Application Series 18
TESTING THE AIR PURITY TESTERS
Patrick Brown and Peter Ritchie of the Institute of Occupational Medicine, Edinburgh,
describe the importance for health and safety of proficiency testing for mineral fibre
measurement and the use of SIR in the Regular Interlaboratory Counting Exchanges (RICE)
scheme.
Background
Asbestos, along with other mineral fibres, invokes an emotive image of fear and danger to
health. But use of these fibres in industry and construction is widespread - and there are
risks of exposure during their clearance and disposal from old installations. Exposure levels
to these airborne fibres are strictly monitored in order to minimise the risks workers in these
industries undertake. For this reason, we must make sure that the measurements of
exposure levels are accurate.
'It is a legal requirement that laboratories making these measurements operate good quality
assurance procedures, including proficiency testing.' explains Patrick Brown, who is
responsible for the administration of the scheme at the Institute of Occupational Medicine
(IOM) in Edinburgh. 'Quality assurance is especially important in subjective measurements
such as these'. With the measurement of airborne fibres, trained microscopists have to count
the number of fibres on a given area of a collecting filter to provide the measurement. Large
variations in results can occur 'So laboratories need to participate in a proficiency testing
scheme to minimise this'.
The RICE scheme
The oldest and largest of such schemes run at the IOM is the Regular Interlaboratory
Counting Exchanges (RICE) scheme. This scheme administers proficiency testing of
laboratories in the UK which measure airborne asbestos fibres. RICE has been operated
since 1984 by the IOM on behalf of the Health and Safety Executive's Committee on Fibre
Measurement. The number of participating laboratories is increasing all the time. 'In the first
round of exchanges there were 175 laboratories' Patrick said 'now there are around 240'.
There are several schemes throughout the world for testing the proficiency of laboratories in
making analytical determinations. 'All of these, including RICE, involve sending reference
samples to laboratories, comparing each laboratory's results to known "true" values for those
samples and measuring the difference' Patrick stated.
How the scheme operates
In the RICE scheme reference samples are counted regularly by each participating
laboratory. Performance is assessed by comparing laboratory counts with reference counts.
'These come from previous counts made by satisfactory RICE participants' explained Patrick.
'The reference samples are subdivided into batches, laboratories are formed into small
groups and RICE is organised in rounds, which take place every 3-4 months. Rounds consist
of the circulation of batches to the groups.' The results are sent to the IOM where laboratory
counts are compared with the reference counts. The assessment is based on the results
averaged over four consecutive rounds. After every round, each laboratory receives
notification of its assessment, and an updated list of participants and their assessments is
published.
The role of SIR
'It is absolutely essential that we maintain the smooth running of this complex scheme as well
as ensure the accuracy and reliability of the data and its analysis' explained Peter Ritchie, a
database manager at the IOM. SIR plays a large part in making this possible. The IOM has
been using SIR since 1984 to store the data from the scheme; to help the administration of
the scheme; to produce analyses, reports, forms and labels; to output to other packages for
further analyses or graphical work. Peter described the origins of SIR in IOM. 'In our
epidemiological work we do a lot of hierarchical case-based studies and SIR was principally
acquired with this in mind'. However, the operational requirements of the RICE scheme
demanded the implementation of a complex network, rather than the hierarchical, database
model. 'Due to its flexibility SIR was capable of the implementation and we were able to
utilise and develop existing and new SIR skills for the scheme rather than having to start with
a new package'.
A very simplified outline of the structure and function of the database is illustrated in the
diagram below:
The four main entities of interest are Rounds, Labs, Samples and Measurements and these
are interlinked. 'A Round is the administrative means by which a Lab receives a Sample. A
Measurement can subsequently contribute towards assessing a Lab's performance or
towards generating a reference value for a new Sample' said Peter, describing the structure
of the database. 'Much use is made of SIR FORMS to enter data, the accuracy of which can
be crucial to a laboratory's assessment and hence its commercial options'. PQL and utilities
are also used for further checks, to initiate and manage rounds and to output data for further
detailed statistical analysis.
RICE is currently run under SIR 2.2 on a PRIME minicomputer. This has proved to be a very
reliable system that depends on the skills and commitment of the staff concerned as much as
the computer system itself. Changes to the scheme over the last few years have been
enabled by the use of SIR as the DBMS. Two main recent changes were alterations to
reflect improvements in the reference counts and in the range of sample types; and
enhancements to enable more sophisticated data analysis. 'Over the next year we will be
transferring the scheme to a PC based system and we want to take advantage of the new
PQL and interface features available, especially for the current menu based utilities' said
Peter, looking forward to the new challenge.
Further Developments
The IOM also operates a smaller international proficiency testing scheme for asbestos fibre
measurement and two similar schemes (one international, one for UK) for the man-made
mineral fibres. Each scheme uses a separate SIR database employing a model of design
and operation similar to RICE. 'In the longer term we are involved in a World Health
Organisation programme to harmonise the measurement methodology and results for
different fibre types via a single international proficiency testing scheme' Peter mentioned.
The organisational and computing methods developed at the IOM can be employed for other
proficiency schemes too. 'We hope this will in due course lead to further tests of the
flexibility of the SIR database'.
For more information on the use of SIR at the IOM contact:
Peter Ritchie
Institute of Occupational Medicine
8 Roxburgh Place
Edinburgh EH8 9SU UK
E-mail: P_Ritchie@iomhq.org.uk
Back Sir Home