Abstract
Background:
Gunshot wounds (GSWs) remain a significant source of mortality in the United States. Timely delivery of trauma care is known to be critical for survival. We sought to understand the relationship of predicted transport time and death after GSW. Given large racial disparities in firearm violence we also sought to understand disparities in transport times and death by victim race, an unstudied phenomenon.
Methods:
Firearm mortality data were obtained from the Boston Police Department 2005–2023. Firearm incidents were mapped using ArcGIS. Predicted transport times for each incident to the closest trauma center were calculated in ArcGIS. Spatial autoregressive models were used to understand the relationship between victim race, transport time to a trauma center and mortality associated with the shooting incidents.
Results:
There were 4,545 shooting victims with 758 deaths. Among those who lived, the median transport time was 9.4 minutes (IQR 5.8, 13.8) and 10.5 minutes (IQR 6.4, 14.6, p=0.003) for those who died. In the multivariable logistic regression, increased transport time to the nearest trauma center (OR 1.024, 95% CI 1.01–1.04) and age (OR 1.016, 95% CI 1.01–1.02) were associated with mortality. There was a modest difference in median transport time to the nearest trauma center by race with non-Hispanic Black at 10.1 minutes, Black Hispanic 9.2 minutes, white Hispanic 8.5 minutes, and non-Hispanic white 8.3 minutes (p<0.001).
Conclusions:
Our results highlight the relationship of transport time to a trauma center and death after a GSW. Non-white individuals had significantly longer transport times to a trauma center and predicted mortality would have been lower with white victim transport times. These data underscore the importance of timely trauma care for GSW victims and can be used to direct more equitable trauma systems.
Level of Evidence:
Level III, Prognostic/Epidemiological
Keywords: Firearm Violence, Trauma Systems, Trauma Care, Equity
Background
Firearm violence remains a source of significant morbidity and mortality in the United States. In 2020, there were nearly 20,000 firearm homicides here.1 Intuitively, quicker access to a trauma center, and thus definitive hemorrhage control or other life-saving intervention, is a significant factor in survival after a gunshot wound (GSW).2–5 GSWs are particularly lethal injuries and hypotensive shooting victims with 10 minute delays or more to the operating room have an associated three-fold increase in death.2 Many delays to hospital based intervention may occur long before the patient arrives at the hospital and a number of studies have evaluated distance to care a surrogate for time to care.
Particularly within urban areas, distance alone may not predict transport times due to traffic and other extant factors. Few studies examining transport time for urban firearm injuries exist and one from Philadelphia, which has a unique transport system relying on police vehicles, showed that up to 23% of firearm fatalities were attributable to differences in transport times.5 There is, in general, a paucity of research on the relationship of transport time and survival, particularly after a shooting. Demonstrating the contribution of transport time to death after GSW could provide evidence-based emphasis on alternative means of transportation, such as adoption of more rapid police transport, in geographies demonstrating such a relationship.
Further, there are substantial racial disparities in gun violence. Black individuals are disproportionately affected by these deaths, accounting for over 50% of all firearm homicide victims despite making up only 13% of the population.6 In fact, homicide remains the leading cause of death for Black men under the age of 45. While not as stark, Hispanic men also have significantly higher rates of firearm injuries when compared to non-Hispanic white men.7 Residential racial segregation can contribute to disparate access to healthcare and may impact time to care. With the Great Migration and subsequent historical segregation practices, Black individuals have been relegated to geographically isolated areas of urban centers.8 Studies have shown associations between geographic distribution of minoritized groups, structurally racist practices, and firearm violence incidence as well as the downstream socioeconomic impacts that mediate these geographic associations.9–11 Similar geographic isolation is seen among other racial and ethnic minority groups, namely Hispanic populations in urban areas, but their underlying historical factors are different than those for Black Americans.12,13 Little is known about how such geographic isolation impacts access to trauma care and survival after the firearm injuries which disproportionately affect these communities.
This study sought to understand the relationship between transport time to urban Boston trauma centers and survival after a GSW as well as to identify racial and ethnic disparities in temporal access to trauma center care and survival. We hypothesized that increased predicted travel time to the nearest trauma center would be associated with death after a gunshot wound and that non-white populations would have longer predicted times to trauma center care.
