A novel machine-learning tool to identify community risk for firearm violence: The Firearm Violence Vulnerability Index.
Journal
The journal of trauma and acute care surgery
ISSN: 2163-0763
Titre abrégé: J Trauma Acute Care Surg
Pays: United States
ID NLM: 101570622
Informations de publication
Date de publication:
01 07 2023
01 07 2023
Historique:
medline:
23
6
2023
pubmed:
5
4
2023
entrez:
4
4
2023
Statut:
ppublish
Résumé
Firearm violence in the United States is a public health crisis, but accessing accurate firearm assault data to inform prevention strategies is a challenge. Vulnerability indices have been used in other fields to better characterize and identify at-risk populations during crises, but no tool currently exists to predict where rates of firearm violence are highest. We sought to develop and validate a novel machine-learning algorithm, the Firearm Violence Vulnerability Index (FVVI), to forecast community risk for shooting incidents, fill data gaps, and enhance prevention efforts. Open-access 2015 to 2022 fatal and nonfatal shooting incident data from Baltimore, Boston, Chicago, Cincinnati, Los Angeles, New York City, Philadelphia, and Rochester were merged on census tract with 30 population characteristics derived from the 2020 American Community Survey. The data set was split into training (80%) and validation (20%) sets; Chicago data were withheld for an unseen test set. XGBoost, a decision tree-based machine-learning algorithm, was used to construct the FVVI model, which predicts shooting incident rates within urban census tracts. A total of 64,909 shooting incidents in 3,962 census tracts were used to build the model; 14,898 shooting incidents in 766 census tracts were in the test set. Historical third grade math scores and having a parent jailed during childhood were population characteristics exhibiting the greatest impact on FVVI's decision making. The model had strong predictive power in the test set, with a goodness of fit ( D2 ) of 0.77. The Firearm Violence Vulnerability Index accurately predicts firearm violence in urban communities at a granular geographic level based solely on population characteristics. The Firearm Violence Vulnerability Index can fill gaps in currently available firearm violence data while helping to geographically target and identify social or environmental areas of focus for prevention programs. Dissemination of this standardized risk tool could also enhance firearm violence research and resource allocation. Prognostic and Epidemiological; Level IV.
Sections du résumé
BACKGROUND
Firearm violence in the United States is a public health crisis, but accessing accurate firearm assault data to inform prevention strategies is a challenge. Vulnerability indices have been used in other fields to better characterize and identify at-risk populations during crises, but no tool currently exists to predict where rates of firearm violence are highest. We sought to develop and validate a novel machine-learning algorithm, the Firearm Violence Vulnerability Index (FVVI), to forecast community risk for shooting incidents, fill data gaps, and enhance prevention efforts.
METHODS
Open-access 2015 to 2022 fatal and nonfatal shooting incident data from Baltimore, Boston, Chicago, Cincinnati, Los Angeles, New York City, Philadelphia, and Rochester were merged on census tract with 30 population characteristics derived from the 2020 American Community Survey. The data set was split into training (80%) and validation (20%) sets; Chicago data were withheld for an unseen test set. XGBoost, a decision tree-based machine-learning algorithm, was used to construct the FVVI model, which predicts shooting incident rates within urban census tracts.
RESULTS
A total of 64,909 shooting incidents in 3,962 census tracts were used to build the model; 14,898 shooting incidents in 766 census tracts were in the test set. Historical third grade math scores and having a parent jailed during childhood were population characteristics exhibiting the greatest impact on FVVI's decision making. The model had strong predictive power in the test set, with a goodness of fit ( D2 ) of 0.77.
CONCLUSION
The Firearm Violence Vulnerability Index accurately predicts firearm violence in urban communities at a granular geographic level based solely on population characteristics. The Firearm Violence Vulnerability Index can fill gaps in currently available firearm violence data while helping to geographically target and identify social or environmental areas of focus for prevention programs. Dissemination of this standardized risk tool could also enhance firearm violence research and resource allocation.
LEVEL OF EVIDENCE
Prognostic and Epidemiological; Level IV.
Identifiants
pubmed: 37012632
doi: 10.1097/TA.0000000000003992
pii: 01586154-202307000-00019
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
128-136Informations de copyright
Copyright © 2023 Wolters Kluwer Health, Inc. All rights reserved.
Références
Everytown Research & Policy. A more complete picture: the contours of gun injury in the United States. Revised November 2, 2021. Effective October 4, 2020. Available at: https://everytownresearch.org/report/nonfatals-in-the-us/ . Accessed November 20, 2022.
National Center for Injury Prevention and Control, Centers for Disease Control and Prevention. Web-based Inquiry Statistics Query and Reporting System (WISQARS). Centers for Disease Control and Prevention. Atlanta, GA. Available at: www.cdc.gov/ncipc/wisqars . Accessed November 20, 2022.
Ssentongo P, Fronterre C, Ssentongo AE, Advani S, Heilbrunn ES, Hazelton JP, et al. Gun violence incidence during the COVID-19 pandemic is higher than before the pandemic in the United States. Sci Rep . 2021;11(1):20654.
Rostron A. The Dickey Amendment on federal funding for research on gun violence: a legal dissection. Am J Public Health . 2018;108(7):865–867.
Stark DE, Shah NH. Funding and publication of research on gun violence and other leading causes of death. JAMA . 2017;317(1):84–85.
Kaufman EJ, Delgado MK. The epidemiology of firearm injuries in the US: the need for comprehensive, real-time, actionable data. JAMA . 2022;328(12):1177–1178.
