STATISTICAL INFERENCE FOR THE WITHIN-DEVICEPRECISION OF QUANTITATIVE MEASUREMENTSIN ASSAY VALIDATION

摘要

Intermediate precision is one of the most important characteristics for evaluation ofprecision in assay validation. The current methods for evaluation of within-deviceprecision recommended by the Clinical Laboratory Standard Institute (CLSI) guidelineEP5-A2 are based on the point estimator. On the other hand, in addition to pointestimators, confidence intervals can provide a range for the within-device precisionwith a probability statement. Therefore, we suggest a confidence interval approachfor assessment of the within-device precision. Furthermore, under the two-stage nestedrandom-effects model recommended by the approved CLSI guideline EP5-A2, inaddition to the current Satterthwaite's approximation and the modified large sample(MLS) methods, we apply the technique of generalized pivotal quantities (GPQ)to derive the confidence interval for the within-device precision. The data from theapproved CLSI guideline EP5-A2 illustrate the applications of the confidence intervalapproach and comparison of results between the three methods. Results of a simulationstudy on the coverage probability and expected length of the three methods arereported. The proposed method of the GPQ-based confidence intervals is also extendedto consider the between-laboratories variation for precision assessment.