 2024). However, no analyses of non-fatal firearm violence across multiple US cities and over time were found. To address this, the current study identified US cities with publicly-available, police data dashboards with non-fatal firearm violence data. We compared overall rates on non-fatal gun violence and assessed trends over time (2020–2025). This type of local data can support the development (and evaluation) of tailored programs and policies, increase awareness of local issues, and allow for evidence-based allocation of resources at the local level. Methods Study design and population To identify cities for inclusion, we started with Everytown's list of “Cities with Real-Time Gun Violence Data” and the dashboard links they provided (Fig. 1) (Everytown, 2026). This list included cities with firearm homicide data from the Federal Bureau of Investigation's (FBI's) annual National Incident-Based Reporting System (NIBRS) and the Supplementary Homicide Report. It included all US cities with: (1) populations of 65,000 or more; and (2) a law enforcement agency that reported 12 months of data per year to the FBI. To look for additional cities, we conducted internet searches: (1) a general search using the search string: (non-fatal OR nonfatal OR injury) AND (firearm OR gun OR shooting) AND (city OR metro OR local) AND (data OR dashboard); (2) the same search string plus the names of the ten largest population cities in the US; and (3) the same search string plus the names of ten cities with the highest homicide rates based on 2024 Everytown data. After identifying these cities, we included only those that had dashboards that met specific criteria: (1) report non-fatal firearm violence data, (2) include data sourced from police departments, (3) include 2025 data, (4) publicly accessible through an interactive dashboard (with no registration required), and (5) available in English. One-time reports and downloadable datasets, as well as dashboards focused on states, counties, or other non-city jurisdictions, were excluded. Searches were conducted between May 28 and June 15, 2026. This study did not meet the definition of human subjects research as it involved only aggregate city-level and national data obtained from publicly available, anonymized sources. This study was reported in accordance with the STROBE guidelines for observational studies. Measures and statistical analysis Of the cities meeting the criteria above, violence data were sourced from the cities' data dashboards. National data was pulled from the Gun Violence Archive (on June 15, 2026) (Gun Violence Archive, 2026). The American Community Survey (5-year estimates) was used for population data. For 2025, 2024 population data were used, as newer data were not released yet. Rates were calculated using the counts of non-fatal, fatal, and total shootings, per 100,000 population. Time trends were described using the last five years of data; at least three years of data were available for all but one city. All analyses were done in Microsoft Excel. Results Existing city-level data sources on non-fatal firearm violence We found public data dashboards that included firearm violence for 17 US cities (Table 1). Most dashboards were located on police department websites; a few were hosted by partner organizations or other initiatives (Appendix Tables A1, A2). For example, Philadelphia data were available through the Philadelphia Center for Gun Violence, which cited its data source as Open Data Philly, a city initiative that provides crime data from the police department (Open Data Philly, 2026).Fig. 1Flow chart representing the search for cities with police data dashboards that included non-fatal violence data for 2025.Fig. 1Table 1Non-fatal firearm violence rates within US cities with publicly available data dashboards (2025).Notes¹ American Community Survey 5-year estimates, 2024, https://data.census.gov/. ² City data sourced from each city's respective crime dashboard; US data from the Gun iolence Archive; see Appendix Table 2 for details. Note that non-fatal shooting numbers are based on number of victims except for the following five cities, which only have data for incidents: Memphis, New Orleans, Pittsburgh, Portland, and Seattle. ³ Calculated using (# Non-fatal shootings/Census population) * 100,000. ⁴ Calculated using (# Fatal shootings/Census population) * 100,000. ⁵ Sum of fatal shootings and non-fatal shootings. ⁶ Calculated using (# Total shooting victimizations/Census population) * 100,000. Flow chart representing the search for cities with police data dashboards that included non-fatal violence data for 2025. Table 1Non-fatal firearm violence rates within US cities with publicly available data dashboards (2025).Notes¹ American Community Survey 5-year estimates, 2024, https://data.census.gov/. ² City data sourced from each city's respective crime dashboard; US data from the Gun iolence Archive; see Appendix Table 2 for details. Note that non-fatal shooting numbers are based on number of victims except for the following five cities, which only have data for incidents: Memphis, New Orleans, Pittsburgh, Portland, and Seattle. ³ Calculated using (# Non-fatal shootings/Census population) * 100,000. ⁴ Calculated using (# Fatal shootings/Census population) * 100,000. ⁵ Sum of fatal shootings and non-fatal shootings. ⁶ Calculated using (# Total shooting victimizations/Census population) * 100,000. Non-fatal firearm violence rates within US cities with publicly