Engaging a hard-to-reach population in research: sampling and recruitment of hired farm workers in the MICASA study.

Hired farm workers provide the majority of the workforce for California's labor-intensive agricultural sector Agriculture is one of the most hazardous occupations, but there has been little research into the etiology of poor health outcomes that occur disproportionately in hired farm worker populations. MICASA is a cohort investigation of occupational and environmental health risks in hired farm worker households in Mendota, California, that employed a two-stage sampling process, including random selection of census blocks and door-to-door enumeration. The aim of this analysis was to evaluate the success of the sampling process and compare demographics of the enumerated population to other regional samples of Latino populations. In the enumeration, 1257 addresses were mapped and 729 hired farm worker households were enumerated. Findings showed no significant differences between the enumerated population and the resulting MICASA study sample; however, the MICASA population was more likely to be male, from Central America, work in agriculture, and have fewer years residency in the U.S. than California Health Interview Survey (CHIS) respondents. Additionally, 9.4% of the enumerated dwellings were back houses or unofficial dwellings and may have been missed by the U.S. Census 2000. Demographic comparisons between the enumerated population, census data, and CHIS data highlight the differences in these sampling methods and suggest possible demographic changes in hired farm workers in California. While difficulties in accessing hired farm workers often account for the lack of population-based research, the MICASA cohort provides an opportunity to examine occupational health patterns relevant to other farm worker populations.