NORC at The University of Chicago. A Blueprint for U.S. Firearms Data Infrastructure. Effective October 14, 2020. Available at: https://www.norc.org/PDFs/Firearm%20Data%20Infrastructure%20Expert%20Panel/A%20Blueprint%20for%20a%20U.S.%20Firearms%20Data%20Infrastructure_NORC%20Expert%20Panel%20Final%20Report_October%202020.pdf . Accessed November 20, 2022.
Kaufman EJ, Passman JE, Jacoby SF, Holena DN, Seamon MJ, MacMillan J, et al. Making the news: victim characteristics associated with media reporting on firearm injury. Prev Med . 2020;141:106275.
Barber C, Cook PJ, Parker ST. The emerging infrastructure of US firearms injury data. Prev Med . 2022;165:107129.
Campell S, Nass D, Nguyen M. The CDC Says Gun Injuries Are on the Rise. But There Are Big Problems With Its Data. The Trace. October 4, 2018. Available at: https://www.thetrace.org/2018/10/cdc-nonfatal-gun-injury-data-estimate-problems/ . Accessed on November 20, 2022.
University of Wisconsin School of Medicine and Public Health. Neighborhood Atlas. Maison, WI. Available at: https://www.neighborhoodatlas.medicine.wisc.edu/ . Accessed November 1, 2022.
diversitydatakids.org . Childhood Opportunity Index 2.0. Available at: https://data.diversitydatakids.org/dataset/coi20-child-opportunity-index-2-0-database . Accessed November 1, 2022.
Centers for Disease Control and Prevention/Agency for Toxic Substances and Disease Registry/Geospatial Research A, and Services Program. CDC/ASTDR Social Vulnerability Index. Available at: https://www.atsdr.cdc.gov/placeandhealth/svi/data_documentation_download.html . Accessed October 1, 2021.
Polcari AM, Hoefer LE, Callier K, Zakrison TL, Rogers SO, Henry M, Slidell MB, Benjamin AJ. Social Vulnerability Index is Strongly Associated with Urban Pediatric Firearm Violence: An Analysis of Five Major U.S. Cities. J Trauma Acute Care Surg . 2023. doi: 10.1097/TA.0000000000003896.
American Community Survey 5-year Data Profiles. United States Census Bureau. Washington, DC. Available at: https://data.census.gov/table?d=ACS+5-Year+Estimates+Data+Profiles . Accessed October 1, 2021.
Opportunity Atlas Data Tool. United States Census Bureau. Washington, DC. Available at: https://www.census.gov/programs-surveys/ces/data/public-use-data/opportunity-atlas-data-tables.html . Accessed June 1, 2022.
Chen T, Guestrin C. XGBoost: A Scalable Tree Boosting System. KDD '16: Proceedings of the 22nd SCM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, August 13–17, 2016, New York, NY: Association for Computing Machinery; 2016. 785–794.
Salati M, Migliorelli L, Moccia S, Andolfi M, Roncon A, Guiducci GM, et al. A machine learning approach for postoperative outcome prediction: surgical data science application in a thoracic surgery setting. World J Surg . 2021;45(5):1585–1594.
Trinidad S, Vancil A, Brokamp C, Moody S, Gardner D, Parsons AA, et al. Relationships between socioeconomic deprivation and pediatric firearm-related injury at the neighborhood level. J Trauma Acute Care Surg . 2022;93(3):283–290.
Stevens J, Leonard J, Reppucci ML, Schroeppel T, Bensard D, Haasz M. Individual and neighborhood level characteristics of pediatric firearm injuries presenting at trauma centers in Colorado. J Trauma Acute Care Surg . 2022;93(3):385–393.
Kim D. Social determinants of health in relation to firearm-related homicides in the United States: a nationwide multilevel cross-sectional study. PLoS Med . 2019;16(12):e1002978.
Houghton A, Jackson-Weaver O, Toraih E, Burley N, Byrne T, McGrew P, et al. Firearm homicide mortality is influenced by structural racism in US metropolitan areas. J Trauma Acute Care Surg . 2021;91(1):64–71.
Formica MK. An eye on disparities, health equity, and racism-the case of firearm injuries in urban youth in the United States and globally. Pediatr Clin North Am . 2021;68(2):389–399.
Kwon EG, Wang BK, Iverson KR, O'Connell KM, Nehra D, Rice-Townsend SE. Interpersonal violence affecting the pediatric population: patterns of injury and recidivism. J Pediatr Surg . 2023;58:136–141.
Gastineau KAB, Williams DJ, Hall M, Goyal MK, Wells J, Freundlich KL, et al. Pediatric firearm-related hospital encounters during the SARS-CoV-2 pandemic. Pediatrics . 2021;148(2):e2021050223.
Abaza R, Lukens-Bull K, Bayouth L, Smotherman C, Tepas J, Crandall M. Gunshot wound incidence as a persistent, tragic symptom of area deprivation. Surgery . 2020;168(4):671–675.
Rowhani-Rahbar A, Quistberg DA, Morgan ER, Hajat A, Rivara FP. Income inequality and firearm homicide in the US: a county-level cohort study. Inj Prev . 2019;25(Suppl 1):i25–i30.
Lipton R, Yang X, Braga AA, Goldstick J, Newton M, Rura M. The geography of violence, alcohol outlets, and drug arrests in Boston. Am J Public Health . 2013;103(4):657–664.
MacDonald J. Community design and crime: the impact of housing and the built environment. Crime Justice . 2015;44:333–383.
University of Chicago Crime Lab New York. The Impact of Street Lighting on Crime in New York City Public Housing. Effective October 2018. Available at: http://www.modresvetlo.cz/PDF/The_Impact_of_Street_Lighting_on_Crime_in_New_York_City_Public_Housing.pdf . Accessed September 20, 2022.