available data dashboards (2025). Notes¹ American Community Survey 5-year estimates, 2024, https://data.census.gov/. ² City data sourced from each city's respective crime dashboard; US data from the Gun iolence Archive; see Appendix Table 2 for details. Note that non-fatal shooting numbers are based on number of victims except for the following five cities, which only have data for incidents: Memphis, New Orleans, Pittsburgh, Portland, and Seattle. ³ Calculated using (# Non-fatal shootings/Census population) * 100,000. ⁴ Calculated using (# Fatal shootings/Census population) * 100,000. ⁵ Sum of fatal shootings and non-fatal shootings. ⁶ Calculated using (# Total shooting victimizations/Census population) * 100,000. City-level firearm violence and death data The cities ranged in size from under 100,000 to nearly 3 million. They were located across the US, with limited representation of the Southwest and Southeast. The number of non-fatal shootings reported in 2025 across these cities ranged from 83 (Pittsburgh, Pennsylvania) to 1515 (Chicago, Illinois). Together, these cities had 23% of the total non-fatal violence incidents nationally in 2025 (26,295). The number of non-fatal shootings per 100,000 population ranged from 13.3 (Seattle, Washington) to 106.0 (Portsmouth, Virginia). All 17 cities had a higher rate than the US as a whole (7.9). All but one city also included data on fatal shootings (or homicides) on their dashboards. The highest number of fatal shootings was reported in Chicago, Illinois (348), followed by Houston, Texas (331). Milwaukee, Wisconsin had the highest rate among these cities in 2025 (24.9 per 100,000), followed by Baltimore, Maryland (23.4) and Memphis, Tennessee (22.8). Fatal shootings in these cities made up approximately 4% of the national total (38,803). The total number of firearm injuries (non-fatal and fatal) was highest in Chicago, Illinois (1863), followed by Houston, Texas (1093). The highest rate was in Portsmouth, Virginia (126.6 per 100,000), followed by Milwaukee, Wisconsin. All of the cities except for Seattle, Washington had higher overall rates of firearm violence than the US (19.4). The non-fatal to fatal ratio indicates that more non-fatal injuries were reported than fatal ones in all 16 cities with complete data (in contrast to the US). Four cities – Denver, Colorado; Rochester, New York; Portsmouth, Virginia; and Seattle, Washington - experienced at least five non-fatal shootings for each fatal shooting. Houston, Texas and Baltimore, Maryland had the lowest non-fatal to fatal ratios (2.3). A related concept, the fatality rate, ranged from 14% in Denver, Colorado to 46% in Baltimore, Maryland. Changes in city-specific non-fatal firearm violence over time Ten cities had data on non-fatal violence for all years, 2020–2025, while all 17 had at least two years of data (Fig. 2). The general trend was an increase in the non-fatal injury rate during the height of the COVID-19 pandemic (in 2021), but an overall decrease between 2020 and 2025. Seven cities saw decreases of at least 50%. Rates in Seattle, Washington and Memphis, Tennessee did not decrease between 2020 and 2025, though post-COVID rates for both were trending in a positive direction.Fig. 2Changes in the non-fatal firearm violence rate in select US cities over time (n = 17, 2020–2025).Fig. 2 Changes in the non-fatal firearm violence rate in select US cities over time (n = 17, 2020–2025). Discussion Firearm violence was responsible for over 38,000 deaths and 26,000 injuries in the US in 2025 alone. Despite advances in data aggregation, surveillance of non-fatal firearm violence remains incomplete, hindering efforts to understand the full scope of firearm victimization, particularly in cities. This study adds to existing literature by summarizing and comparing the most recent publicly available, police-reported non-fatal firearm violence data for 17 US cities, including trends over time. The cities' dashboards revealed substantial variation in non-fatal firearm violence rates in 2025, and city rates consistently exceeded the national rate. Fortunately, most of the cities experienced declines in non-fatal firearm violence between 2020 and 2025, including nearly 50% reductions in about half of them. These cities only represent a small fraction of the US population. Large cities with significant firearm violence burdens such as Chicago and Baltimore had dashboards, but many cities with similar per capita firearm violence burdens (St. Louis, Detroit, Birmingham) did not (Gun Violence Archive, 2026). Further, our search did not locate dashboards with non-fatal violence data for either New York City or Los Angeles, the two most populous US cities. Cities represented in this study span the US, but we identified a paucity of firearm violence dashboards from cities in the Southeast and Southwest. Notably, we did not find data dashboard for cities within Mississippi or Alabama, two states with high per capita firearm violence (The Trace, 2026). The variation in city-level representation is likely multifactorial. Data dashboards require funding to build and maintain, and cities with higher violence rates or a focus on violence prevention, may be more inclined to invest in these data aggregation tools (Butts et al., 2015; Casas Fuentes et al., 2026). Cities with legislative efforts around firearm violence, such as Seattle, may also be more likely to fund and maintain dashboards (Rowhani-Rahbar et al., 2020). Recent reductions in federal