Recommendations for Future Research on
Alcohol and Injuries in ER Studies

Post-conference meeting, Berkeley, CA, October 7, 2005

Jason Bond, Guilherme Borges, Cherpitel Cherpitel, Mariana Cremonte, Norman Giesbrecht, Tom Greenfield, Daniel Hungerford, Scott Macdonald, Jacek Moskalewicz, Vladimir Poznyak, Robin Room, Tim Stockwell, Grazyna Swiatkiewicz, Robin Touquet, Yu Ye

The following were discussed as main areas/topics that warrant more research in the ER setting:

I. Epidemiology of the magnitude of alcohol involvement in injury and involvement by type and
cause of injury

While we know that alcohol consumption is associated with injury, further studies are needed of the magnitude of alcohol involvement in injury occurrence and alcohol involvement by type and cause of injury. These can include secondary analyses of existing data, including aggregations of different datasets. Issues to be investigated in these studies include: the distribution on level of alcohol impairment of those with a particular injury; dose-response relationships; the existence and extent of co-factors which contribute to the occurrence of the particular injury (e.g., other drug use, icy conditions); distribution in terms of the context of occurrence of the particular type of injury; distribution of usual drinking patterns (volume and heavy episodic consumption) and injury; and, hangover effects associated with drinking in the event. Contextual variables affecting associations at the study level are also an important consideration. Particularly for intentional injuries, studies are needed which measure and analyze cultural variations in alcohol expectancies and in the excuse value of alcohol. The impact of alcohol involvement by severity of injury and extent of disability, and alcohol dependence and abuse as predictors of severity and disability are also important to examine.


II. Clinical assessment of alcohol intoxication

Research is needed on optimal ways of recording estimated BAC, degree of intoxication and other clinically significant information for use in replacing or reformulating the alcohol codes (Y90 and Y91) in ICD-10, looking towards their adoption in ICD-11.


III. Methodological comparisons of control subjects for estimating risk of injury in ER patients

Studies of general populations concerning injuries and alcohol involvement in the injury are needed. Analyses of such population surveys and of ER samples in the same catchment area can be used to illuminate the effect of alcohol on whether and when an injured person presents to an ER, and can illuminate factors that distinguish injury-prone from other heavier drinkers.

Studies are needed with designs which allow for estimation of the relative risk of injury from a given blood-alcohol level in particular contexts and with other particular co-factors. These can include case-crossover and other designs that yield a control condition.

IV. Methods of obtaining BAC estimates

Development and testing of methods of obtaining BAC estimates with the least possible intrusion is needed. The forensic significance of accurate measurements makes them much more difficult and unlikely to actually be collected on a routine basis. Given this, the emphasis should be on methods which yield estimates which are sufficient for monitoring and epidemiological research purposes, but not accurate enough to be useful forensically.


V. Brief intervention in the ER

Additional studies are needed to explore the implementation and efficacy of brief intervention in the emergency room setting, who delivers the intervention, type of ER setting, context of injury and recidivism outcomes, including the development of a brief intervention package that is “transportable” and easily distributed across cultures in a multitude of ER settings.


VI. Comparisons of individual-level risk of injury estimates from ER studies with aggregate-level data

Comparative analysis is needed of individual-level estimates of risk of injury obtained from ER studies with those estimates obtained from aggregate-level data. These analyses will contribute data on the magnitude of alcohol’s causal role in injury as well as inform and improve the estimates of the global burden of disease for injury attributable to alcohol consumption.


VII. Implementation of a national alcohol surveillance system

Developmental and evaluative studies that prepare the way for implementation of a national surveillance system for alcohol involvement in ER presentations are needed. Such a system could be similar to the CDC National Drug System and the National Epidemiologic Injury Surveillance System (NEISS).


VIII. Dissemination of major research findings of recent ER studies

Studies are needed of the best approaches for disseminating major research findings from recent ER studies to inform prevention activities and to encourage consideration of the ER as an epidemiologic resource and a resource to inform brief intervention and policy development.