Bedside glucose monitoring quality control practices. A College of American Pathologists Q-Probes study of program quality control documentation, program characteristics, and accuracy performance in 544 institutions.

OBJECTIVE

To investigate the adequacy of bedside glucose monitoring (BGM) quality control documentation and monitoring, characterize program structure and organization, and identify characteristics associated with the ability to produce accurate results.

DESIGN AND SETTING

College of American Pathologists Q-Probes laboratory quality improvement study in 544 institutions.

MAIN OUTCOME MEASURES

Percent compliance with quality control (QC) documentation, appropriate corrective action, and frequency of inappropriate patient testing, and the percentage of BGM results within +/-10% and +/-15% of a corresponding clinical laboratory glucose result.

RESULTS

Five hundred forty-four institutions reviewed a total of 19543 individual QC paper documents from 2543 separate BGM instruments. Ninety percent of QC determinations that should have been performed and noted on these documents were recorded; of those performed, 2.8% of QC results were outside of the acceptable range. Thirty-two percent of the out-of-range QC results had no record of corrective action. There were 527 reported instances of one or more patients being tested when there was no record of corrective action for out-of-range QC results. There were 20665 undocumented potential QC events, with 2053 instances of one or more patients being tested when there was no documentation. Two hundred forty-two institutions submitted 6653 paired BGM and clinical laboratory results for comparison. Approximately 56% of BGM results were within +/-10%, and 74% were within +/-15% of the corresponding clinical laboratory result. Factors associated with better accuracy performance are discussed.

CONCLUSIONS

There is a need for improving compliance with QC documentation, improving appropriate corrective action follow-through, decreasing the frequency of inappropriate patient testing, and improving BGM accuracy performance. We provide recommendations for improvement.