An Introduction to Quality Assurance in Analytical Science

1
An Introduction to Quality Assurance in
Analytical Science

• Dr Irene Mueller-Harvey
• Mr Richard Baker
• Mr Brian Woodget

2
Part 4 - Regulation and Accreditation

• Contents
• Accreditation (slide 3,4)
• ISO 17025 (slides 5-7)
• Inter-laboratory proficiency testing (slides
  8-11)
• Evaluating analytical quality (slides 12-13)

The presentation contains some animation which
will be activated automatically (no more than a
2 second delay), by mouse click or by use of the
page down key on your keyboard.
3
Accreditation
Accreditation Certificate

• Accreditation is defined as
• Formal recognition that
• a testing laboratory is competent to carry out
  specific tests

4
Accreditation which standard?
A number of universally recognised Accreditation
systems
ISO 9000 applicable mostly to manufacturing
services - emphasis on records and customer
service
Good laboratory (GLP) and good manufacturing
practice (GMP) recognised for pharmaceutical and
clinical measurements - emphasis on records,
procedures, reproducibility
ISO 17025 used by analytical laboratories for
work outside the health sector - emphasis on
valid data, traceability, comparisons and data
recognition in a court of law
5
ISO 17025 Accreditation
Main features of accreditation

• Guarantee to customers
• work to agreed standards
• Laboratory
• independent stringent assessment
• Agreed specified methods used
• All measurements traceable to
• national international standards

6
ISO 17025 Accreditation
How do you become accredited?
1 Install a quality system in the laboratory that
complies with the ISO 17025
requirements 2 Decide on the schedule of tests
that are to be accredited 3 Apply to
the United Kingdom Accreditation
Service (UKAS) 4 Have a Pre-assessment visit
from the UKAS assessors 5 Have a formal
Assessment visit from UKAS 6 Clear any
non-compliances they find 7 Become Accredited
for the chosen schedule of tests
7
ISO 17025 Accreditation
How do you remain accredited?

• pay an annual subscription
• perform satisfactorily in inter-laboratory
  proficiency tests
• undergo and pass regular horizontal and
  vertical audits

Mostly concerned with the overall
quality management system
Will trace sample paths from initial acceptance
to final report
8
Inter-laboratory Proficiency Tests
Many laboratory tests are used for international
or national standard and regulations, for example

• food labelling,
• water quality,
• animal feeds,
• pesticide residues.

The authorised laboratories must show that their
results are comparable with others doing the same
tests.
9
Proficiency Schemes (1)
Participation in proficiency testing schemes
offer the following benefits

• Show labs how well they compare with others
• Help them to reduce the overall variability in
  testing

• Give the regulatory authorities and consumers
  confidence that their quality criteria are
  meaningful

10
Proficiency testing (2)
Organisers of such schemes must ensure that

• all participating labs receive identical samples
• a true value is assigned for the result of a
  test (for example it may be the mean of all
  participants results)
• participants do not know the true result before
  they do the test.

11
Proficiency testing (3)
From the results submitted by the participating
laboratories, the organisers can

• rank the laboratorys performance by calculating
  its z-score
• identify the laboratories that can produce
  acceptable results for the test
• show divergent laboratories how they need to
  improve their performance of the test.

12
Calculation of z scores
A laboratorys z-score is calculated from
(x - X)/SD x the lab result X the
true or accepted result SD target
value for the standard deviation
The graph above shows the z scores obtained by
120 laboratories who participated in a particular
proficiency testing scheme
The best performing laboratories have z values
close to zero. The acceptable range is 2 to -2
13
Variation of CV with analyte level
We can expect a CV to be approx. 5 at a 1g/l
level and approx. 15 at a 0.001g/l level
60
50

40
30
Coefficient of variation (CV) or relative
standard deviation (RSD)

20

• 10

0

-10
-20

-30
-40
-50
-60
10
0.1
1 ppm
1 ppb
10-1
10-3
10-6
10-9

• Concentration