funding for public health research, staffing cuts at the Centers for Disease Control and Prevention (CDC) and other agencies, and the disruption of federal datasets will likely affect the surveillance and dissemination of violence data. The incidents captured in this study represented 23% of non-fatal firearm violence in the US in 2025, likely because firearm violence is disproportionately concentrated in large urban areas. We found substantial variation in non-fatal firearm injury rates across cities, ranging from 13.3 to 106.0 per 100,000 residents, an almost eightfold difference. Some of the cities with relatively high rates of non-fatal firearm violence included Baltimore, Chicago, and Philadelphia; all have historically suffered high rates of firearm violence and become leaders in firearm violence prevention efforts such as community violence interruptors (Butts et al., 2015; Cosey-Gay et al., 2023; Flowers, 2025; Zinny et al., 2024). The city with the highest per-capita rate of non-fatal firearm violence and a dashboard was Portsmouth, Virginia. While their rate was almost double that of Chicago and Baltimore, it reflected only 103 non-fatal shootings. Other small cities with high levels of violence per capita would do well to follow their example in proactively tracking incidents to increase awareness, inform programming, and monitor trends. The dashboards identified for this study captured less than 4 % of fatal firearm violence in the US in 2025, a much smaller percentage than captured for non-fatal incidents. Fatal firearm violence, particularly self-inflicted injury or suicide, has different risk factors and is more geographically dispersed, affecting rural persons at high rates (Reeping et al., 2023). In 2023, over half of fatal firearm injuries were self-inflicted, most of which were excluded from the crime data used here. This may also contribute to this study's findings that all cities with dashboards had non-fatal to fatal ratios greater than one, representing a higher volume of non-fatal injuries compared to fatal ones. This contrasts with the US as a whole, which has a non-fatal to fatal ratio of 0.7. Examining trends and comparing metrics in these 17 cities and the US overall enables us to better understand the whole spectrum of US firearm violence and where there is a greater concentration of non-fatal firearm violence. Almost all cities demonstrated a decrease in non-fatal firearm violence from 2020 to 2025. This is consistent with research from multiple cities and the US overall demonstrating a recent decrease in fatal firearm violence (The Trace, 2026). In 2024, the US Surgeon General declared firearm violence a public health crisis, and significant federal funding and attention was directed to firearm violence prevention. Many cities were able to expand, strengthen, and evaluate community violence intervention efforts. For example, an evaluation of Chicago CRED (Create Real Economic Destiny), a street outreach–based community violence intervention program that provides intensive life coaching, education, and workforce supports, found that it significantly reduced the likelihood of criminal legal involvement for participants (Semenza et al., 2023). Additionally, firearm purchasing, leading to firearm access and carrying, both risk factors for injury, increased during the COVID-19 pandemic, but decreased substantially over the 2020–2025 time period (Miller et al., 2024; Schleimer et al., 2021). To attempt to address the challenge of fragmented data and dashboards, organizations such as the Gun Violence Archive and The Trace have aggregated multiple data sources to create a national data dashboard. Recent studies leveraging the Gun Violence Archive and datasets derived from it, such as the American Violence Project, have demonstrated the feasibility of examining fatal and non-fatal shootings across large samples of US cities and neighborhoods (Lyons et al., 2022; Semenza et al., 2023; Semenza et al., 2024); however, these aggregated data sources still depend on heterogeneous public reporting systems and do not eliminate underlying variation in how jurisdictions define, collect, and report non-fatal firearm violence. While these national dashboards represent a critical step forward, they still miss significant amounts of injuries, estimated at more than one-quarter of deaths for the Gun Violence Archive (Anchan et al., 2025). In conducting this study, we compared non-fatal gun violence injury totals in the study cities to those reported by Gun Violence Archive and compiled on The Trace and found that while victim counts were similar in some cities, the number of nonfatal injuries reported by the Gun Violence Archive for 2025 ranged from a low of 43% of the total reported in Portsmouth, Virginia to a high of 109% of the total in Rochester, New York (The Trace, 2026). Others have already investigated differences between this archive and police data for capturing injuries related to gun violence, and found differences may arise due to the inclusion of officer-involved shooting injuries and due to the use of media reporting to compile data, which may have resulted in a bias toward the inclusion of shooting incidents with multiple injured individuals and those involving women or children (Anchan et al., 2025; Gobaud et al., 2023). Health surveillance systems such as the CDC's National Syndromic Surveillance Program have been utilized to collect health records at near real-time speed, though it is