Methods
Data Collection
To perform this cross-sectional study, we obtained data from a number of sources. First, we obtained shooting data including outcomes, type of shooting (multiple or single victim), demographics (gender, age, race and ethnicity), date of shooting and location of shootings from the Boston Police Department. Victims aged ≥14 were included as younger patients are not typically kept at adult centers in the city. While some young teenagers may very well be brought to pediatric centers, the trauma center closest to by far the most shootings is also a pediatric trauma center and would typically care for shooting victims much younger (and older) than this. The only stand alone children’s hospital is located in close proximity to and ringed by the other centers in an area with sparse shootings. Shooting data spanned all available Boston Police Data from January 1st, 2005 to March 31st, 2023. Boston trauma center locations were obtained from the American College of Surgeons listing of trauma centers and Boston Emergency Medical Services (BEMS) station locations from the BEMS website. STROBE guidelines for cross-sectional studies were followed in the preparation of this manuscript (Supplemental Digital Content, http://links.lww.com/TA/D426). Institutional review board approval was obtained.
Database Descriptions
Boston Police Department (BPD) Data
Data included only shootings occurring within the city of Boston and thus within the BPD’s jurisdiction. This dataset does not include self-inflicted gunshot wounds nor shootings the BPD has determined to be legally justified as per department policy. These determinations are updated based upon analysis by the BPD’s Boston Regional Intelligence Center.
US Census Bureau Topographic Integrated Geographic Encoding and Referencing (TIGER)/Line Shapefiles Database
This database houses selected geographic and mapping information from the US Census Bureau’s master address and TIGER database.
Data Analysis and Methodology
We utilized multiple statistical methods for this study. Police data were entered and locations mapped using provided coordinates overlayed on the shapefile of Boston using ArcGIS Pro (Esri; Redlands, CA). Hospital addresses and BEMS station addresses were geocoded using the ArcGIS World Geocoding Service. Population characteristics by census tract were also mapped and then spatially joined to the shooting data.
For each shooting, we utilized route analysis tools in ArcGIS to calculate predicted transport time (in minutes) and transport distance (in kilometers) from the shooting location to the nearest trauma center. This methodology uses typical road conditions, speed limits and average traffic based upon historical data to generate typical drive times from a given location to another. All nearest trauma centers were in Boston. As Tufts Medical Center did not obtain trauma center verification until February 2009, shootings prior to this date did not include it as a destination and instead utilized only the other four trauma centers. None of these 34 shootings were excluded from analysis. The same is true of the co-located BEMS station which opened in 2010 and the same analysis was performed for BEMS stations, but utilizing a route analysis from the stations to the shooting locations. Both trauma centers and EMS stations were assigned a mask (an assigned number) to avoid identifying specific centers and stations in our results.
Data were then abstracted from ArcGIS and Stata 18.0 (StataCorp LLC, College Station, TX) utilized for the remainder of the statistical analysis. Our primary outcome of interest was death after a shooting occurred. Our primary exposure of interest was predicted transport time to the nearest trauma center.
Univariate logistic regression was performed and all predictor variables significant at p<0.20 were then entered into a final multivariable model. Year was included in the model to account for changes in care over the long time period. To test for spatial confounding, model’s residuals were then assessed for spatial autocorrelation using a global Moran’s I test. This initial model showed statistically significant residual spatial autocorrelation (index 0.04, p<0.001). To address this, a spatial weights matrix for the shooting locations was created using inverse distance weighting and spectral normalization. We then performed a spatial autoregressive logit model and assessed the residuals for spatial autocorrelation.15 Finding no residual spatial autocorrelation (index < −0.001, p=0.753), this became the final model. Statistical significance of the final model was defined as p≤0.05 for multivariable assessment.
To assess for possible racial inequity in deaths associated with disparate access to care, we used our final regression model to predict mortality outcomes under a counterfactual scenario in which every shooting’s transport time was set to the median transport times of white non-Hispanic individuals. We stratified by race/ethnicity and compared actual mortality outcomes to these simulated outcomes.
Additional comparisons were made by Chi2 tests for proportions and Mann Whitney U or Kruskal Wallis tests for non-parametric continuous data.
