Agriculture Reference
In-Depth Information
non-satisfactory outcomes to undertake some further
investigation of the method and to demonstrate the
implementation of corrective action.
Unfortunately, there are relatively few proficiency
testing schemes available to scientists working in the
field of veterinary drug residues. Some such schemes,
such as those organised by the EU CRL network, are
frequently available to EU member states only. Other
schemes, however, are open to all participants.
applied and the concentration detected in each sample
calculated. From this, the mean concentration, the
standard deviation and the coefficient of variation (%) of
the fortified samples should be calculated. These data
may be used to calculate CCα and CCβ.
Proficiency testing
The use of proficiency tests, where different laborato-
ries test the same materials using their own methods
(providing that those methods are normally run under
routine conditions), is a method of assessing laboratory
reproducibility. Evaluation of laboratory performance is
usually made by calculating a consensus value for the
test material, based on the analytical results obtained by
the participants. The calculation of a robust mean,
following the elimination of outliers, is the approach
most commonly adopted. This approach has some draw-
backs - most notably that the consensus mean may be
biased when poorly controlled methods are used by a
large number of participants, leading to a wide range in
analytical results. Numerical assessment of an individual
laboratory's performance is usual effected with a
z
-score.
This is essentially a measure of the deviation of the
individual result from the consensus value. Underlying
the use of the
z
-score is an assumption that individual
z
-scores will approximate to a normal distribution, with
a mean of zero and a standard deviation of 1. Usually,
z
-scores are interpreted as follows:
z
< 2, satisfactory;
2 ≤
z
≥ 3, questionable; and
z
> 3, unsatisfactory. The
numerical values can be either positive or negative, indi-
cating that the measurement was greater or less, respec-
tively, than the consensus mean. The main value of a
proficiency test is to highlight possible problems with an
analytical method and to spur the participants with
References
Commission Decision 98/179/EC (1998)
Official Journal of the
European Communities
, L65, 31.
Commission Decision 2002/657/EC (2002)
Official Journal of the
European Communities
, L221, 8.
Commission Decision 2005/34/EC (2005)
Official Journal of the
European Communities
, L16, 61.
Commission Regulation (EU) No 37/2010 (2010)
Official Journal of
the European Communities
, L15, 1.
Commission Regulation (EC) No 136/2004 (2004)
Official Journal of
the European Communities
, L21, 11.
Council Directive 91/496/EEC (1991)
Official Journal of the European
Communities
, L268, 56.
Council Directive 96/23/EC (1996)
Official Journal of the European
Communities
, L125, 10.
Council Directive 97/78/EC (1998)
Official Journal of the European
Communities
, L24, 9.
Directive 2003/74/EC of the European Parliament and of the Council
(2005)
Official Journal of the European Communities
, L262, 17.
Directive 2008/97/EC of the European Parliament and of the Council
(2008)
Official Journal of the European Communities
, L318, 9.
Regulation (EC) No 178/2002 of the European Parliament and of the
Council (2002)
Official Journal of the European Communities
, L31, 1.
Regulation (EC) No 470/2009 of the European Parliament and of the
Council (2009)
Official Journal of the European Communities
, L152, 11.
Regulation (EC) No 1831/2003 of the European Parliament and of the
Council (2003)
Official Journal of the European Communities
, L268, 29.