Background Identifying emergent themes in safety begins with data that illustrate changes in rates of injury and illness. Oftentimes, these data are borne from surveillance systems. Dairy farming is a useful case study, as it is among the most hazardous domestic jobs and increasingly technology-driven. Thus, surveillance of injury and illness on dairy farms is a chore ripe with challenge as well as necessity. Located in the second largest dairy producing state and with electronic health records covering over 90% of the healthcare service population, the study provides a valuable example in surveillance.

Description of the problem The agricultural and healthcare industries in the U.S. are as decentralised and fragmented as the data sources that track them. Prior and current agricultural injury surveillance systems suffer from a dearth of data and difficulty determining farm exposures from existing resources. Response rates on injury surveys tend to be low, thus combining data from different sources and with different operationalizations is necessary but provides an additional problem to effective surveillance.

Results A surveillance program is built by first leveraging the grounded knowledge of agriculturalists. Relationship-building is imperative in accessing the best data available and appropriately linking disparate data sets. Dairy license information, farm vehicle registrations, and federal farm subsidy recipients are used to enumerate a sample of dairy farmers. This information is then linked to electronic health records, health payer claims, and mail surveys in order to cull injury and illness events, as well as current safety practices. Rates of injury and illness are examined at regular intervals.

Conclusions The current project may serve as a model for creatively overcoming the unique challenges of passive, ongoing surveillance in a decentralised system, with broader implication for public health safety surveillance programs in other world regions.