Detection and correction of systematic laboratory problems by analysis of clustered proficiency testing failures.

Context: The Laboratory Accreditation Program of the College of American Pathologists monitors the performance of its subscribers in proficiency testing (PT). Failure to perform as expected prompts the program to query the laboratory.

Objective: To determine whether laboratories are correcting apparent problems when contacted by the program about repeatedly unacceptable performance in a diagnostic test.

Design: Retrospective analysis of 1 year's records (2002-2003) from the College's Proficiency Testing Exception Summary correspondence, which identifies clusters of PT failures. The analysis focused on those laboratories in which the Proficiency Testing Exception Summary algorithm identified repeated failures over 3 or 4 testing events; PT performance is monitored as a condition of accreditation. During 1 survey year, approximately 6300 accredited laboratories collectively tested approximately 1,205,000 analytes and submitted results to their PT providers on more than 3,500,000 PT challenges. During the period of observation, 14,085 Proficiency Testing Exception Summary reports were mailed to participants. Educational materials were included to help laboratories identify and correct each PT failure.

Results: There were only 1304 cases of repeated PT failures after the initial correspondence from the accreditation program (9.3%). Of these, there were only 119 cases of unsatisfactory results on the subsequent PT event (9.1%). All systematic problems were resolved by the conclusion of the third round of correspondence.

Conclusions: Accredited laboratories generally perform well in proficiency testing. Identification of clusters of PT failures by the accreditation provider can help those laboratories having analytic difficulties to investigate and correct the problems.