8 |
Klinisk Biokemi i Norden · 1 2017
Validation and verification in clinical chemistry
Elvar Theodorsson, Department of Clinical Chemistry and Department of Clinical and
Experimental Medicine, Linköping University, Linköping, Sweden
Abstract
Measurement systems used in
clinical chemistry are usually
manufactured by international
companies working according to
in vitro directive (IVD) – regu-
lations. This means that medical
laboratories do not need to vali-
date commercially available measurement systems,
but are obliged to verify that their properties found
during the producers validation can be reproduced
in the users laboratories. Developments in healthcare
and accreditation standards including ISO 15189
focus increasingly on customer needs which may
call for the validation of entire conglomerates of
laboratories and measurement systems catering for
the needs for the same patient population. Such vali-
dation practices are, however, still in their infancy.
The purpose of the present paper is to offer prac-
tical details on the common practice of verification
of single measurement systems and add perspectives
on the validation of conglomerates of laboratories
and measurement systems. Method validation across
conglomerates of laboratories using verified com-
mercially available measurement systems can only be
performed by the laboratories – users themselves in
their own circumstances. The use of patient samples
in split- sample techniques is especially valuable in
this contexts in order to avoid compatibility issues.
Introduction
Validation and verification of measurement met-
hods are procedures aiming at establishing realistic
expectations with the analyst and confidence with
the end-user that the methods are fit for the intended
purposes (1, 2).
Measurement systems already validated by their
manufacturers need only to be
verified
by the end
users in their own circumstances. Verification means
that the users study the bias and imprecision of the
measurement system and compare their results with
the ones observed by the manufacturers during their
validation (1, 3-11).
The basic principles and procedures of validation
are shared amongst all fields of bioanalysis with
some differences in emphasis depending on the par-
ticular fields of work (12-14). In the early 1990s the
U.S. Food and Drug Administration (FDA) initiated
and supported conferences and harmonization work
on bioanalytical method validation which in 2001
resulted in the guidelines “FDA Guidance for Indu-
stry – Bioanalytical Method Validation” (13, 15). This
document is widely used as standard reference for
validation of bioanalytical measurement methods in
addition to the European EMA Guidelines on Valida-
tion of Bioanalytical Methods (16) and the Eurachem
guide for method validation (17).
There are theoretically no limits to the extent of
validation and verification procedures, but also sub-
stantial time- and economic constraints in practice.
It is therefore crucial that verification and validation
efforts are optimized in order to minimize their cost/
benefit ratio.
Method verification
Single laboratory method validation is not needed
when a measurement system is manufactured by
a company or other responsible source which has
performed proper method validation (18, 19) and
is providing you with the detailed results of that
validation. A study of the bias and imprecision of
the new measurement system is needed, establis-
hing whether you are able to reproduce in your own
laboratory the data found by the company during
method validation.
The most common instance of method verification
is when old measurement systems are replaced by
new. This is the example dealth with here.