Results
Over the study period, there were 4,545 shooting victims, including 758 deaths (Figure 1, Table 1). In all, 1,329 (29%) of victims were injured in a shooting with at least one other victim. There was no significant difference in death (16.1% vs 16.9%) for these multi-victim shootings. Overall, 9.6% of victims were female and the median victim age was 25 years (IQR 20, 31). The median predicted transport time from the nearest BEMS station was 4.16 minutes (IQR 3.2, 5.1) and 1.6 kilometers (IQR 1.2, 2.0) and from the nearest trauma center was 9.51 minutes (IQR 5.8, 14.0) and 3.9 kilometers (IQR 2.2, 7.0). These results can be found in Table 1. Mortality rates increased approximately linearly with increasing transport times (Figure 2).
Figure 1:
Geographic distribution of survivors and non-survivors of shootings in Boston mapped over predicted trauma center transport times.
Table 1:
Overall Characteristics of Shooting Survivors and Non-Survivors.
| Variable | Survivors | Non-Survivors | Significance |
|---|---|---|---|
| Total | 3,787 (83.3%) | 758 (16.7%) | - |
| Median Age (IQR) | 24 (20, 30) | 25 (21, 32) | p=0.001* |
| Gender | |||
| Male | 3,454 (91.3%) | 691 (91.2%) | - |
| Female | 330 (8.7%) | 67 (8.8%) | p=0.900 |
| Race and Ethnicity | |||
| Victim Black Non-Hispanic | 2,370 (77.6%) | 513 (78.3%) | - |
| Victim Black Hispanic | 230 (7.5%) | 38 (5.8%) | - |
| Victim White Hispanic | 343 (11.3%) | 80 (12.2%) | - |
| Victim White Non-Hispanic | 111 (3.6%) | 24 (3.7%) | p=0.436 |
| Multi-victim Shooting | 1,115 (29.4%) | 214 (28.2%) | p=0.512 |
| Most Common Nearest Masked Trauma Center (TC), (%) | TC 1 (47.4%) | TC 1 (45.0%) | p=0.221 |
| Most Common Nearest Masked BEMS Station (AS), (%) | AS 1 (22.1%) | AS 4 (22.6%) | p=0.027* |
| Median Predicted Transport Time to a Trauma Center in Minutes (IQR) | 9.4 (5.8, 13.8) | 10.5 (6.4, 14.6) | p=0.003* |
| Median Predicted Transport Distance to a Trauma Center in Kilometers (IQR) | 3.8 (2.2, 6.9) | 4.2 (2.4, 7.9) | p=0.005* |
| Median Predicted Time from an Ambulance Station in Minutes (IQR) | 4.1 (3.2, 5.1) | 4.2 (3.2, 5.1) | p=0.322 |
| Median Predicted Distance from an Ambulance Station in Kilometers (IQR) | 1.6 (1.2, 1.9) | 1.6 (1.2, 2.0) | p=0.101 |
Denotes significance at the p<0.05 level.
BEMS: Boston Emergency Medical Services
IQR; interquartile range.
Figure 2:
A scatterplot of shooting mortality and predicted transport time (in four minute intervals) to the closest trauma center.
Those who died had a predicted transport time to the nearest trauma center of 10.5 minutes (IQR 6.4, 14.6) while those who survived had a predicted transport time of 9.4 minutes (IQR 5.8, 13.8, p=0.003, Table 1).
Univariate Analysis
In univariate logistic regression on the outcome of mortality, victim race (OR 0.933, 95% CI 0.741–1.180, p=0.559), ethnicity (OR 0.959, 95% CI 0.771–1.200, p=0.713), gender (OR 0.986, 95% CI 7.500–1.300, p=0.918), time from the nearest ambulance station (OR 1.03, 95% CI 0.972–1.090, p=0.326) and multi-victim shootings (OR 0.94, 95% CI 0.793–1.120, p=0.504) were not associated with mortality and did not meet the inclusion threshold for multivariable modeling. Travel time to a trauma center (OR 1.03, 95% CI 1.012–1.045, p=0.001), victim age (OR 1.02, 95% CI 1.007–1.023, p<0.001), year (categorical, eg 2010 OR 1.59, 95% CI 1.046–2.420, p=0.030), trauma center (trauma center 4 OR 1.26, 95% CI 1.010–1.579, p=0.044) and nearest BEMS station 15 (OR 2.40, 95% CI 0.786–7.203, p=0.125) met inclusion thresholds.
Multivariable Analysis
In our logistic spatial autoregressive model, predicted transport time per minute to the nearest trauma center (OR 1.024, 95% CI 1.010–1.043, p=0.010), age (OR 1.016, 95% CI 1.010–1.024, p<0.001) and categorical year (eg 2010 OR 1.60, 95% CI 1.128–1.064, p =0.013) were significantly associated with death after a shooting.
