Aims: To determine the acute lethal toxicity of a range of psychoactive substances in terms of the dose customarily used as a single substance for non-medical purposes.
Design and method: A structured English-language literature search was conducted to identify experimental studies and clinical reports that documented human and non-human lethal doses of 20 abused substances that are distributed widely in Europe and North America. Four inclusion criteria were specified for the reports, and approximately 3000 relevant records were retrieved from search engines at Biosis, Science Citation Index, Google and the National Library of Medicine's Gateway. In order to account for different drug potencies, a 'safety ratio' was computed for each substance by comparing its reported acute lethal dose with the dose most commonly used for non-medical purposes.
Findings: The majority of published reports of acute lethal toxicity indicate that the decedent used a co-intoxicant (most often alcohol). The calculated safety ratios varied between substances by more than a factor of 100. Intravenous heroin appeared to have the greatest direct physiological toxicity; several hallucinogens appeared to have the least direct physiological toxicity.
Conclusions: Despite residual uncertainties, the substantial difference in safety ratios suggests that abused substances can be rank-ordered on the basis of their potential acute lethality.