A review of the relationship between quality control and donor sample results obtained from serological assays used for screening blood donations for anti-HIV and hepatitis B surface antigen

Quality control principles, initially designed for manufacturing, have been applied to clinical chemistry and later to serological testing for infectious diseases. Methods for setting control limits have been included using the effects of assay performance on clinical outcomes. However, this approach assumes that the reactivity of patient samples change proportionally with the results of quality control samples tested in the same assay. Although this may be the case for clinical chemistry, this assumption has never been tested for serological assays. During a 9-month period, 177910 and 185684 donor test results for HBsAg and anti-HIV, respectively, were analysed and compared with 712 HBsAg and 710 anti-HIV results obtained from a single batch of a multimarker quality control sample obtained from the same assays. Results of the analysis showed that the negative donor results did not change in proportion with the changes experienced by the quality control sample results as determined by regression coefficients, particularly when results obtained from different reagent batches were compared. There were insufficient results obtained from positive samples for a similar analysis. However, investigations show that, even if the reactivity of positive donor samples changed in proportion with the change in reactivity of the quality control sample, few false-negative donor test results would have occurred.     

Comprehensive quality control programme for serology and nucleic acid testing using an Internet-based application

The National Serology Reference Laboratory, Australia (NRL) has quality assured the serology for high risk blood-borne infections since 1985, commencing with anti-Human Immunodeficiency Virus (HIV) enzyme immunoassays and later extending the programmes to other blood-borne virus testing and to nucleic acid testing. A quality control (QC) programme was considered the most appropriate manner in which to achieve real-time monitoring. An Internet-based application, EDCNet, facilitates the entry of results of QC sample testing and returns the analysed results instantaneously. Results can be displayed in a variety of tables and charts, so that QC results from blood service and diagnostic laboratories can be monitored. Comparison of results between laboratories using the same system offers monitoring of accuracy, while within-laboratory comparisons offer monitoring of the assay precision. More than 200,000 data points were submitted to EDCNet in 2002 from blood service laboratories as well as from diagnostic laboratories. Analysis of reagent batch variability was determined, e.g. the coefficient of variation between batches of seven assays used to detect anti-hepatitis C virus (HCV) antibodies ranged from 4.9% to 21.6%. Using EDCNet, laboratories can monitor both precision and accuracy of results by comparison with the results of other laboratories. The system may be a highly cost-effective method for maintaining quality.

Identifying the critical cut-points of a quality control process for serological assays: results from parametric and semiparametric regression models

Accreditation and Quality Assurance - Quality control programs rely on continuous monitoring which may generate large volume of complex data. Assessing the precision of a biological assay using...