目的：设计和分析常规化学定量检测项目目标不确定度的确定方式，定期评估、评审不同阶段检验结果的测量不确定度变化，探讨其在质量改进和持续改进中的价值。方法以生物变异质量规范、“CNAS-TRL-001 CNAS 技术报告”为基础，设计5种目标不确定度的确定方式：方式1“基础方式”（不同等级目标不精密度和目标偏移）；方式2生物学变异不同等级总允许误差；方式3目标变异偏倚相对扩展不确定度评估值；方式4生物学变异不同等级目标相对扩展不确定度；方式5分析质量指标目标相对扩展不确定度。设计并尝试使用不确定度评价指数（UEI）评审不同阶段不确定度的变化。结果针对2013年常规化学具有计量溯源性并参加卫生部室间质评的14个项目，方式2、3、4达到目标不确定度的评估结论无明显差异。方式5中9个项目达到目标不确定度要求，占64．3％。UEI＜0分别有 TP、ALT、BUN、UA、CK，占35．7％；其余9项 UEI＞2％。结论方式5分析质量指标目标相对扩展不确定度，同时兼顾质量指标对实验室内复现性精密度和偏倚的要求，容易为实验室接受；方式4生物学变异不同等级目标相对扩展不确定度，与检验结果不确定度的评估方式相同，优于方式2和方式3；方式1是基础方式，是检验结果未达到目标不确定度的具体原因分析方式。UEI 评审不同阶段检验结果不确定度的变化，较质量管理方式可以更敏感地发现检验结果准确性的改变，其价值和实用性需要在实践中进行验证。%Objective To discuss the value in quality improvement and continuous improvement through the way for determina-tion of target uncertainty in chemical quantitative detection project and regular evaluation of measurement uncertainty in different phase.Methods Based on the biological variability of quality specification and CNAS-TRL-001:CNAS technical report,to compare the five kinds of determination of target uncertainty.Method 1:the fundamental way(calculate the different levels of target impre-cision and bias);Method 2:biological variation of different grades of total allowable error;Method 3:the relative expanded uncer-tainty evaluation value based on target imprecision and bias;Method 4:the target relative expanded uncertainty based on biological variation of different grades;Method 5:the target relative expanded uncertainty based on the quality index of analysis.We used un-certainty evaluate index(UEI)to review the changes of uncertainty in different phase.Results The 14 conventional chemistry tests in 2013 with metrological traceability and participate in the Ministry of health EQA were as the target.There was no significant difference among the 2,3,4 method,the ratio of which reached the ideal value of uncertaninty target were not significantly different. In method 5,9 projects achieved the target of uncertainty requirements,accounted for 64.3%.TP,ALT,BUN,UA,CK,these5 pro-jects′UEI were less than 0,accounted for 35.7%;other 9 projects′UEI were more than 2.0%.Conclusion Method 5:the target relative expanded uncertainty based on the quality index of analysis which is based on "WS/T403-2012" can give consideration to the quality standard of repeatability precision and bias in the laboratory at the same time,and is easy to be accepted for laboratory;method 4:the target relative expanded uncertainty based on biological variation of different grades is in the same way with the eval-uation of test results uncertainty,is better than method 2 and 3;method 1 is the fundamental way,can give the specific reasons when the test results cannot get the target uncertainty.Use UEI to assess the changes of uncertainty in different phase is more sensitive to changes of the test results′accuracy and its usefulness needs to be confirmed in practice.
展开▼
