The 131st Annual Meeting (November 15-19, 2003) of APHA

Abstract #57492

Scott Macdonald, PhD¹, Cheryl J. Cherpitel, DrPH², Norman A. Giesbrecht, PhD³, Guilherme Borges, ScD⁴, Tim Stockwell, PhD⁵, and A. DeSouza¹. (1) Centre for Addiction and Mental Health, 100 Collip Circle, Suite 200, London, ON N6G 4X8, Canada, 519 679-1892, scotm@uwo.ca, (2) Alcohol Research Group, 2000 Hearst Avenue, Berkeley, CA 94709, (3) Addiction Research Foundation Site, Centre for Addiction & Mental Health, 33 Russell Street, Toronto, ON M5S 2S1, Canada, (4) National Institute of Psychiatry, Calzada Mexico Xochimilco No. 101, Col. San Lorenzo Huipulco C.P. 10610, Mexcio DF, Mexico, (5) National Drug Research Institute, Curtin University of Technology, Level 4, 679 Murray Street, West Perth Western Australia, Perth, 6005, Australia

This paper is based on data collected from injury patients at 30 Emergency Room (ER) Departments using similar methodology in six countries. These data are analyzed with the goal of determining whether alcohol is a likely cause of violence through an analysis of traditional means of investigating causation and an application of some main criteria for establishing causation outlined by Sir Bradford Hill: (1) temporal sequence, (2) strength of association, (3) an observed statistical relationship cannot be better explained by a third variable, (4) specificity (5) consistency of findings, (6) biological gradient (i.e. dose-response relationship), (7) coherence with biological background and previous knowledge. The data collection methods and results supported temporal sequence of events and specificity. Odds ratios of violence versus accidental injury for a blood alcohol concentration (BAC) over 80mg% were significant for all countries, ranging from 2.77 for Mexico to 9.45 for Canada, supporting both strength of associations and consistency of findings. No third variables were found that better explain these relationships. A dose-response relationship between BAC and likelihood of violence was found at lower levels of BAC, but the observed deterioration of this relationship at very high BAC levels was likely due to the competing influence of alcohol on accidents and medical conditions. Finally, the results appear consistent with existing knowledge of alcohol’s effect on aggression. Overall, data from this meta-analysis points to a causative role of alcohol in violence.

Learning Objectives:

• At the conclusion of this session the learner will have a better understanding of alcohol's association with violence-related injuries across countries.

Keywords: Alcohol Use, Emergency Department/Room

Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.