Article Text

This article has a correction. Please see:

Download PDFPDF
105 Rates of violence and substance use among a high-risk sample during the flint water crisis (fwc): results from the flint youth injury study
  1. Patrick Carter1,
  2. Jason Goldstick1,
  3. Justin Heinze2,
  4. Maureen Walton3,
  5. Marc Zimmerman2,
  6. Jessica Roche1,
  7. Kaneesha Wallace1
  1. 1US University of Michigan Injury Centre
  2. 2US University of Michigan School of Public Health
  3. 3US University of MIchigan School of Medicine


Statement-of-purpose On April 25, 2014, Flint’s water source was changed without adding essential anti-corrosive agents, leading to lead contamination of the potable water supply. Individual-level consequences resulting from this community-wide traumatic event have yet to be examined. We examined rates of violence and substance use following the initial date of the Flint Water Crisis (FWC) exposure among youth already enrolled in an on-going longitudinal study.

Methods Drug-using youth (age: 14–24) seeking assault-injury treatment and a comparison cohort of youth seeking ED care for other reasons were enrolled in a longitudinal study (Baseline: 09/2009–12/2011; Follow-ups: 6, 12, 18, 24 months). Eligible youth were re-contacted and enrolled in a 2nd wave study (Baseline: 03/2014–01/2016) with two additional biannual follow-ups. For this analysis, we conducted pre/post analyses of paired data for participants completing follow-ups within the 24 months preceding/following the initial date of exposure and examined differences in rates of substance use and violence.

Results 352-youth enrolled in both studies; 313 (88.9%) had follow-up data within 24 months pre- and post-initial exposure date. In the analytic sample, baseline age was 20.1 (SD:2.4), 49.8% were male, 75.4% required public assistance, and 59.4% were assault-injured. Despite overall decreasing trends in violence and substance use among this cohort at-large, we found that the prevalence of violence (37.7%-vs.-49.5%,p<0.001) increased in the post-period, including rates of victimisation (31.9%-vs.-45.4%,p<0.001) and aggression (30.0%-vs-39.9%,p<0.001), and rates of non-partner (24.6%-vs.-32.3%,p<0.05) and partner (28.1%-vs.-40.9%,p<0.001) violence. Pre-post differences in alcohol (AUDIT-C:M=1.3,SD=2.9,p<0.001) and marijuana (ASSIST: M=1.2,SD=8.8,p<0.05) were also both positive.

Conclusion/Significance While causal attributions are limited, we observed increases in violence and substance use among this high-risk sample after the initial FWC exposure, suggesting that negative health effects associated with this on-going crisis extend beyond the environmental exposure to lead and highlights the need for additional resources and research focused on addressing the consequences of this wide-scale community trauma.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles

  • Correction
    BMJ Publishing Group Ltd