Comparing the Distressed Communities Index and Area Deprivation Index in Predicting Firearm-Related Emergency...

Area of Science:

• Public Health
• Socioeconomic Determinants of Health
• Injury Prevention

Background:

• Firearm-related injuries are a significant public health issue in the U.S., leading to numerous Emergency Department (ED) visits.
• Neighborhood socioeconomic status (SES) is a critical factor in understanding firearm injury patterns.
• The Area Deprivation Index (ADI) and the Distressed Communities Index (DCI) are tools used to measure neighborhood SES.

Purpose of the Study:

• To compare the predictive accuracy of the DCI and ADI in identifying ED visits for firearm-related injuries.
• To assess the performance of models using DCI, ADI, and a combination of both indices.

Main Methods:

• Retrospective analysis of Maryland ED data from 2019-2020 for firearm-related injuries.
• Construction of three logistic regression models: DCI-only, ADI-only (stratified), and combined DCI/ADI.
• Model performance evaluated using Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and Receiver Operating Characteristic (ROC) curves, controlling for demographic and clinical factors.

Main Results:

• Out of 2,392,813 injury-related ED visits, 2,504 (0.1%) were firearm-related.
• Firearm injuries disproportionately affected Black individuals (84.7%) and males (88.2%).
• The DCI-only model (AIC=12,703.88) and ADI-only model (AIC=12,748.6) showed comparable predictive accuracy for firearm-related ED visits; combining them offered no significant improvement.

Conclusions:

• Both the Distressed Communities Index (DCI) and the Area Deprivation Index (ADI) are effective and comparable predictors of firearm-related ED visits.
• The DCI offers a viable alternative to the traditional ADI for assessing neighborhood-level risk factors in firearm injury research and prevention.