Objectives The objective was to estimate the agriculture injury rate in in the United States. The Bureau of Labour Statistics conducts Census of Fatal Occupational Injuries and Survey of Occupational Injuries and Illness but it excludes workplaces with 10 or fewer employees or self-owned farm operations and may underestimate the agricultural injury rate.

Method The Central States Centre of Agricultural Safety and Health partnered with National Agriculture Statistics Service to annually administer agricultural injury survey. In 2012, 6953 surveys were administered to a stratified random sample of 2007 Census of agriculture respondents in seven Midwestern States. The survey included questions on demographics, type, location and source of injury, body part injured, lost work time, and cost. The data were linked to Census of agriculture for farm level attributes. Univariate and multivariate logistic regressions were used to evaluate factors associated with adult operator injuries.

Results The cumulative incidence was 60.6 injuries per 1000 farm operators. Injury incidence was significantly higher in part-time compared to full-time farmers (79.3 vs 42.6 per 1000, p < 0.0001); farm size 1000 or more acres compared to 180–999 and 1–179 acres (91.6 vs. 60.5 and 45.4 per 1000, p = 0.002); at least one livestock compared to none (77.1 vs. 44.3 per 1000, p = 0.0004); and having a tractor with 100 or more horsepower (71.8 per 1000, p = 0.006).

Conclusions There were substantial differences in injury incidence by individual and farm attributes. These results may be used to develop targeted interventions to reduce agricultural injuries in the Midwestern United States.