External Quality Assessment of Point-of-Care Methods: Model For Combined Assessment of Method Bias and Single-Participant Performance by the Use of Native Patient Samples and Noncommutable Control Materials

BACKGROUND: An important objective in external quality assessment (EQA) is to evaluate systematic deviations between methods. However, this is not possible when noncommutable control materials are used. The aim of this study was to develop an EQA model that incorporates a method bias evaluation using native patient samples into EQA schemes in which noncommutable materials are used. METHODS: The model was applied twice in a point-of-care (POC) international normalized ratio survey among 1341 and 1578 participants. To estimate bias, about 100 native patient samples for each POC method were analyzed by a selected group of "expert" primary healthcare centers and on a designated comparison method. In addition, the expert centers as well as all the other EQA participants analyzed 2 noncommutable control materials, and method-specific target values were established. Both method bias and the deviation of a single-participant result from the method target value were evaluated against analytical quality specifications, making combined assessment possible. The best-case scenario occurred when both results were within the quality specifications. RESULTS: Two POC methods fulfilled the quality specification for bias, whereas one did not. The best-case scenario was achieved by more than 90% of the participants using the methods with no bias, whereas none of the participants using the method with unacceptable bias achieved this result. CONCLUSIONS: We propose an EQA model for which the bias of POC methods can be evaluated in situations in which commutable control materials are not available.