Access to Care and Outcomes by Race
Of shooting victims with a known race, 77.4% were Black alone, 7.5% were Black Hispanic, 11.4% were white Hispanic and 3.6% were white alone and 0.1% were another race. Death rates were similar in unadjusted analysis (p=0.436). Non-white victims were younger than their white counterparts. This data can be found in Table 2.
Table 2:
Race Specific Metrics.
| Characteristics | Black Non Hispanic | Black Hispanic | White Hispanic | White Non Hispanic | Significance |
|---|---|---|---|---|---|
| Median Age (IQR) | 25 (20, 31) | 24 (20, 30) | 24 (20, 31) | 28 (21, 37) | p=0.002* |
| Male Gender N, % | 2,621 (91.0%) | 240 (89.6%) | 392 (92.7%) | 118 (88.1%) | p=0.318 |
| Multi-victim Shooting N, % | 814 (28.2%) | 74 (27.6%) | 114 (27%) | 32 (23.7%) | p=0.672 |
| Most Common Nearest Masked Trauma Center (TC) N, % | TC1 1,447 (50.2%) | TC 1 149 (55.6%) | TC 1 207 (48.9%) | TC 1 54 (40.0%) | p=0.029* |
| Most Common Nearest Masked BEMS Station (AS) N, % | AS 1 631 (21.9%) | AS 1 61 (22.8%) | AS 1 83 (19.6%) | AS 1 27 (20.0%) | p=0.617 |
| Median Predicted Transport Time to a Trauma Center in Minutes (IQR) | 10.1 (6.1, 14.4) | 9.2 (5.3, 12.7) | 8.5 (5.3, 11.7) | 8.3 (5.3, 13.3) | p<0.001* |
| Median Predicted Transport Distance to a Trauma Center in Kilometers (IQR) | 4.1 (2.3, 7.9) | 3.5 (2.0, 5.3) | 3.3 (1.9, 5.1) | 3.3 (1.8, 6.0) | p<0.001* |
| Median Predicted Time from an Ambulance Station in Minutes (IQR) | 4.1 (3.2, 5.0) | 4.1 (3.2, 5.1) | 4.3 (3.3, 5.3) | 4.1 (2.9, 5.4) | p=0.531 |
| Median Predicted Distance from an Ambulance Station in Kilometers (IQR) | 1.6 (1.2, 1.9) | 1.6 (1.2, 2.0) | 1.7 (1.2, 2.0) | 1.4 (1.0, 2.1) | p=0.090 |
Denotes significance at the p<0.05 level
IQR; interquartile range.
BEMS: Boston Emergency Medical Services
Eighty two percent of Black non-Hispanic (p=0.014), 6% Black Hispanic (p=0.031), 8% white Hispanic (p<0.001) and 3% non-Hispanic white (p=0.035) victims had predicted transport times above the overall median of 9.5 minutes.
The median transport time to the nearest trauma center for Black non-Hispanic victims was the highest followed by Black Hispanic, white Hispanic and finally white non-Hispanic victims (p<0.001). A depiction of shooting location hotspots, the median center shooting location and the predicted transport time to the nearest trauma center can be found in Figure 3. The majority of each race went to the same trauma center and were closest to the same EMS station.
Figure 3:
Hotspots of shootings across the city by race with predicted transport times. The median center of all shooting locations is also depicted for each race.
Using our spatial autoregressive model outputs with the predicted transport time for each racial group replaced with the median transport time for white non-Hispanic victims, Black non-Hispanic victim predicted mortality dropped from 17.8% to 16.9% (a difference of 25 deaths), Black Hispanic victim mortality dropped from 15.2% to 14.9% (a difference of one death) and white Hispanic mortality dropped from 19.4% to 18.8% (a difference of 3 deaths).
Discussion
In this study, we found a significant association between an increase in predicted transport time to a trauma center and the odds of mortality after a GSW in Boston. Additionally, we identified modest but statistically significant racial and ethnic inequity in temporal access to life-saving trauma care. Black non-Hispanic shooting victims faced the longest transport times to a trauma center followed by Black Hispanic victims, white Hispanic victims and finally white non-Hispanic victims. The current study is novel in its use of spatial autoregressive models to understand the relationship between travel time to a trauma center, racial disparities in access to trauma care, and gunshot wound mortality in a large database of firearm assaults in Boston.
