Around half a million people die from pesticide poisoning each year in Asia and the Western Pacific. Most deaths are due to organophosphorus (OP) pesticides; other frequently lethal pesticides are paraquat and phosphide rodenticides. Both primary prevention and improved medical management could rapidly reduce deaths. However, there is little activity on either front. Pesticide regulation is sporadic and usually based on animal data. Current protocols for management of pesticide poisoning are based on little evidence and are difficult to deliver in the resource-poor settings in which most poisonings occur. The few systematic reviews conducted suggest there is no good quality human evidence that any antidote other than atropine is of benefit. Building on a collaboration currently funded by the Wellcome Trust, we propose to establish an Australian/Sir Lankan research collaboration to evaluate methods to reduce deaths from deliberate self-poisoning with pesticides. Tackling the problem on four complementary fronts (research into pathophysiology, antidotes, preventive strategies and improved delivery of evidence based clinical care) we believe we can achieve a 50% reduction in deaths. We will systematically review the evidence for the effectiveness of treatments for pesticide poisoning using Cochrane collaboration methods and test strategies to incorporate available evidence into clinical practice. We will collect data on the relative human toxicity of different pesticides, to guide a strategy to prevent poisoning through pesticide regulation. The most toxic pesticides will be restricted in selected districts. If deaths are reduced, then the restriction will be applied across the country. The neurotoxic effects of OP (Intermediate Syndrome and delayed neuropathy) cause many deaths through respiratory failure. We will conduct serial clinical and neurophysiological examinations to determine if these can be predicted, treated or prevented. A simple algorithm to predict prognosis in paraquat poisoning is needed to identify patients suitable for inclusion in antidote studies - we will determine if changes in creatinine can be used. We will do five phase II studies on new antidotes for pesticides: sodium bicarbonate for chlorphenoxy herbicides; antioxidant therapy for paraquat; sodium bicarbonate, clonidine and diazepam for OP. The effectiveness of the most promising antidote will be tested in a large Phase III study. By including prospective health-economic assessments on new and standard treatments, the costs and benefits of this will be apparent. All strategies are chosen such that they may be adopted by similar programs in other countries of the region.