Online Program Home
My Program

Abstract Details

Activity Number: 358 - Contributed Poster Presentations: Section on Statistics in Epidemiology
Type: Contributed
Date/Time: Tuesday, July 30, 2019 : 10:30 AM to 12:20 PM
Sponsor: Section on Statistics in Epidemiology
Abstract #301790
Title: Application of Variance Analysis to Proficiency Test Data from CDC Lead and Multi-Element Proficiency (LAMP) Program
Author(s): Po-Yung Cheng* and Kathryn Vance and Cynthia Ward and Robert Jones and Amir Makhmudov
Companies: CDC and CDC and CDC and CDC and CDC
Keywords: variance analysis; proficiency test; z-score
Abstract:

LAMP is a voluntary performance and quality assurance program designed to promote high quality of whole blood lead, cadmium, mercury, selenium, and manganese measurements. We applied the variance analysis to LAMP data (blood cadmium (BCd)) and used the obtained standard errors and their relevant target values of proficiency test samples to determine a linear regression between them. Based on this linear regression, we calculated the z-scores and used these scores to determine the performance for the LAMP participating laboratories. There are two methods for BCd measurement: ICP-MS and GFAAS. For ICP-MS, R2 for fitting a quadratic equation of the standard errors on the target values was 0.975. For GFAAS, R2 for its quadratic equation was 0.972. The variance for ICP-MS was smaller than the one for GFAAS (median: 0.1635 vs. 0.1949, respectively; range: 0.0197-3.2369 vs. 0.0549-14.3742, respectively). We used the standard errors from the quadratic equation of ICP-MS to calculate the z-scores for all measurement values for LAMP data. For ICP-MS, 6.6% of z-scores were out of the acceptable range (within ±3). In contrast, 11.4% of z-scores were out of the acceptable range for GFAAS.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program