This paper was presented at the Research Society for Alcoholism Annual Meeting, Vancouver, British Columbia, Canada, 26 June−1 July 2004.
Multi-level analysis of alcohol-related injury among emergency department patients: a cross-national study*
A research report from the Emergency Room Collaborative Alcohol Analysis Project (ERCAAP) and the WHO Collaborative Study on Alcohol and Injuries
Article first published online: 23 NOV 2005
Volume 100, Issue 12, pages 1840–1850, December 2005
How to Cite
Cherpitel, C. J., Ye, Y., Bond, J., Rehm, J., Poznyak, V., Macdonald, S., Stafström, M. and Hao, W. (2005), Multi-level analysis of alcohol-related injury among emergency department patients: a cross-national study. Addiction, 100: 1840–1850. doi: 10.1111/j.1360-0443.2005.01257.x
- Issue published online: 23 NOV 2005
- Article first published online: 23 NOV 2005
- Submitted 14 September 2004; initial review completed 5 December 2004; final version accepted 25 April 2005
- international emergency department;
- multi-level modeling
Aim The aim of this analysis was to examine the average rate and variation of alcohol-related injury across emergency department (ED) studies, the effect of usual drinking on likelihood of alcohol-related injury, whether cross-study variation in rate of alcohol-related injury can be explained by between-study difference in usual consumption and whether social-cultural contextual variables help explain cross-study variations, after between-study difference in usual consumption has been controlled.
Design Data were merged from the Emergency Room Collaborative Alcohol Analysis Project (ERCAAP) and the WHO Collaborative Study on Alcohol and Injuries, together representing 28 studies in 16 countries, and include 8423 (drinking) injury patients who arrived in the ED within 6 hours after injury. Alcohol-related injury was based, separately, on a positive blood alcohol concentration (BAC) and self-reported drinking within 6 hours prior to injury. A multi-level design and hierarchical generalized linear models were used for analysis in which patients were nested within studies.
Findings Overall prevalence of alcohol-related injury was 24% and 29% for positive BAC and self-report, respectively. At the patient level, log-transformed alcohol consumption in the last 12 months was a significant predictor of alcohol-related injury. At the study level significant variation in rates of alcohol-related injury was observed; studies with higher overall average consumption reported a higher rate of alcohol-related injury. When volume was controlled, societies with higher detrimental drinking pattern and higher legal level for intoxication while driving were more likely to have an increased rate of alcohol-related injury.
Conclusions Alcohol-related injury varies across EDs and countries. While it is associated with an individual's usual alcohol consumption, it is also affected by a number of societal drinking characteristics including the aggregate volume of consumption, overall drinking pattern and legislative policies to control drinking and related harms.