Search

E the correlation between laboratory personnel standards and the accuracy and reliability of laboratory results was one of the original studies called for in the Clinical Laboratory Improvement Amendments of 1988 (CLIA).1 A review of the research literature, which was published in 1992, found that most published research focused on the relationship between performance in proficiency testing and the credentials of personnel employed by the participating laboratory.2 The Centers for Disease Control and Prevention funded a study by Christian and colleagues,3 which explored ways in which personnel competency was being assessed in a small group of hospital, blood bank, independent, physician office, and group practice laboratories. This study found that no consistent method was being used for competency assessment. Given the difficult task of finding a tool that measures how well an employee actually performs his or her job, it is not surprising that competency assessment remains a challenging assignment. In a recent review of the methods commonly used to assess the clinical competence of residents in internal medicine, Holmboe and Hawkins4 found no single assessment method that could successfully evaluate clinical competence. The College of American Pathologists Q-Probes study5 of the tools used by 522 institutions to assess laboratory employee competence adds

Objective. —To examine the frequency and characteristics of problems in laboratory testing in primary care physicians' offices and their impact on health care. Design. —Prospective descriptive study in which participating office-based primary care clinicians reported each occurrence of any laboratory incident during a 6-month study. Each identified problem was reported on a structured data collection instrument with an open-ended description of the problem. Setting. —Primary care physicians' offices in North America. Participants. —One hundred twenty-four primary care clinicians in 49 practices of the Ambulatory Sentinel Practice Network (ASPN). Main Results. —A total of 180 problems were reported, yielding a crude rate of 1.1 problems per 1000 patient visits. Problems involving test ordering and specimen handling were the most common (56%), while those attributable to the test analysis itself accounted for 13% of the total. In the judgment of the practice staff, 27% of the reported problems had an impact on patient care. Of the 24 cases for which the specific impact was known and reported, half of the impacts were minor and about half were significant, as judged by whether or not the diagnosis and/or treatment of the patient was measurably affected. Conclusions. —Clinically apparent problems with laboratory testing in primary care were found at a rate of 1.1 problems per 1000 patient visits. Of the laboratory problems that occurred in this study, 27% were judged by the physician to have an effect on patient care. (JAMA. 1996;275:635-639)

Enzymic activities of both γ-glutamyl transpeptidase and 5'-nucleotidase have been evaluated in patients in whose cases hepatobiliarγ-tree disease was suspected. γ-Glutamyl transpeptidase was a better indicator of diseases of the liver, bile duct, or pancreas than 5'-nucleotidase in nine of the ten conditions studied. 5'-Nucleotidase was found to be almost as useful as γ-glutamyl transpeptidase in indicating hepatobiliary disease in patients with granuloma of the liver, acute pancreatitis, congestive heart failure, rheumatoid arthritis, and carcinoma metastatic to the liver. γ-Glutamyl transpeptidase was most elevated in patients with primary carcinoma of the pancreas. Patients with carcinoma metastatic to bone or with neurologic conditions were found to have both enzymes elevated, but elevations occurred in a much lower proportion of patients than in those with hepatobiliary-tree disease. Liver scan findings correlate better with γ-glutamyl transpeptidase than with 5'-nucleotidase or alkaline phosphatase enzymic activity.