Jornal de Química Clínica e Medicina Laboratorial

Jornal de Química Clínica e Medicina Laboratorial
Acesso livre

Abstrato

Evaluation of Sigma Metrics of Commonly Assayed Biochemical Parameters in a Clinical Laboratory

Sharma Kumar, Reshma Datta, Neelu Shetty

Background: Six Sigma is a quality management strategy to improve the quality of processes and lays emphases on identification and removal of defects. Implementation of Six Sigma across laboratory processes allows identification of errors and introduction of novel approaches towards cost reduction without sacrificing quality. Keeping this in view, the study laboratory aimed to gauge the process performance of 19 routinely assayed parameters on sigma scale– that will help in assessing the laboratory’s performance and will enable in working out and choosing the correct approach towards improvement of problem analyte performance.
Methods:
Quality Control data was harvested retrospectively from August 2019 to December 2019. Sigma metrics was calculated for 19 biochemical parameters tested on Vitros-5600 using Total Allowable Error (TEa), Coefficient of variation (CV%) and bias (%). Quality Goal Indices of the problem analytes were calculated to identify the cause of error.
Results:
The following problem analytes were identified in this study having a sigma score of <3- Urea, ALT, ALP, Sodium, Calcium and Iron. QGI was calculated for these parameters to identify the area requiring improvementimprecision, inaccuracy.
Conclusion: The study concluded that sigma metrics is a good quality tool to assess the analytical performance of a clinical chemistry laboratory and stringent internal QC rules need not be adopted for methods with sigma ≥ 6. Also, false rejections in such cases can be minimised by relaxing control limits to 3S. However, for a problem analyte with sigma metric below 3, root cause analysis should be performed along with improvement in method performance before it can be routinely used.

Isenção de responsabilidade: Este resumo foi traduzido com recurso a ferramentas de inteligência artificial e ainda não foi revisto ou verificado.
Top