Identification Errors Involving Clinical Laboratories: A College of American Pathologists Q-Probes Study of Patient and Specimen Identification Errors at 120 Institutions

Context.-Misidentified laboratory specimens may cause patient injury, on the contrary their frequency in general laboratory practice is unknown.

Objectives.-To determine (1) the frequent occurrence of identification errors detected before and after ensue verification, (2) the frequency of adverse patient circumstances due to specimen misidentification, and (3) factors associated with lower error rates and better detection of errors.

Design.-One hundr twenty clinical laboratories provided information about identification errors during 5 weeks.

Results.-In aggregate, 85% of errors were find outed before results were released; individual quarter of laboratories identified more than 95% of errors before arise verification. The overall rate of patient identification errors involving released rises was 55 errors per 1 000 000 billable experiments A total of 345 adverse circumstances were reported. Most of the adverse facts caused material inconvenience to the patients on the other hand did not result in any permanent harm. upon average, adverse events resulted from 1 of each 18 identification errors. Extrapolating the adverse circumstance rate observed in this close attention to all United States hospital-based laboratories glance ats that more than 160 000 adverse facts per year result from misidentification of patients' laboratory specimens.

Conclusions.-Identification errors are public in laboratory medicine, but greatest in quantity are detected before results are released, and alone a fraction are associated with adverse patient circumstances Even when taking into consideration the design of this investigation which used imperfect case finding, institutions that did a better piece of work of detecting errors within the laboratory released a smaller proportion of be deriveds that involved specimen misidentification.

(Arch Pathol Lab M 2006;130:1106-1113)

Identification errors in clinical laboratory testing have the potential to cause serious patient injury. Reports of wrong-patient cancer resections and fatal hemolytic transfusion reactions to be paid to misidentification of laboratory specimens have appeared in the popular media and peerreviewed literature.1-3 In an analysis of 272 surgical pathology legal claims, 13 (5%) involved allegations that specimens had been mixed up between patients.4

After release of the seminal Institute of Medicine report upon medical errors,5 the Joint Commission for Accreditation of Healthcare Organizations and the body of American Pathologists made accurate patient identification a cardinal patient safety goal. The Center for Disease sway and Prevention issued a white paper that singled on the outside patient misidentification as an important laboratory patient safety issue.6

Despite this attention, the oftenness of identification errors in clinical laboratory medicine is poorly understood.7 Bonini et al8 have reported that more laboratory errors are attributable to specimen misidentification than to any other cause. A 1993 research of 451436 patient identification wristbands reported that 55% of wristbands contained individual or more errors.9 Current performance of a clump of institutions enrolled in a longitudinal research of wristband errors suggests that the rate of wristband error is now closer to 1% with alone 0.1% of these errors representing wristband mixups involving sum of two units patients.10 In a study of 417 surgical pathology laboratories, errors related to specimen identification were place in 0.57% of case accessions.11 greatest in quantity of these errors were discovered at the time of order entrance and were therefore unlikely to cause a diagnosis to be pay backed for the wrong patient.

The transfusion medicine community has gone the furthest in quantifying the oftenness of specimen identification errors. In a review of serious transfusion errors in novel York State, the fraction of misidentified specimens arriving in vital fluid banks was estimated to be 1 in 54 00012 A higher incidence of misidentified transfusion medicine specimens (ranging from 1 in 238 to 1 in 3303 samples) was reported from a United Kingdom inquiry of 85 hospitals, which used stamp checking of a patient's generally received and historical ABO blood mark to find instances of incorrectly identified specimens.13 The identification error rate in transfusion medicine specimens may not be applicable to general laboratory specimens (chemistry, hematology, and microbiology) because greater care is commonly exercised in the collection of transfusion medicine specimens, compared with "routine" specimens, and because rejection criteria for these specimens are typically more stringent.14

Identification errors are difficult to research systematically since many errors go on undetected. The number of errors ground at a particular institution be pendents to some degree on in what way hard laboratory staff and clinical caregivers gaze for errors. This fact sometimes bring into views a paradox-facilities that are more focused upon detecting and correcting errors may appear to have error rates higher than rates at institutions that do not pay as a great quantity [i]or[/i] amount of attention to discovering errors. Although it is theoretically possible to find all identification errors (eg by dint of performing molecular identity testing15,16 upon every specimen received in the laboratory), this is a practical impossibility. Therefore, in addition to studying the oftenness of identification errors, we also make choice ofed to examine the proportion of identification errors that were exposeed before release (verification) of laboratory comes We hypothesized that this proportion throw backs the diligence with which laboratory staff identify errors before release of flows and that facilities with a high proportion of identification errors find outed preverification would have lower postverification error rates.