Method of Characterization

There are also RMs which are prepared for a specific application and are used for validation of relevant methods. Cobbaert et al. (1999) made use of Ion Selective Electrode (ISE)-protein-based materials when evaluating a procedure which used an electrode with an enzyme-linked biosensor to determine glucose and lactate in blood. Chance et al. (1999) are involved with the diagnosis of inherited disorders in newborn children and they prepared a series of reference materials consisting of blood spotted onto filter paper and dried, from which amino-acids can be eluted and 

Apart from innovative work, RMs are essential during exerdses such as the introduction to a laboratory of a method from elsewhere or the transfer of an established method onto new instrumentation. Even where the conditions for the analysis have been standardized by the manufacturer of a reagent kit, some validation work should still be undertaken so as to have documented data for quality assurance purposes, e.g. accreditation, as a basis for IQC, for later reference when problems which may be related to equipment, reagents or staff etc. need to be investigated. 

Validation work indudes the determination of within-batch and between-batch imprecision, linearity, sensitivity and limit of detection - which can be carried off with any suitable spedmen. However, to demonstrate the accuracy of the procedure, to investigate possible interferences and to establish reference ranges, it is necessary to analyze RMs with known concentrations, ideally using certified materials. Where none are available, accuracy may be assessed using other approaches, but these are less reliable and are more complicated to perform (Taylor 1995). 

Fraser has extensively discussed this relationship between laboratory work and clinical needs (Fraser and Hyltoft Peterson 1993) and has recently addressed the role of documented analytical quality as derived from measurements of RMs (Fraser and Hyltoft Peterson 1999). Among the concepts proposed by Fraser and his colleagues, it is suggested that analytical imprecision should be 0.50 CVi and bias should be 0.25 (CVi + where CVj is the within-subject biological variation (i.e. changes from day to 

Evaluation of a new method may include comparison of results obtained by the analysis of reference materials using established procedures. The purpose is to demonstrate that the novel approach provides results that are at least as reliable as an accepted technique. This same approach may also be used to discover which, of a number of techniques, are preferred. Comparison of analytical techniques and/or methods is considered further in the discussion of external quality assessment, below. 

---