Comparable laboratory performances in the analysis of lead in control samples and in fresh human blood

The validity of quality control programs is based on the assumption that the control samples can be commuted with the real samples, so that, according to the results obtained by a given laboratory on control samples, it is possible to produce reasonable predictions about the reliability of results obtained on real samples. During the implementation of a program of biological monitoring of the general population against the risk of saturnism, a great number of data have been collected allowing the evaluation--as far as blood lead determination is concerned--of the real predictive power of the results obtained by a laboratory during a quality control program, according to its specific level of analytical reliability. The results obtained by eleven laboratories in the analysis of control samples have been compared, by regression analysis, with the results obtained on the same samples by the reference laboratory. The same procedure has been adopted for the results obtained by each laboratory and by the reference laboratory in the duplicate analysis of about 10% of the real samples collected by each center during the biological monitoring program. The comparison between the regression parameters obtained, in both cases, for each laboratory has not produced evidence of systematic differences. Furthermore, a non-parametric evaluation of the data (based on the magnitude of the differences between the results of the laboratories and the results of the reference center) shows, in most cases, similar laboratory performances in the analysis of control samples and of real samples.