Authors
Crisostomo, L., Molina, VV.

Title
Pregnancy outcomes among farming households of Nueva Ecija with conventional pesticide use versus integrated pest management.

Journal
International Journal of Occupational and Environmental Health, 8:232-242. 2002.

Summary
Throughout the world, the increased use of pesticides has enhanced the health and economies of countries by amplifying crop production, decreasing food spoilage, and reducing pests. Yet, about one third of pesticides marketed in developing countries do not meet internationally accepted quality standards. Many pesticides that are banned or strictly controlled in the developing world are freely available in developing countries. A number of experimental and epidemiologic studies have suggested that some pesticides may be associated with specific pregnancy outcomes. Using a retrospective cohort design, the authors of this study evaluated the risk of preterm births, spontaneous abortions and birth defects among two farming municipalities, San Antonio and Munoz, of Nueva Ecija in the Philippines.

San Antonio and Muñoz are located 60 km apart from one another. Close to 93% of the farmers in San Antonio use pesticides conventionally while 75-80% of Muñoz farmers practice Integrated Pest Management (IPM). Conventional pesticide users (CPUs) were defined as those who apply pesticides at levels beyond the "spot spraying" method. In contrast, those who practiced IPM did zero spraying of pesticides or spot spraying as a last resort on heavily damaged areas. To be included in the study farming had to be the occupation of at least one spouse in the household during pregnancies that occurred between January 1998 and December 1999. Farming was defined as work that involved more than backyard farming. In addition, families had to have been residents for two years or more in the study area and both the wife and husband had to be physically present in the study area (village) within the period of 3 months before conception up to the first three months of pregnancy. Families from San Antonio who practiced IPM, and families from Muñoz who practiced CPU were excluded. Families were also excluded if farmers did not mix or apply pesticides, if they refused consent, or if the family had any of the following pregnancy outcomes: induced abortion, ectopic pregnancy or hydadiform mole.

Each village is subdivided into puroks (household clusters). House-to-house surveys were conducted by village health workers in 1/2 to 2/3 of the total number of puroks chosen randomly. In total 9/16 villages in San Antonio and 20/37 in Muñoz were surveyed. A pre-tested questionnaire was used to collect sociodemographic data, pesticide exposure and pregnancy outcomes. Birth outcomes were verified through actual physical examination, in the case of birth defects, and health records. Over 600 households were surveyed with 345 CPU households and 333 IPM households included in the study.

The authors found that 1374 brands of pesticides were used in 345 CPU households while only 399 brands were used in 331 IPM households. Mixture spraying in which two or more brands of pesticides are mixed in a single tank was practiced by 37% of CPU farmers and less than 1% of IPM farmers. Over 50% of the pesticides used in both study groups were classified as moderately hazardous by the WHO Classification system while, 21% of the pesticides used in the CPU and 7% of the IPM group were classified as extremely hazardous.

Using the Fisher's Exact Test, the authors examined several sociodemographic, health and reproductive factors that may have been associated with the exposure (pesticide use) and the outcomes (birth defects, spontaneous abortion, preterm birth). After controlling for past birth defect and preterm birth, birth defects were found to be more than 4 times as likely in the CPU group verses the IPM group (RR=4.56, 95% CI=1.21-17.09). In addition, birth defects were more likely if father's mixed or applied pesticides during the first trimester of their wife's pregnancy (p=0.05). Preterm birth was not found to be significantly associated with conventional pesticide use however, the CPU group was over 6 times as likely to have a spontaneous abortion (RR=6.17, 95% CI=1.37-27.86).

Importantly, this study looked at two very comparable population groups that differed mainly in their pesticide use practices. Also, the authors reviewed medical records to confirm pregnancy outcomes and took into account critical windows of exposure. The authors were unable to analyze the effect of specific pesticides and pesticide exposures were not directly measured, making the exposure measurements less precise. As well, pregnancy outcomes such as spontaneous abortion and birth defects are often inaccurately represented. It has been estimated that about half of all pregnancies end in spontaneous abortion, and thus any malformations that are part of syndromes incompatible with fetal life are missed. Consequently, the real incidence of birth anomalies and spontaneous abortions is difficult to ascertain. The results demonstrate that, in the population studied, exposure to pesticides in mixtures can be hazardous to fetal health, however, it will be difficult to determine what mixture, specific pesticide or concentration of pesticides is responsible for the adverse outcomes documented.