J Clin Pathol Qual Control.  2000 Dec;22(2):293-301.

External Quality Assessment in anti-HIV Enzyme Immunoassay in Korea

Affiliations
  • 1Department of Virology, National Institute of Health, Korea. aidshiv@nuri.net
  • 2Department of Clinical Pathology, Yonsei University College of medicine, Seoul, Korea.
  • 3Blood Transfusion Research Institute, Korea National Red Cross, Seoul, Korea.

Abstract

BACKGROUND: We performed the quality assessment of HIV antibody test since 1998 to improve the quality of laboratory testing and to identify problems in laboratory testing processes. This report is the result from May 1999 to April 2000.
METHODS
The quality assessment panels for HIV antibody testing were composed of undiluted 5 HIV antibody positive sera and 5 negative sera for 3rd and 4th panel. The panels were distributed to 260 participating laboratories of hospital, blood bank, public health. Among the sera, one positive serum showing low titer obtained from blood donor of early HIV infection and duplicated negative sera were included to challenge the high accuracy and to evaluate the reproducibility of HIV testing respectively.
RESULTS
Enzyme immunoassay (EIA) method was used by 230 out of 260 laboratories for 3rd panel and 244 out of 251 laboratories for the 4th panel. The analytic sensitivity and specificity of EIA HIV1/2 antibody test were 98.55% and 99.78% in 3rd trial and 100% and 98.98% in 4th trial. Analytic accuracies in EIA were similar those of previous year. The data of each sample were analyzed by traditional (mean+/-2 standard deviation) and robust statistics to find outliers. Robust statistics detected more outlier. Youden analysis for duplicate negative sample was represented that 22 out of 75 laboratories using B assay were positioned outside the 99% ellipse confidence showing 10 laboratories of random error and 12 laboratories of systemic error.
CONCLUSIONS
Robust statistics about single and duplicate negative sample seems to be more efficient for the identification of some results showing significant variation from the median than traditional statistics by mean and standard deviation. Further studies are needed for analysis of the quality assessment results including the advanced analysing method.


MeSH Terms

Blood Banks
Blood Donors
HIV
HIV Infections
Humans
Immunoenzyme Techniques*
Korea*
Public Health
Sensitivity and Specificity
Full Text Links
  • JCPQC
Actions
Cited
CITED
export Copy
Close
Share
  • Twitter
  • Facebook
Similar articles
Copyright © 2024 by Korean Association of Medical Journal Editors. All rights reserved.     E-mail: koreamed@kamje.or.kr