Building and evaluating the autoverification of coagulation items in the laboratory information system

The increasing number of specimens, the decreasing number of proficient staff, and clinical demands make autoverification important in the laboratory's future development. However, although autoverification has been widely applied, a full coverage of its technique is hard to achieve in a laboratory. The establishment of the autoverification rules and parameters is still unclear. Therefore, the aim of this study was to fill the vacancy by the method described below.

First, all logic processes and autoverification rules were established with reference to the Clinical and Laboratory Standards Institute (CLSI) document Auto10-A-Autoverification of Clinical Laboratory Test Results-Approved Guideline. Second, we established rules in the LIS, collected clinical specimens for validation, and assessed the results in a "live" environment.

The whole passing rate of autoverification was 82.0% and the manual intervention rate was 18.0%. The highest passing rate of a single project was activated partial thromboplastin time (APTT), approximately 96.1%, followed by prothrombin time (PT), approximately 95.1%, and fibrinogen, approximately 90.9%. The turnaround time (TAT) was shortened by 31.8% (90 minutes vs. 132 minutes).

Through implementing the autoverification, which accelerated verification and decreased the TAT and the odds of human review errors in the released results, we can save more time and concentrate on verifying the abnormal results and proceeding emergency tests.