BACKGROUND * Acute alcohol intoxication often complicates acute organophosphorus pesticide poisoning. * No data are available on how alcohol intoxication affects outcome in acute organophosphorus pesticide poisoning. * In particular, the relationships between plasma alcohol concentration and plasma organophosphorus concentration or outcome are unclear. CONCLUSIONS * Alcohol co-ingestion is associated with higher concentrations of the organophosphorus insecticide dimethoate, probably due to larger ingestions. * The higher concentrations of dimethoate found with alcohol co-ingestion increase the risk of death in dimethoate poisoning. There was no detectable effect of the alcohol itself on outcome. * Efforts to reduce deaths from insecticide self-poisoning may benefit from concurrent efforts to reduce alcohol consumption. OBJECTIVE Many patients acutely poisoned with organophosphorus insecticides have co-ingested alcohol. Although clinical experience suggests that this makes management more difficult, the relationship between plasma concentration of alcohol and insecticide is unknown. We aimed to determine whether acute intoxication results in ingestion of larger quantities of insecticide in dimethoate self-poisoning and a worse clinical outcome. METHODS We set up a prospective study of acute dimethoate self-poisoning in Sri Lankan district hospitals. An admission plasma sample was analysed to identify the ingested insecticide; in patients with detectable dimethoate, plasma alcohol was measured. RESULTS Plasma from 37 of 72 (51.4%) dimethoate-poisoned patients had detectable alcohol [median concentration 1.10 g l(-1)[110 mg dl(-1)][interquartile range (IQR) 0.78-1.65]] a median of 3 h post ingestion. The median plasma dimethoate concentration was higher in patients who had ingested alcohol [479 micromol l(-1) (IQR 268-701) vs. 145 micromol l(-1) (IQR 25-337); P < 0.001]. Plasma dimethoate concentration was positively correlated with plasma alcohol (Spearman's rho= 0.34; P= 0.0032). The median alcohol concentration was higher in the 21 patients who died compared with survivors (0.94 vs. 0.0 g l(-1), P= 0.018). Risk of death was greater amongst individuals who consumed alcohol [odds ratio (OR) 4.3, 95% confidence interval (CI) 1.2, 16.4]; this risk was abolished by controlling for dimethoate concentration (OR 0.3, 95% CI 0.0, 8.8), indicating that deaths were not due to the direct toxic effects of alcohol. CONCLUSIONS Alcohol co-ingestion is associated with higher plasma concentrations of dimethoate and increased risk of death. Larger studies are required to assess this finding's generalizability, since efforts to reduce deaths from self-poisoning may benefit from concurrent efforts to reduce alcohol consumption.