This pilot program was designed to: 1) determine whether farmers could use a simple self-reporting system to estimate the incidence of occupationally related, self-limited pesticide poisonings; 2) describe the frequencies and types of pesticide products used in spraying operations; and 3) assess whether self-reporting and feedback would influence spraying behaviors. For 12 months, 50 farmers in northern Vietnam recorded after every spraying session any adverse health effect and the pesticide used. Data were also gathered from 50 controls. Of the 1,798 recorded spray operations, 8% were asymptomatic, 61% associated with vague ill-defined effects, and 31% accompanied by a least one clear or symptom of poisoning. After six months, the self-reporting farmers' spraying frequency and use of highly hazardous products (Ia/Ib) had declined more significantly than in the controls, as had their moderate adverse health effects. This low-cost surveillance method influenced the behaviors of farmers given access to IPM farmer field schools. The study also demonstrated the value of farmers as informants.