Organophosphorus pesticide self-poisoning is an important clinical problem in rural regions of the developing world, and kills an estimated 200,000 people every year. Unintentional poisoning kills far fewer people but is a problem in places where highly toxic organophosphorus pesticides are available. Medical management is difficult, with case fatality generally more than 15%. We describe the limited evidence that can guide therapy and the factors that should be considered when designing further clinical studies. 50 years after first use, we still do not know how the core treatments--atropine, oximes, and diazepam--should best be given. Important constraints in the collection of useful data have included the late recognition of great variability in activity and action of the individual pesticides, and the care needed cholinesterase assays for results to be comparable between studies. However, consensus suggests that early resuscitation with atropine, oxygen, respiratory support, and fluids is needed to improve oxygen delivery to tissues. The role of oximes is not completely clear; they might benefit only patients poisoned by specific pesticides or patients with moderate poisoning. Small studies suggest benefit from new treatments such as magnesium sulphate, but much larger trials are needed. Gastric lavage could have a role but should only be undertaken once the patient is stable. Randomised controlled trials are underway in rural Asia to assess the effectiveness of these therapies. However, some organophosphorus pesticides might prove very difficult to treat with current therapies, such that bans on particular pesticides could be the only method to substantially reduce the case fatality after poisoning. Improved medical management of organophosphorus poisoning should result in a reduction in worldwide deaths from suicide.