Summary
Methods and results from the quality assurance program of the Atherosclerosis Risk
in Communities (ARIC) Study regarding hemostasis variables are presented, following
up previous reports in this journal on standardized procedures for blood collection
and processing (7) and an organized plan for the performance of those procedures (8).
Efforts were made to control for and assess all sources of variability, from venipuncture
to laboratory analysis, including also local field center processing and sample shipping.
The quality control program included (a) a standardized protocol for blood collection
and processing; (b) training, certification, and annual recertification of field center
personnel for blood collection and processing; (c) monitoring of fasting times, phlebotomy
times, processing times, and shipping problems; (d) hemostatic laboratory internal
quality control; (e) a replicate blood sample program; (f) an intraindividual variability
study; and (g) continual monitoring of quality control and study participants’ data.
This paper focused on items (c), (d), and (e). Measures of Variation, generally Standard
deviations and coefficients of Variation, are estimated for replicate blood sampling
and internal quality control data, for activated partial thromboplastin time, fibrinogen,
factor VII and VIII activity, von Willebrand factor, antithrombin-III, and protein
C. The results demonstrate that it is possible to reliably measure these hemostatic
variables in a large multicenter study.