limited by hospital participation (62% in 2019) and lacks city-level granularity (Zwald et al., 2022). The FBI's NIBRS includes information on all crime incidents at the national and local level. However, city-level data were not available in an easily accessible format. Researchers have called for the development of comprehensive, national or state-sponsored databases documenting these incidents (Anchan et al., 2025; Barber et al., 2022). This may be an opportunity for academic or research institutions to lead in strengthening firearm injury surveillance. These institutions have the ability to integrate disparate data sources and develop standardized definitions that could improve data quality, and they also serve as neutral conveners across police departments, hospitals, and community organizations, with the ability to enable data-sharing agreements that are often difficult for city agencies to establish independently. The analysis was restricted to cities with publicly available dashboards reporting non-fatal firearm violence from police data, which reduces generalizability beyond the studied cities. To reduce selection bias, we used predefined inclusion criteria and a standardized search strategy. However, we may have missed some cities with relevant dashboards as many are not easily discoverable through standard search terms. Several appear embedded within police transparency portals, public health dashboards, or open data systems. Beyond finding the dashboards, many are difficult to navigate and do not include adequate documentation or definitions. For most cities, the available data provided limited information on incident characteristics or context. Variables such as firearm type, incident intent or circumstances (e.g., intimate partner violence), location (e.g., school or home), time of occurrence, number of people injured, and association with other reported offenses (e.g., robbery) were unavailable or inconsistently reported. These differences in data definitions and reporting systems may affect comparability between cities. Additionally, many cities have more limited forms of dissemination, including annual reports, presentations, webinars, and practitioner-facing incident notification systems, such as Chicago's Crime Prevention Information Center alerts to community violence intervention workers (Bocanegra & Aguilar, 2025), which were not included in our study. Future research should systematically examine and compare the range of state and local approaches used to disseminate firearm violence data. Further, the dashboards relied on police data, which may underestimate the number of injuries (and deaths). Finally, the data for many cities changed on their respective dashboards throughout the study period, particularly within the most recent year of data, highlighting the dynamic (and likely imprecise) nature of this type of data. Finally, rates are not age-standardized. Despite these limitations, this study has several strengths. It provides one of the first large-scale comparisons of non-fatal firearm violence counts, rates, and trends across U.S. cities using publicly available police data dashboards. By systematically identifying and synthesizing data from 17 cities spanning a wide range of population sizes and geographic regions, this analysis demonstrates the feasibility of using local data systems to improve firearm violence surveillance. These data can inform and evaluate local prevention programs and policies while enabling cross-city comparisons that identify effective strategies and best practices. Expanding firearm surveillance to include non-fatal violence, not just deaths, can support more comprehensive policies that reduce the overall burden of firearm violence on individuals, families, and communities (Wolf et al., 2026). Conclusions Non-fatal firearm violence remains a substantial and unevenly distributed public health burden in the United States. This analysis provides one of the first large-scale comparisons of non-fatal firearm violence across US cities. Given limited resources, high-quality local data are needed to identify communities most affected by firearm violence and guide prevention efforts. Strengthening systems to collect and standardize local firearm injury data is essential to enable cross-city comparisons, improve surveillance, and inform effective, sustained violence prevention efforts. CRediT authorship contribution statement Maureen R. Benjamins: Writing – review & editing, Writing – original draft, Visualization, Supervision, Project administration, Methodology, Formal analysis, Data curation. Cameron De La Huerta: Writing – review & editing, Writing – original draft, Visualization, Methodology, Data curation. Nazia S. Saiyed: Writing – review & editing, Writing – original draft, Supervision, Methodology. Madison Offstein: Writing – review & editing, Writing – original draft, Methodology, Funding acquisition. Melissa Gutierrez-Kapheim: Writing – review & editing, Project administration, Funding acquisition, Data curation, Conceptualization. Ashley Wolf: Writing – review & editing, Writing – original draft, Methodology, Data curation, Conceptualization. Funding This research was, in part, supported by staff working on the Chicago Gun Violence Research Collaborative (CGVRC) Fellowship. The CGVRC Fellowship is supported by the Shure Charitable Trust, a supporting foundation of the Jewish Federation of Metropolitan Chicago. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.