Our results align with previous studies showing higher mortality from firearm injury with increasing distance from a trauma center and build on this by testing transport time, which has rarely been evaluated and which we considered a more suitable measure of trauma center access.16–21 We show that increasing transport time to a trauma center is associated with a 2% increased odds of mortality per minute. Karrison, et al in their study of patients in Chicago showed a significantly increased mortality for trauma victims in general with increased transport time, corresponding to a 0.26% increased odds of mortality per 2 minute increase.22 This suggests, based on our substantially larger effect size, that transport time may be even more important for shooting victims. This concept is further reinforced in a study of hypotensive patients with penetrating chest trauma who had increased mortality with increased transport time, and another study showing prolonged scene time and transport time associated with higher 24 hour mortality for all patients.23,24 The only study using similar geospatial methodology to ours had similar findings (death OR 1.03 per minute of transport time).5 However, that study had a substantially lower median predicted transport time (5.6 minutes as opposed to 9.5 minutes in Boston), perhaps reflective of differing trauma center distributions, shooting hot spots or typical traffic patterns in the two cities.5 It had similar mortality (20% in Philadelphia and 17% in Boston).5 While intuitive, the relationship of transport time and death highlight the importance of prehospital time as a modifiable determinant of death after a GSW.2 For Boston in particular, Level 1 trauma centers are concentrated within the city center without even distribution throughout the outlying areas, potentially contributing to geographic inequities in access to trauma care and patients of all races in some areas at risk for unnecessary death.25
The current findings reinforce many others demonstrating a strikingly higher rate of firearm victimization among Black and Hispanic (both Black and white) individuals when compared to white non-Hispanic people.1,26 We also find that non-white shooting victims in Boston have modestly longer transport times when compared to white victims. Based on our predictive modeling, around 30 non-white shooting deaths may have been related to inequities in trauma center access during the study period. The etiology behind these disparities is manifold and are grounded in the socioeconomic implications of historic segregation and structural racism that have locked Black communities in areas of cities that are politically stripped of resources and isolated newer immigrant communities relegated to poverty.9,12 Studies of Philadelphia and Boston have shown higher rates of firearm violence among redlined areas, which was shown to be influenced by downstream socioeconomic factors.9,27
While not as stark, similar disparities have been seen in Hispanic communities, with shooting victimization being over three times higher among Hispanic individuals when compared to non-Hispanic white.1 Studies evaluating Hispanic victims typically include both Black and white Hispanic individuals which is in contrast to our more nuanced evaluation of Black Hispanic and white Hispanic victims. Though not studied directly in the context of firearm victimization, Hispanic segregation can have detrimental effects on the health of US-born Hispanic individuals, partly mediated through neighborhood poverty.12 Our results build on this by showing intersectional disparities of race and ethnicity with Black Hispanic individuals having a higher transport time to hospitals when compared to white Hispanic, highlighting the interplay between structural racism focused on race and issues of immigration and settlement.
The concept of trauma deserts arose as an offshoot of food deserts whereby socially marginalized neighborhoods lack ready access to reliable trauma care, particularly capable trauma centers.3,28 Though not studied here, the location of urban hospitals may also be driven by the racial distribution of socioeconomic assets within a city. This aligns with previous studies showing that Black individuals are more likely than white individuals to live in trauma deserts in two major metropolitan areas studied.29 There were differing findings for Hispanic communities, though without our Black and non-Black Hispanic delineations, which were not found to be situated in trauma deserts in the three studied cities.29 While these studies showed the geographic distance by which individuals are separated from trauma centers, disparate outcomes after trauma were not studied. Our findings differed and showed that shootings of Black, Black Hispanic and white Hispanic individuals are, on average, situated further from trauma centers in Boston. Despite differences in overall transport times for non-white and non-Hispanic shooting victims, there was no significant difference in overall mortality rates between Black and white individuals in our study, which may be related to the relatively small difference in transport times and other covariates that differ between groups – in particular, significantly younger ages of Black, Black Hispanic and white Hispanic victims which may portend improved survival despite longer transport times. The impact of age may be due to fewer comorbidities and better physiologic reserve of younger patients although the effect of small differences in age is unclear. In simulated results utilizing the transport times of white victims for all other races/ethnicities, the predicted mortality was lower for all non-white groups.
