Evaluation of Preanalytical Point-of-Care Testing Errors and Their Impact on Productivity in the Emergency Department in the United States

Abstract

Abstract Background Preanalytical errors due to poor sample quality or improper sample handling may occur with point-of-care testing (POCT). Methods A retrospective analysis was conducted using deidentified records for 15 479 i-STAT® cartridges run at the Oregon Health & Science University Emergency Department (ED) between December 2015 and August 2016. Data were collected from electronic health records and device middleware for CG4+, CHEM8+, and Troponin cartridges. The frequency of POCT errors was evaluated by cartridge type. The effect of user experience on error frequency, impact of error on hospital length of stay (LOS), and test turnaround time (TAT) were all evaluated. Direct costs incurred due to Chem8+ and Troponin cartridge waste and indirect costs as avoidable nursing staff labor were estimated over 2 years. Results A total of 935 erroneous results were identified (6.0% of all cartridges). Three hundred seventy-two (2.4%) were unusable results, and 563 (3.6%) were cartridge errors, of which 163 were classified by device error codes as poor sample quality/improper sample handling. Error rates were inversely correlated with user experience based on number of tests performed during the 9-month period. Compared to nonerroneous results, test TATs and LOS were significantly longer with erroneous results (P < 0.01). Over 2 years, direct costs incurred due to cartridge waste was $45 000, and indirect cost was estimated between 486 and 729 h in avoidable nursing labor. Conclusions Preanalytical POCT errors were inversely correlated with user experience and significantly impacted clinical productivity in the ED based on LOS and test TAT.