The study found no statistically significant relationship between the rate of heroin-related deaths and emergencies and the concentration of diacetylmorphine of street heroin samples confiscated in Vienna in 1999. The street heroin sold in Vienna during the study period was generally of low quality and in base form with a high proportion of added caffeine, which is more suitable for smoking than for injection. The study results do not rule out the possibility that some accidental heroin-related deaths and/or heroin-related emergencies in Vienna in 1999 occurred because addicts purchased heroin with an unexpectedly high content of diacetylmorphine and/or total amount of opiates; however, the findings do challenge the widely held belief that the number of heroin-related deaths is directly related to fluctuations in the purity of street heroin being sold to users. All street heroin samples confiscated by Vienna police between January 1, 1999, and December 31, 1999, were analyzed using gas chromatography. Purity was determined by the concentration of diacetylmorphine. The total amount of opiates was also determined by adding the concentrations of diacetylmorphine and certain impurities. Data on drug-related deaths in Vienna were obtained from autopsy records, since Austria requires an autopsy in all cases of sudden death. Data on drug-related emergencies were obtained from the Vienna Ambulance Service. An opiate-related emergency was assumed if the diagnosis made by the physician was based on clinical findings of impaired respiration and/or recovery after naloxone administration. 1 table, 2 figures, and 49 references