Limitations:
There are important limitations to our study. It represents a retrospective cohort and causal inference cannot be gleaned. Similarly, race and ethnicity were reported by the police, and may be inaccurate. Our distinction of Black Hispanic and white Hispanic may limit comparison to other studies which typically denote only Black, Hispanic or white. While mortality data is included in the dataset, there is no available information on injury severity or anatomic location.
We have no reason to think that overall injury patterns or severity would vary with distance or time to care, but we cannot assess this here. Further, shootings deemed legally justified are missing from this dataset as per reporting practices of the Boston Police Department. Additionally, transport times are predictive and cannot reliably determine the actual route that a transporting vehicle took, the time from injury to arrival or the obstacles they actually faced on that journey, instead reporting average conditions for transport. Finally, this represents a single city in the US and may not be representative of other urban areas where geographic segregation and structural inequities may differ.
Conclusions:
In Boston, greater transport times to a level one trauma center were associated with higher mortality for victims of firearm violence. There were modest disparities among transport times between different racial and ethnic groups. These findings can be used to guide similar study in other cities and support local and national policy changes focused on developing trauma systems.
Supplementary Material
Conflict of Interest Statement:
The authors have no conflicts of interest to disclose nor funding related to this study. All authors will provide conflict of interest statements as JTACS disclosure statements (http://links.lww.com/TA/D427).
Footnotes
Media Statement:
This work finds that longer transport times after a shooting are related to death and that racial and ethnic minorities are more likely to have longer transport times to care.
References
- 1.Rees CA, Monuteaux MC, Steidley I, Mannix R, Lee LK, Barrett J, et al. Trends and Disparities in Firearm Fatalities in the United States, 1990–2021. JAMA Netw Open. 2022;5(11):e2244221. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Meizoso JP, Ray JJ, Karcutskie CA, Allen CJ, Zakrison TL, Pust GD, et al. Effect of time to operation on mortality for hypotensive patients with gunshot wounds to the torso: The golden 10 minutes. J Trauma and Acute Care Surg. 2016; 81(4): 685–91. [DOI] [PubMed] [Google Scholar]
- 3.Crandall M, Sharp D, Unger E, Straus D, Brasel K, Hsia R, et al. Trauma Deserts: Distance From a Trauma Center, Transport Times, and Mortality From Gunshot Wounds in Chicago. Am J Public Health. 2013; 103(6): 1103–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Scantling D, Orji W, Hatchimonji J, Kaufman E, Holena D. Firearm Violence, Access to Care, and Gentrification: A Moving Target for American Trauma Systems. Ann Surg. 2021;274(2):209–217. [DOI] [PubMed] [Google Scholar]
- 5.Byrne JP, Kaufman E, Scantling D, Tam V, Martin N, Raza S, et al. Association between Geospatial Access to Care and Firearm Injury Mortality in Philadelphia. JAMA Surg. 2022;157(10):942–949. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Kegler SR, Simon TR, Zwald ML, Chen MS, Mercy JA, Jones CM, et al. Vital Signs: Changes in Firearm Homicide and Suicide Rates — United States, 2019–2020. MMWR Morb Mortal Wkly Rep. 2022; 71(19): 656–663. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Centers for Disease Control and Prevention. Leading Causes of Death-Non-Hispanic black Males - United States. Available from: https://www.cdc.gov/minorityhealth/. Accessed July 23, 2023. https://www.cdc.gov/minorityhealth/lcod/men/2016/nonhispanic-black/index.htm
- 8.Berlin I The Making of African America: The Four Great Migrations. Vol 9. The Penguin Group; 2010. [Google Scholar]
- 9.Poulson MR, Neufeld MY, Laraja A, Allee L, Kenzik KM, Dechert T. The effect of historic redlining on firearm violence. J Natl Med Assoc. 2023; 115(4): 421–7. [DOI] [PubMed] [Google Scholar]
