Related Experiment Video
Updated: Jun 28, 2025

Inverse Probability of Treatment Weighting Propensity Score using the Military Health System Data Repository and National Death Index
Published on: January 8, 2020
Historical Redlining and Present-Day Nonsuicide Firearm Fatalities
Ayesha Dholakia1, Kendall J Burdick1, Catherine Kreatsoulas2
1Department of Pediatrics, Boston Children's Hospital, and Department of Pediatrics, Boston Medical Center, Boston, Massachusetts (A.D., K.J.B.).
Background:
Redlining began in the 1930s with the Home Owners' Loan Corporation (HOLC); this discriminatory practice limited mortgage availability and reinforced concentrated poverty that still exists today. It is important to understand the potential health implications of this federally sanctioned segregation.
Objective:
To examine the relationship between historical redlining policies and present-day nonsuicide firearm fatalities.
Design:
Maps from the HOLC were overlaid with incidence of nonsuicide firearm fatalities from 2014 to 2022. A multilevel negative binomial regression model tested the association between modern-day firearm fatalities and HOLC historical grading (A ["best"] to D ["hazardous"]), controlling for year, HOLC area-level demographics, and state-level factors as fixed effects and a random intercept for city. Incidence rates (IRs) per 100 000 persons, incidence rate ratios (IRRs), and adjusted IRRs (aIRRs) for each HOLC grade were estimated using A-rated areas as the reference.
Setting:
202 cities with areas graded by the HOLC in the 1930s.
Participants:
Population of the 8597 areas assessed by the HOLC.
Measurements:
Nonsuicide firearm fatalities.
Results:
From 2014 to 2022, a total of 41 428 nonsuicide firearm fatalities occurred in HOLC-graded areas. The firearm fatality rate increased as the HOLC grade progressed from A to D. In A-graded areas, the IR was 3.78 (95% CI, 3.52 to 4.05) per 100 000 persons per year. In B-graded areas, the IR, IRR, and aIRR relative to A areas were 7.43 (CI, 7.24 to 7.62) per 100 000 persons per year, 2.12 (CI, 1.94 to 2.32), and 1.42 (CI, 1.30 to 1.54), respectively. In C-graded areas, these values were 11.24 (CI, 11.08 to 11.40) per 100 000 persons per year, 3.78 (CI, 3.47 to 4.12), and 1.90 (CI, 1.75 to 2.07), respectively. In D-graded areas, these values were 16.26 (CI, 16.01 to 16.52) per 100 000 persons per year, 5.51 (CI, 5.05 to 6.02), and 2.07 (CI, 1.90 to 2.25), respectively.
Limitation:
The Gun Violence Archive relies on media coverage and police reports.
Conclusion:
Discriminatory redlining policies from 80 years ago are associated with nonsuicide firearm fatalities today.
Primary Funding Source:
Fred Lovejoy Housestaff Research and Education Fund.
More Related Videos
Related Concept Videos
Bystander Effect
Censoring Survival Data
Statistical Methods for Analyzing Epidemiological Data
Longitudinal Research
Psychological and Sociocultural Causes of Schizophrenia
Framing Effects

