Point-of-care testing error: sources and amplifiers, taxonomy, prevention strategies, and detection monitors.

CONTEXT

In a survey performed 4 years ago, testing venues doing only point-of-care testing (POCT) made up 78% of sites for patient testing licensed under federal regulations.

OBJECTIVES

To identify sources of POCT error, to present a classification of such errors, to suggest strategies to prevent errors, and to describe monitors that assess and reduce the frequency of errors.

DESIGN

To identify sources of POCT error, large studies of error among US Federal Certificate of Waiver laboratories (CoWs) and practitioner-performed microscopy certificate holders were reviewed. To facilitate investigation and management of POCT error, a taxonomy of such errors (modified from a classification previously published by Gerald Kost) was used to identify 4 steps with error potential in each of the 3 phases (ie, preanalytic, analytic, and postanalytic) of the POCT process. To prevent observed POCT errors, 4 strategies are suggested: direct observation of instrument/method functionality, structured observation of method performance, proficiency testing/use of relevant test scenarios, and autonomation. To assess frequency of errors, a quartet of indices are introduced as detection monitors: order documentation, patient identification, specimen adequacy, and result integrity.

RESULTS

Three sources of POCT error were identified: operator incompetence, nonadherence to test procedures, and use of uncontrolled reagents and equipment. Three other characteristics of many point-of-care tests amplify their risk of error: incoherent regulation, rapid availability of results, and the results' immediate therapeutic implications. Two members of the quartet of detection monitors, order documentation and specimen adequacy, are relatively difficult to measure and are controversial, but the other 2, patient identification and result integrity, are easier to assess and are relatively widely accepted.

CONCLUSIONS

Point-of-care testing errors are relatively common, their frequency is amplified by incoherent regulation, and their likelihood of affecting patient care is amplified by the rapid availability of POCT results and the results' immediate therapeutic implications. The modified Kost taxonomy offers a reasonable approach to the identification of POCT errors. Direct observation of test functionality, structured observation of test performance, and testing the competence of POCT operators, as well as autonomation of devices, are strategies to prevent such errors. In this context, we suggest monitoring POCT order documentation, patient identification, specimen integrity, and result reporting to detect errors in this sort of testing.