Alcohol consumption, setting, gender and activity as predictors of injury: a population-based case-control study.

OBJECTIVE A population-based case-control design was employed to quantify the risk of injury after consumption of alcohol as a function of setting, concurrent activity and usual drinking habits. METHOD A total of 797 cases (66.6% men) and 797 controls (57.7% women) were interviewed. The response rate was 83% for eligible cases approached for an interview. Cases were injured patients from a hospital emergency department. Community controls used were each paired with a case on suburb of residence, and interviewed regarding their activities in the 6-hour period preceding their paired case's injury. RESULTS Self-reported alcohol consumption was consistent with both medical records and breath-analyzer tests. Drinking any alcohol and using prescribed medication in the prior 6 hours were both associated with significantly increased risk of injury when controlling for demographic and setting variables. Use of illicit drugs (mainly cannabis) was associated with reduced risk of injury. Setting (e.g., recreational, work) and activity (e.g., sport, travel, work) variables were also independently associated with risk of injury. The risk of injury for women was significantly elevated for any consumption of alcohol; for men it was elevated only when consumption exceeded 90 grams. CONCLUSIONS These data confirm earlier findings that risk of injury for women for a given level of consumption is greater than for men. They extend earlier findings by identifying significant setting, activity and drug use variables predictive of injury. In addition, when these latter variables are controlled, it is found that for women, but not for men, the risk of injury is significantly elevated even at low levels of alcohol intake.