Syndromic surveillance in veterinary medicine using laboratory submission data: lessons learned from two systems

Syndromic surveillance can be characterised as a process involving the continuous analysis of health data to provide immediate feedback. The data sources scanned should therefore represent high population coverage, be acquired continuously, in an automated manner, and be available in digital format. Data from diagnostic test requests often meet these requirements. Building on the experiences from developing syndromic surveillance in two institutions, the steps from data extraction to eventual aberration detection are described in this paper, and can be summarised as: classification of records into syndromes; retrospective evaluation of data to create baseline profiles following the removal of excessive noise and aberrations, and the identification of temporal effects; prospective evaluation of detection algorithms; and finally real-time monitoring and implementation. All steps described were implemented using open source software, and could be readily reproduced in other institutions