Comparison of the results from two different External Quality Assessment Schemes supports the utility of robust quality specifications

Background: The definition of quality goals to evaluate the performance of laboratories participating in External Quality Assessment Schemes (EQAS) is currently not homogeneous. The aim of the work was to verify the applicability of quality goals based on biological variation comparing the results from two different EQAS. Methods: We evaluated the performance of the laboratories participating in two Italian EQAS, presenting similar characteristics in terms of number of participants, type of EQA-samples, and program organization. The results were obtained during 2007 for 27 components in the regional scheme of Lombardy (RGL) and in the national scheme of Prolarit (PRL). The percentage total error of single measurements was calculated for each reported EQA value. The total error values at the 68th percentile, at selected critical concentration values, were compared with maximum tolerable error derived from biological variation. Results: The performance of laboratories participating in the RGL scheme was significantly better. The frequency of satisfactory performance at, respectively, minimum, desirable and optimum levels was 98%, 80% and 59% in the RGL scheme, and 73%, 56% and 22% in the PRL scheme. Conclusions: Due to the different performance shown in the two programs, objective analytical goals based on biological variability constitute the optimal solution.