Patient based quality control and integration of results to the approval system

Aim: Laboratory test results are very important in diagnosis and treatment of diseases. So, ensuring quality assurance of laboratory test results is a requirement. In the clinical laboratories, internal quality control (IQC) procedures are performed before every run; but, the continuity of the analytical performance during the run is arguable. On the other hand, IQC procedures couldn?t define the errors while patient samples are being analyzed. In this study, we aimed to evaulate whether monitoring median results of population which is one of the patient based quality control (PBQC) procedures, could be used in improving the safety of the test results. Materials and Methods: The test results of patiens who applied to the Central laboratory of Dokuz Eylul University Hospital between 01.01.2011 and 01.01.2012 for routine clinical chemistry analysis were taken from laboratory information system (LIS). First, number of the test results per day were determined; then, the test results having the number of 50 or higher were enrolled in the study. These tests are Na, K, Cl, glucose, BUN, creatinine, AST, ALT, GGT, ALP, CK, LDH, uric acid, TG, cholesterol, HDL-C, total protein, albumin, total bilirubin, direct bilirubin, calcium, iron, UIBC, phosphorus, vitamine B12, folic acid, CRP and ferritin. Because of the differences of the patient population, median values were calculated for each subgroups of inpatient, outpatient, daytime and shift. By using the results of means and medians obtained from the data for one-year period, the graphics were made: histograms to evaluate the distributions, line graphics to compare of the median results and mean results, and control charts to evaluate the process stability. Results: Based on the results of on-year period, we defined the significant differences between means and medians; furthermore, most of the tests didn?t have normal distribution. By using the X-MR control charts and mean/median ratios, we determined higher error detection performance in all groups for Na, K, Cl, glucose, BUN, creatinine, AST, ALT, ALP, LDH, uric acid, TG, cholesterol, HDL-Cholesterol, total protein, albumin, total bilirubin, direct bilirubin, calcium, UIBC, phosphorus, vitamine B12 and folic acid; lower error detection performance in the groups of daytime for CRP, iron, ferritin and in the groups of shift for GGT, CRP ve CK. Conclusion: As a complementery of IQC procedures, PBQC procedures could improve the analytical quality assurance. As PBQC adopted to LIS, in particular, the laboratories which have the higher and moderate test capacity could reach and assess the data easily; so, analytical shifts during run could be defined early, and the erroneous results could be prevented. In additon, with retrospective evaluation of the control charts, useful information could be obtained about the analytical system. Keywords: Patient based quality control, quality control, median, X-MR charts