- 10.Poulson M, Neufeld MY, Dechert T, Allee L, Kenzik KM. Historic redlining, structural racism, and firearm violence: A structural equation modeling approach. The Lancet Regional Health - Americas. 2021; e100052. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Garen JW. The epidemiology of firearm violence in the twenty-first century United States. Annu Rev Public Health. 2015;36:5–19. [DOI] [PubMed] [Google Scholar]
- 12.Do DP, Frank R, Zheng C, Iceland J. Hispanic Segregation and Poor Health: It’s Not Just Black and White. Am J Epidemiol. 2017;186(8):990–9. [DOI] [PubMed] [Google Scholar]
- 13.Pathman DE, Konrad TR, Schwartz R. The proximity of predominantly African American and Hispanic rural communities to physicians and hospital services. Journal of Rural Health. 2002; 18(3): 416–27. [DOI] [PubMed] [Google Scholar]
- 14.United States Census Bureau Data. United States Census Bureau. Published 2020. Accessed January 31, 2020. Available from: https://data.census.gov/cedsci/
- 15.Spinelli D Fitting spatial autoregressive logit and probit models using Stata: The spatbinary command. The Stata Journal. 2022; 22(2): 293–318. [Google Scholar]
- 16.Nathens A, Brunet F, Maier R. Development of Trauma Systems and Effect on Outcomes After Injury. The Lancet. 2004;363:1794–1801. [DOI] [PubMed] [Google Scholar]
- 17.Pigneri DA, Beldowicz B, Jurkovich GJ. Trauma Systems: Origins, Evolution, and Current Challenges. Surgical Clinics of North America. 2017;97(5):947–59. [DOI] [PubMed] [Google Scholar]
- 18.Brown JB, Rosengart MR, Billiar TR, Peitzman AB, Sperry JL. Distance matters: Effect of geographic trauma system resource organization on fatal motor vehicle collisions. J Trauma and Acute Care Surg. 2017;83(1):111–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Band RA, Salhi RA, Holena DN, Powell E, Branas CC, Carr BG. Severity-adjusted mortality in trauma patients transported by police. Ann Emerg Med. 2014;63(5): 608–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Wandling MW, Nathens AB, Shapiro MB, Haut ER. Association of prehospital mode of transport with mortality in penetrating trauma a trauma system-level assessment of private vehicle transportation vs ground emergency medical services. JAMA Surg. 2018;153(2):107–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.McCoy CE, Menchine M, Sampson S, Anderson C, Kahn C. Emergency medical services out-of-hospital scene and transport times and their association with mortality in trauma patients presenting to an urban level i trauma center. Ann Emerg Med. 2013;61(2):167–74. [DOI] [PubMed] [Google Scholar]
- 22.Karrison TG, Philip Schumm L, Kocherginsky M, Thisted R, Dirschl DR, Rogers S. Effects of driving distance and transport time on mortality among Level i and II traumas occurring in a metropolitan area. J Trauma and Acute Care Surg. 2018;85(4):756–65. [DOI] [PubMed] [Google Scholar]
- 23.Swaroop M, Straus DC, Agubuzu O, Esposito TJ, Schermer CR, Crandall ML. Pre-hospital transport times and survival for Hypotensive patients with penetrating thoracic trauma. J Emerg Trauma Shock. 2013;6(1):16–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Esmaeili Ranjbar A, Mayel M, Movahedi M, Emaeili Ranjbar F, Mirafzal A. Pre-hospital time intervals in trauma patient transportation by emergency medical service: association with the first 24-hour mortality. Journal of Emergency Practice and Trauma. 2015;2(2):37–41. [Google Scholar]
- 25.Mackenzie EJ, Hoyt DB, Sacra JC, Jurkovich GJ, Carlini AR, Teitelbaum SD, et al. National Inventory of Hospital Trauma Centers. JAMA. 2003;289(12):1515–22. [DOI] [PubMed] [Google Scholar]
- 26.Lanfear CC, Bucci R, Kirk DS, Sampson RJ. Inequalities in Exposure to Firearm Violence by Race, Sex, and Birth Cohort From Childhood to Age 40 Years, 1995–2021. JAMA Netw Open. 2023;6(5):e2312465. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Jacoby SF, Dong B, Beard JH, Wiebe DJ, Morrison CN. The enduring impact of historical and structural racism on urban violence in Philadelphia. Soc Sci Med. 2018;199:87–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Circo GM. Distance to trauma centres among gunshot wound victims: Identifying trauma “deserts” and “oases” in Detroit. Injury Prevention. 2019;25:39–43. [DOI] [PubMed] [Google Scholar]
- 29.Tung EL, Hampton DA, Kolak M, Rogers SO, Yang JP, Peek ME. Race/Ethnicity and Geographic Access to Urban Trauma Care. JAMA Netw Open. 2019;2(3): e190138. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.