Agricultural worker populations in many countries show distinctive exposure and disease profiles. These populations appear to have lower risk of some diseases such as colon and lung cancer, cardiovascular disease and allergic disease, which has been attributed to frequent exposure to microbial agents and healthier habits, including reduced tobacco use and increased physical activity [1–5]. On the other hand, regular exposure to certain pesticides, UV radiation, diesel exhaust and solvents and high dust levels has been reported to be associated with increased morbidity, including risk of several cancer types [6], respiratory disease including non-allergic asthma and chronic obstructive pulmonary disease (COPD) [7,8], neurotoxic [9] and reproductive outcomes [10]. Studies of agricultural populations are of great interest in their own right (i.e., agricultural workers make up a large proportion of the working population worldwide), but they also contribute to a better understanding of disease risks associated with pesticides, other chemical, biological, and physical hazards for the general population because those exposures also occur outside agriculture. Additionally, these studies are suited to identify factors that may protect against particular types of cancer as well as allergies and other non-malignant conditions. Indeed, studies of agricultural populations have the potential to inform effective interventions to reduce disease burden in the general population.

In 2006, the US-National Cancer Institute (US-NCI) brought together principal investigators of a number of agricultural cohorts to develop an international agricultural cohort consortium to study cancer and other health outcomes in association with agricultural exposures. Investigators from thirteen cohorts from Canada, France, New Zealand, Norway, Republic of Korea and USA attended this first workshop. To foster a transition into an active consortium with the aim of studying exposure-disease associations not easily addressed by single cohorts, a second workshop was convened by the International Agency for Research on Cancer (IARC) and the US-NCI in October 2010 in France. At this meeting the representatives from nine countries and 18 studies agreed to form an active consortium coordinated by IARC, named AGRICOH.

The purpose of AGRICOH is to promote and sustain collaboration and data sharing/pooling to assess the association between various agricultural exposures and a wide range of health outcomes with a particular focus on those associations that cannot easily be investigated in individual studies because of rare exposures (e.g., use of infrequently applied chemicals) or relatively rare outcomes (e.g., cancer, neurologic and auto-immune diseases). AGRICOH will seek to identify potential health hazards as well as protective factors that may affect both agricultural and non-agricultural populations. These aims will be supported by assembling background information from participating cohorts, developing a plan for harmonization of core exposure and outcome variables, and pooling of data, effectively increasing sample size to yield statistically powered and robust data analyses. In addition, availability of biological specimens in 16 cohorts in the consortium offers the opportunity to pool biological material in support of specific research projects with a molecular focus. Future aims of AGRICOH are to identify additional existent cohort studies focusing on agricultural populations to join the consortium and to encourage establishment of new agricultural cohorts, in low- and medium-income countries, where the range is wider and the intensity of exposures is expected to be higher, but their role in health and disease is rarely documented.

The consortium focuses on cohort studies with a broad definition of agricultural exposures and populations. These include crop and animal farming activities and environments, active and retired agricultural workers, farm owners and their families, including the occupational and the residential milieu, professional groups exposed to specific agents used in farming, such as pesticides, or generated during farming, such as organic dust produced in settings such as grain or poultry production. Availability of biological specimens is not a pre-requisite to join the consortium. The majority of cohorts in AGRICOH research health outcomes in relation to occupational and environmental exposures with a focus on agricultural settings. The Korean Multi-Center Cancer Cohort, the Janus Serum Bank of Norway and the Ontario Health Study on the other hand, are general population cohorts with the first two studies encompassing a significant number of individuals from agricultural populations and the latter having potential to oversample in agricultural areas.

As of February 2011, AGRICOH comprises 22 cohorts from five continents. The studies are from South Africa (1), Canada (3), Costa Rica (2), USA (6), Republic of Korea (1), New Zealand (2), Denmark (1), France (3) and Norway (3) (see Table 1). In total, ten cohorts offer data on cancer incidence: the New Grain Worker’s Study (Saskatchewan, Canada), Cancer and Mortality among Workers of Banana Plantations (Costa Rica), the Agricultural Health Study (USA), the Next Generation Cohort of Agricultural Health Study (USA), the Marshfield Epidemiologic Study Area (MESA) Farm Cohort (USA), the Korean Multi-center Cancer Cohort, the Agriculture and Cancer Cohort (France), the Janus Serum Bank (Norway), the Cancer in the Norwegian Agricultural Population Cohort and the Ontario Health Study.

Other health outcomes studied in the past or planned for future research by cohorts in AGRICOH include respiratory (15 cohorts), neurologic (nine cohorts), auto-immune (five cohorts) and cardiovascular (five cohorts) diseases as well as reproductive outcomes (seven cohorts), allergic disorders (12 cohorts) and injuries (10 cohorts) (Table 1). The cohorts contain a wide range of population sizes (from a few hundred to over half-million persons), sub-groups (men/women, adults/children/infants, rural/urban, emerging farmers) and degrees of implementation (from completed follow-up to recently initiated recruitment). Sixteen cohorts have access to biological specimens on at least a sub-sample of participants. The type and extent of characterization of agricultural exposures varies widely across these studies. The majority use questionnaires (including exposure journals) to gather most of the exposure data. A few cohorts obtain information on occupation, duration of employment and information on lifestyle factors from direct record linkage to agricultural census data and/or national public health administration databases. Personal, household, and farm environment sampling offer additional exposure data in several cohorts. A number of cohorts have the capacity to conduct exposure assessment using stored biologic specimens or through field studies. Exposure to pesticides, fertilizers, endotoxins, mycotoxins, viruses, organic and inorganic dust, diesel and other exhaust gasses, and solvents are the main exposures documented in the cohorts in AGRICOH. Published articles providing a description of the cohorts in AGRICOH, when available, are cited in Table 1.

A Steering Committee of nine members encompassing a broad spectrum of expertise, in terms of health outcomes, (agricultural) exposures assessment and data pooling experience, will lead the discussion on issues related to AGRICOH during and in between annual meetings and will guide the development of future AGRICOH activities. A coordinator from IARC (Scientific Secretariat) will provide support to the Steering Committee and consortium. Meetings dedicated to the consortium to consolidate objectives and future plans are planned to be held on an annual basis.

The Consortium welcomes project proposals involving pooling of data from all interested researchers. The ultimate decision about whether to contribute data and participate in any particular project rests with the individual AGRICOH consortium members, i.e., the principal investigators of the cohorts. To facilitate future projects the need for data harmonization of selected core variables was discussed at the 2010 Workshop and will be initiated in May 2011. Each approved project will require some data harmonization. Eighteen research concepts for data pooling were discussed during the AGRICOH Workshop at IARC in October 2010 involving the study of cancer, respiratory, neurologic and other health outcomes in association with pesticides, organic dust and other exposures. These were supported by the consortium members and more developed proposal plans to guide data pooling in support of each of those concepts are in preparation.

AGRICOH, including 22 cohorts from nine countries in five continents welcomes new cohorts and proposals to research the association between agricultural exposures and health outcomes. Additional information on these procedures or on the consortium is featured in the webpage (http://agricoh.iarc.fr). The next annual meeting of the consortium is being planned for 12–13 September 2011 in Barcelona.