Discussion on evaluation mode of proficiency testing for heavy metal and harmful element detection in medicines / 中国中药杂志

This study aims to explore the evaluation model for the proficiency testing of heavy metal and harmful element residues in pharmaceuticals,and to provide reference for the proficiency testing program and proficiency testing result in the field of residue analysis. The proficiency test result of cadmium determination in honeysuckle as an example. The algorithm A,NIQR,and Horwitz function are used to calculate the assigned value and the standard deviation. Z was obtained at the same time. If | Z | ≤2,the result is satisfactory. If 2< | Z | <3,the result is questionable. If | Z | ≥3,the result is unsatisfactory. In addition,the median value is the assigned value,and deviation(D%) is used. If D% is not more than 16%,the result is satisfactory; if D% is more than 16%,the result is unsatisfactory. After analysis,in the results of questionable or dissatisfied laboratories calculated by algorithm A and NIQR,the deviation error of some data is within the scope of the standard. In the results of the satisfactory laboratory evaluated by the Horwitz function,some data deviation errors far exceed the standard range. The evaluation result of the D% meets the requirements. According to heavy metal and harmful element trace analysis methods,this study is the first to apply D% to the evaluation of the detection ability of heavy metals and harmful elements in pharmaceuticals. This method makes the evaluation result more reasonable,and has important reference significance for the evaluation of other proficiency test results.

Review on the statistic methods used in the proficiency testing in the drug control field / 中国药学杂志

OBJECTIVE: To explore the statistical methods for proficiency testing in the drug control field. METHODS: The statistical methods were sorted out with data processing used in inter-laboratory and proficiency testing by various industries both at home and abroad. The Grubbs test and Dixon test in classical statistical method were focused on as well as quartile method and iterative method in the robust statistical method, and the characteristics of the different statistic methods was summarized respectively. RESULTS: Every method has its advantages and application conditions. CONCLUSION: It is suggested that proper statistical method be chosen according to the data characteristics of the test results.

ISO 1 5 1 8 9:2 0 1 2 and External Quality Assessment / 现代检验医学杂志

External quality assessment (EQA)as an effective tool for quality control can help clinical laboratories to discover quality problems during testing process and prompt clinical laboratories to improve their testing quality.Samples used for traditional EQA frequently have matrix effects which assign the target value by grouping.Ideal samples for EQA are com-mutable which have the same numeric relationship between different measurement procedures as that expected for patients’ samples.And they can assess accuracy against a reference measurement procedure or a designated comparison method.When analyzing,robust statistical method can be used to reflect the variability of testing results between laboratories more truly. And the evaluation standards should be desirable.If unsatisfactory results appear,we should investigate every aspect of tes-ting process,find the root cause,and take corrective action.On condition that no official EQA scheme is launched for some analytes,it is necessary for us to evaluate the performance of analytes by alternative evaluation procedures.