PURPOSE: The lack of population-based occurrence data for many chronic conditions is considered to be a major gap in US disease surveillance capacity. The purpose of this pilot study was to evaluate the Marshfield Epidemiologic Study Area (MESA) for describing the incidence and prevalence of selected health conditions in a general population, and to identify possible differences by farm residency status. METHODS: MESA, an established resource for population-based research, covers 24 ZIP codes in Wisconsin where nearly all of the 77,000 residents receive their health care from the Marshfield Clinic and affiliated hospitals. Individual follow-up time is tracked, and farm residents are periodically enumerated (5,400 in 1998). We selected 88 mostly chronic health conditions of potential importance to farmers using archived diagnostic codes. MESA residents from 1995 were followed through 2002 to estimate age-specific and age-adjusted rates of diagnosis. Poisson and logistic regression were used to evaluate the effect of farm residency. RESULTS: Farm residents were at greater risk of diagnosis with alveolitis (prevalence odds ratio 11.4, 95% confidence interval 8.4-15.5), crushing injuries (1.7, 1.4-2.2), segmental dysfunction (1.6, 1.3-2.0), foreign body injuries (1.5, 1.4-1.7), dislocations (1.4, 1.2-1.5), and open wounds (1.2, 1.1-1.3). Male, but not female, farm residents were at greater risk of diagnosis with osteoarthritis (1.2, 1.1-1.4). Stress (0.7, 0.6-0.8), depression (0.8, 0.7-0.9), and substance use (0.5, 0.5-0.6) were diagnosed less among farm residents, suggesting possible influence of health care utilization. Overall incidence ranged from under 1 per 100,000 person-years for certain intestinal infections to 4429/100,000 for unspecified injuries. The incidence of selected cancer diagnoses in MESA was somewhat higher than published US rates, a possible indication of reduced specificity in case ascertainment. CONCLUSION: This pilot study demonstrates MESA can be a useful resource for surveillance of clinically detected disease among farm and non-farm residents in a general population, but further study must involve greater case validation and assessment of differential health care utilization.