Drugs and alcohol often occur together in fatal poisonings, complicating the process of determining the cause of death. Especially when found in concentrations generally regarded as toxic but not lethal, the question arises whether the combination of sublethal amounts was the likely cause of death. In this study, we examined poisoning deaths involving amitriptyline, propoxyphene and promazine, which are, after benzodiazepines, the most frequently occurring drugs in Finnish alcohol-related poisonings. From the forensic toxicology database, covering the years 1995-2002, we extracted 332 fatal poisonings, calculated median blood alcohol and drug concentrations, constructed concentration-concentration and concentration-response curves and evaluated the significance of the presence of therapeutic amounts of benzodiazepines. Median amitriptyline and propoxyphene concentrations were lower in alcohol-related cases than in clean drug poisonings. Correspondingly, the median blood alcohol concentrations in all drug-related poisonings were 1.5-2.2 mg/g lower than that found in clean alcohol poisonings. Alcohol concentration proved to be a more sensitive indicator of alcohol-drug interaction than drug concentration. This result suggests that when alcohol is present, relatively small overdoses of the studied drugs may result in fatal poisoning. In this context, fatal drug and alcohol concentrations and the issue of determining the most important agent in fatal drug-alcohol intoxications are discussed.