Cannabis Was the Most Commonly Detected Substance in Assault-Injured Black Men in Baltimore | Research

The takeaway

Among young Black men treated for assault injuries at a Baltimore trauma center, cannabis was detected in 42.5% of cases, more than any other substance, and was more associated with firearm injuries than blunt force or stabbing.

Public health researchers studying violence and substance use, trauma professionals, and policymakers addressing urban violence.

42.5% of assault-injured men tested positive for cannabis

What the researchers found

Cannabis was the most frequently detected substance (42.5%), followed by alcohol (24.4%), opioids (22.3%), and cocaine (3.6%). Men who screened positive for cannabis had 32% lower odds of stabbing versus shooting and 50% lower odds of blunt force assault versus shooting. In contrast, alcohol was associated with stabbing and blunt force rather than firearm assault.

Why it matters

This study provides real-world data on substance use patterns among a population disproportionately affected by violence. The differential associations between specific substances and injury types suggest that contextual and environmental factors linked to different substances may influence the type of violence encountered.

The numbers in context

N = 1,922. Cannabis detected: 42.5%. Alcohol: 24.4%. Opioids: 22.3%. Cocaine: 3.6%. Multiple substances: 25.8%. Cannabis users: 32% lower odds of stabbing vs shooting, 50% lower odds of blunt force vs shooting.

How the study worked

Retrospective analysis of 1,922 Black men aged 18-34 admitted to a Baltimore trauma center with assault-related injuries and toxicology screens between 2013 and 2017. Multinomial logistic regression examined associations between substance detection and injury type (blunt force, stabbing, firearm).

What this study cannot tell us

Toxicology screens detect recent use but cannot determine whether the person was intoxicated at the time of assault. The study cannot establish whether substance use caused or contributed to the assault. Trauma center data may over-represent more severe injuries. Baltimore-specific findings may not generalize to other cities.

How to read the evidence

Large trauma center cohort with objective toxicology data, but cross-sectional design cannot establish causation between substance use and injury type.

When this study was published

Published in 2025 with data from 2013-2017.

The bigger picture

The finding that cannabis and alcohol are associated with different types of assault injuries reflects the complex interplay between substance use, social environments, and exposure to violence. The high rate of cannabis detection does not necessarily imply that cannabis contributed to the violence itself.

Questions still open

Why is cannabis more associated with firearm injuries while alcohol is more associated with other assault types? Do these patterns reflect different social environments where each substance is used?

Common questions

Does cannabis cause violence?

This study cannot answer that question. It found cannabis was commonly detected in assault victims but cannot determine whether cannabis contributed to the assault or was incidental. The substance-injury type associations likely reflect social and environmental contexts rather than direct pharmacological effects.

Why was cannabis detection so high?

Cannabis remains detectable in urine for days to weeks after use, so the 42.5% detection rate reflects recent use rather than intoxication at the time of assault. Cannabis use prevalence is also high in the demographic studied.

Read the original research

Association of Substance Use with Types of Assault-Related Injury Among Black Men in Baltimore, Maryland.

Substance use & misuse, 60(8), 1117-1125

View on PubMed ↗Publisher’s paper ↗

Citation

Marineau, L A; Perrin, N A; Johnson, R M; Uzzi, M; Alexander, K A; Irvin, N A; Thurman, P; Campbell, J C. (2025). Association of Substance Use with Types of Assault-Related Injury Among Black Men in Baltimore, Maryland.. Substance use & misuse, 60(8), 1117-1125. https://doi.org/10.1080/10826084.2025.2487974