New strategies for agricultural injury surveillance using multiple electronic data sources

Background and Objective: The inadequacy of injury surveillance within agriculture is well-known. This is particularly true for non-fatal injury. Researchers are developing an agricultural injury surveillance system based on five sources of existing electronic data: ambulance reports, hospital discharge and emergency department data, death certificates, and electronic news clippings. Previous research suggests that these datasets contain farm injury records, yet many are incomplete. Because different fields tend to be missing depending on the data source, our objective is to study the effect of case-matching and merging on completeness and overall yield of farm injury cases. Methods: Through electronic match-merging, record completeness will be evaluated between all matching records. By the conference date we will be able to specify the decrease of missing data in key fields as a result of merging, as well as quantify the increase in the total number of farm cases identified. Results: Based on previous research, we anticipate increases in complete farm injury case records of at least 25%. Preliminary research data also indicate that falls, horse-related injuries and injuries with hand-held tools and machinery are more common farm injuries than is seen among fatal injuries. Conclusions: From this analysis, we will make a determination as to whether record-merging is an effective method of increasing the yield of farm injury cases for farm injury surveillance. Methodological challenges will be discussed.