| Abstract: | We developed of computer programs to evaluate the clinical reliability of test results by comparing each new result with previous results for the same patient, and to signal discrepancies in real time. These "delta check" discrepancies are noted, and they must be reviewed by the laboratory staff before results can appear on a patient's record. During a month, I reviewed 1403 such delta check messages and detected 55 (3.9%) that could not be explained on the basis of the patient's clinical condition. Of these, 23 represented true laboratory errors, which were corrected. The recognition of discrepancies before they appear on patients' reports has facilitated the operation of the clinical chemistry laboratory. Mislabeled and otherwise mishandled specimens are discovered before erroneous results appear on a patient's record. |
