Thermo Scientific Phadia 100, User Manual

Page: 182 / 357

Phadia 100 - User Manual Quality Assurance
12-3501-20/12 Page 8.7
8
Quality control of other allergy,
autoimmunity and asthma methods
Controls of preferably different concentration levels are used. Sets of control
samples are included in each run.
Minimum 20 results per control sample are collected from different runs and the
mean values ±2 SD and ± 3 SD are calculated and used in the control charts.
Preferable way to run QC
Use the minimum- and maximum-range values given in the Directions for Use
for the Quality Control, or for EliA on the package certificate. Run a minimum
of 20 assay runs to get the mean for the expected value and expected standard
deviation in the Quality Control log. Now you can use these obtained values for
expected concentration and standard deviation. It is recommended to use 3 as
the standard deviation limit.
The mean value and SD should be recalculated with regular intervals similar to
ImmunoCAP Specific IgE Control.
The obtained mean value for each control sample of ImmunoCAP Gliadin IgA/
IgG Control LMH and ImmunoCAP Specific IgG Control LMH should fall
within the range given in the package insert.
For EliA ANA Control, EliA ANCA/GBM Control, EliA CCP Control, EliA APS
Control and EliA Celiac Control the ranges are given on the package certificate.
For ImmunoCAP Total IgE Control LMH, ImmunoCAP ECP Control,
ImmunoCAP Tryptase Control, ImmunoCAP TG IgG Antibodies Control,
ImmunoCAP TPO IgG Antibodies Control, ImmunoCAP Specific IgE f1
Control, ImmunoCAP Specific IgG/IgG4 i1 Control H, ImmunoCAP Specific
IgG/IgG4 Control L, ImmunoCAP Gliadin IgA Control H, ImmunoCAP Gliadin
IgA/IgG Control L and ImmunoCAP Gliadin IgG Control H the ranges are given
on the vial label.
Control Charts
A control chart is used to detect deviations from “normal” performance.
Consequently there is a need for rules which, if transgressed, inform the
operator that the performance is out of control. The rules are often in the
form of control limits inserted in the control charts.To discover a systematic
error, a control rule should be based on a number of consecutive runs. For
changes occurring occasionally, only the actual run needs to be considered.