A meta-analysis of alcohol toxicology study findings among homicide victims

Aim To synthesize the results of alcohol toxicology reports for homicide victims and examine variations in these results across person and setting characteristics. Methods We meta-analyzed 61 independent studies from 57 published manuscripts which met the study inclusion criteria and reported alcohol toxicology test results for homicide victims. A total of 71 031 toxicology test results, derived from 78 265 homicide victims across 13 countries (most from the United States), were examined. Results On average, 48% of homicide victims tested positive for alcohol and 33% (using the 0.08 threshold) or 35% (using the 0.10 threshold) were determined to be intoxicated. The proportion of homicide victims testing positive for alcohol appeared to be decreasing over time. Further, the proportion testing positive increased with age is higher for female than for male victims, and differs by race. Finally, the overall estimates were relatively stable across study sites. Conclusion Alcohol toxicology test results remain an important method for measuring the success of efforts to manage the consequences of alcohol. However, future toxicology studies should focus upon collecting information on evidence processing time, establishing measurement standards for reporting data and ensuring that subgroup estimates are included for purposes of cross-site comparisons.