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

Handan Wand,Wayne Dimech,Robert Freame,Kathy Smeh
DOI: https://doi.org/10.1007/s00769-017-1265-9
2017-05-19
Accreditation and Quality Assurance
Abstract:Quality control programs rely on continuous monitoring which may generate large volume of complex data. Assessing the precision of a biological assay using quality control processes is essential to evaluate the daily variations in a testing system. Variation can be introduced by reagents, instruments and operators, as well as biological changes in the populations screened. We have proposed a statistical analysis framework which combines novel statistical analysis and visualization techniques to determine the functional relationship between the quality control and blood donor’s negative results. Flexible semiparametric regression techniques were used to determine the functional relationships between blood donor’s hepatitis B surface antigen and anti-HIV test results and the reactivity of quality control samples over a period of 9 months. We demonstrated that the use of semiparametric regression models in conjunction with the probabilistic approaches may bring comprehensive insight into understanding the significant temporal features of the data and its impact on patient’s test results. In the absence of clinically relevant cut-point(s), data-driven methodologies, such as the one described in this study may potentially have significant benefits and widespread applications.
chemistry, analytical,instruments & instrumentation
What problem does this paper attempt to address?