Health effects from pesticide use in Costa Rica : an epidemiologic approach

Sammanfattning: The use of toxic pesticides is increasing in developing countries, often under unsafe conditions. Poisonings are a well known but underreported public health problem, and studies on long-term effects are sparse. Therefore, a survey was carried out to determine the incidence and the risk factors for pesticide poisonings in Costa Rica, integrating data from national registries of occupational accidents, hospitalizations, and fatalities between 1980-1986. A case series further assessed the exposures that can lead to a fatal outcome in nonintentional paraquat poisonings. A cross-sectional study among 211 banana plantation workers tested the hypothesis that chronic neurobehavioral deficits occur as a consequence of previous intoxication with cholinesterase inhibiting pesticides. An ecologic study tested the hypothesis that geographical differences in cancer incidence are associated with pesticide use. A cohort of 34,457 banana plantation workers with more than 400,000 person years was established to assess the cancer pattern among workers with a high level of exposure to pesticides (in particular dibromochloropropane), and to examine the feasibility of computerized record linkage between various registries in Costa Rica. In addition, health hazards from pesticide use in the Third World were reviewed, and some possibilities and limitations for epidemiologic research in developing countries were assessed. Cholinesterase inhibitors and paraquat caused the vast majority of the poisonings. Young workers, female workers and banana plantation workers were the groups at highest risk for occupational poisonings. Suicide as a cause of severe poisonings was less frequent than expected. It was estimated that, yearly, 1.5% of the agricultural workers suffer a work-related pesticide poisoning requiring medical attention. The case series suggested that paraquat may be fatal by dermal absorption of diluted paraquat, by ingestion of very low doses, and possibly by inhalation of diluted spray. Mild neurobehavioral deficits were found among previously poisoned workers, both by organophosphates and carbamates; after a poisoning, the workers seemed to become more susceptible to subsequent low dose exposures to cholinesterase inhibitors. Pesticide use was associated with increased cancer incidence in the most rural areas of Costa Rica, including lung and female hormone related cancers. Male banana plantation workers were at increased risk of melanoma and penile cancer, and female workers of cervical cancer and leukemia. Pesticides appear to be a public health problem throughout the Third World. Epidemiologic research to assess these problems is difficult because of methodological and infrastructure restrictions. However, high exposures provide good opportunities to carry out etiologic research; another advantage is that participation rates are often very high. Registry based studies may be feasible, but sometimes after costly improvements of the registries. This thesis indicates that pesticide exposure in Costa Rica is associated with an increased risk of various health effects.