Context: Major clinical decisions are based on the laboratory test results where preanalytical errors are an important cause of repeat collections in patients. Identification of problem areas and continuous training of phlebotomy staff are important tools in reducing these errors.
Aims: In this study, we looked at the most common causes of sample rejection in our setting and the efficacy of the corrective measures and training processes for staff in reducing preanalytical errors.
Settings and designs: This prospective study was conducted at the laboratory diagnostic services of a tertiary care oncology center, with a hematopoietic stem cell transplant unit during the period of 2012-2017 in two phases. Sample rejections from various wards were analyzed for types of rejections.
Materials and methods: In the first phase, we analyzed the problem areas (year 2012). Following a root cause analysis, current practices of training were altered. In the second phase (2013-2017), we studied the effects of these measures.
Statistical analysis used: The percent variation and P value for significance in sample rejections were calculated.
Results: During the year 2012, 0.36% samples were rejected by laboratory. Following interventions in the period from 2013 to 2017, samples rejected dropped to 0.19% (P < 0.0001), 0.09% (P < 0.0001), 0.09% (P = 0.8387), 0.05% (P = 0.0004), and 0.05% (P = 0.329), respectively. The reduction was significant from surgical oncology ward (P = 0.0107) and intensive care unit (P = 0.0007). From 2013 to 2017, errors significantly reduced to 0.015% for hemolyzed samples (P = 0.0001), 0.005% for contaminated samples, 0.036% for clotted samples, and 0.019% for labeling errors.
Conclusion: Intervention in the form of targeted training helps reduce errors and improves the quality of results generated and contributes to better clinical outcomes.
Keywords: Intervention; laboratory errors; quality assurance; root-cause analysis; sample rejection; training efficacy.
Copyright: © 2019 Journal of Laboratory Physicians.