Proficiency testing for determination of lead and arsenic in cosmetics: comparison of analytical procedures and evaluation of laboratory performances

Abstract

A proficiency testing scheme (CNAS T0419) is described involving 217 laboratories in China as participants using their regular analytical methods for the determination of lead and arsenic in foundation cream cosmetics. The metrological comparability of results from various approaches was tested using simple linear regression analysis and analysis of variance as valid techniques. There were no significant differences in the values obtained after elimination of outliers by the Grubbs test. The normality of the distribution of the results submitted by the participants was examined with the one-sample Kolmogorov–Smirnov test. The distribution characteristics of result points for the paired samples were investigated using the Youden plot, which could greatly simplify the data analysis process. The assigned values of test materials for proficiency assessment were derived directly from the reported results of the participating laboratories. The measurement capabilities of laboratories for determining heavy metals, especially lead and arsenic in complex matrices were objectively assessed using z-scores, which could readily be used to compare each participant's value relative to the others and enabled the outlier laboratories to find out the dominant error sources such as systematic and random variations. The percentages of acceptable results of between- and within-laboratories were 86.0% and 84.2% for lead, and 84.6% and 83.6% for arsenic, respectively.