National Farmworker Database: Establishing a farmworker cohort for epidemiologic research

Abstract

Background There is little research into the long-term health effects of pesticides and other agricultural exposures among seasonal and migrant farmworkers in the United States. We present results of a feasibility study that established a cohort of farmworkers for use in epidemiologic research. Methods Subjects consisted of migrant and seasonal farmworkers who joined the cohort while seeking social services through members of the Association of Farmworker Opportunity Programs (AFOP) and were entered in the National Farmworker Database (NFD) between the end of 1997 and March 1999. We designed an add-on interview with information that enhanced the utility of the database for epidemiologic research. Results We recruited and obtained basic demographic and employment information on 5,597 farmworkers at very modest cost and effort. Subjects were mostly seasonal (61.5%), female (56.7%), and Hispanic (67.4%), with a median age of 27. Most (62.6%) had not completed high school; almost all (99.1%) reported being U.S. citizens or permanent residents, an eligibility requirement for some of the services provided by AFOP. The majority (62.5%) had engaged in farmwork for less than 10 years, but had performed a wide variety of tasks on different crops, including row crops and tree fruits. Picking was the most common task reported. Most subjects had performed farmwork in Florida, North Carolina, Texas, Michigan, or Georgia. For usual source of health care, 63.7% reported use of U.S. hospitals or emergency rooms/clinics, 42.0% U.S. private physicians, and 29.7% migrant health clinics. Among subjects reporting a prior diagnosis of cancer, primary sources of health care for treatment of that cancer included U.S. private physicians (61.9%), U.S. hospitals or emergency rooms/clinics (23.8%), and migrant health clinics (10.5%). Conclusions Results suggest that by adding a brief interview to the existing NFD data collection process, which was designed for other purposes, it is feasible to create an efficient tool for conducting longitudinal epidemiologic research among farmworkers. Am. J. Ind. Med. 40:612–618, 2001. © 2001 Wiley-Liss, Inc.