Use of middleware data to dissect and optimize hematology autoverification

In this 2021 journal article published in the Journal of Pathology Informatics, Starks et al. of the University of Iowa Hospitals and Clinics present their case for the value of autoverification data evaluation despite the inherent challenges of extracting and analyzing data related to autoverification rules from informatics system. Noting that most autoverification rules were housed in their middleware solution and (and not the LIS, which would require additional fields currently not present), they examined instrument- and middleware-generated flags generated from “complete blood count (CBC) with white blood cell (WBC) count differential (Diff) and the ‘a la carte‘ ordering of individual CBC components.” Their analysis resulted in two significant changes to their autoverification rules related to those tests, resulting in “improved efficiency and lower rerun rates.” However, these insights did come at some investment in time and resources, requiring third-party data retrieval methods and “extensive cleanup and formatting” of the data to actually identify flagged specimens. The authors conclude that these downsides highlight “opportunities for improvement in software tools that allow for more rapid and routine evaluation of autoverification parameters.”

Please to read the entire article.