biology lab guidelines

8/8/2019 Biology Lab Guidelines

1/36

NABLNATIONAL ACCREDITATION

BOARD FOR TESTING ANDCALIBRATION LABORATORIES

NABL 162

GUIDELINES FOR

PROFICIENCY TESTINGPROGRAMfor TESTING AND

CALIBRATIONLABORATORIES

ISSUE NO : 04 AMENDMENT NO : 00ISSUE DATE: 24.03.2008 AMENDMENT DATE: --


2/36

National Accreditation Board for Testing and Calibration LaboratoriesDoc. No: NABL 162 Guidelines for Proficiency Testing Program for Testing and Calibration Laboratories

Issue No: 04 Issue Date: 24.03.2008 Last Amend No: 00 Amend Date: -- Page No: i

AMENDMENT SHEET

Sl PageNo.

ClauseNo.

Date ofAmendment

Amendment made Reasons SignatureQM

SignatureDirector

1

2

3

4

5

6

7

8

9

10


3/36


Issue No: 04 Issue Date: 24.03.2008 Last Amend No: 00 Amend Date: -- Page No: 1/ 31

Contents

Part-1: Development and Operation of Proficiency Testing Programs

SL. SECTION PAGE NO.

1. Introduction 2

2. Definitions 3-5

3. Scope 6

4. Types of Proficiency Testing 7-9

4.1 Interlaboratory Testing 74.2 Split Sample Testing 7

4.3 Measurement Comparison 8

4.4 Qualitative Schemes 9

4.5 Known Value Schemes 9

4.6 Partial Process Schemes 9

5. Organisation and Design 10-12

5.1 Frame Work 10

5.2 Staff 10

5.3 Test and Data processing equipments 11

5.4 Statistical Design 11

5.5 Test Item Preparation 11

5.6 Test Item Management 12

5.7 Choice of Method/Procedure 12

5.8 Evolution of Proficiency Testing Programs 12

6. Operation and Reporting 13-15

6.1 Coordination & Documentation 13

6.2 Instructions 13

6.3 Packaging and Transportation 13


4/36



SL. SECTION PAGE NO.

6.4 Testing/Calibration 14

6.5 Data Analysis & Records 14

6.6 Program Reports 15

7. Evaluation of Proficiency Test Results 16

8. Communication with Participants 17

9. Confidentiality/Ethical Considerations 18

9.1 Confidentiality 18

9.2 Collusion and Falsification of results 18

Annexure A 19-25

Examples of statistical methods for treatment of proficiency test data

A.1 Determination of the assigned value and its uncertainty 19

A.2 Calculation of performance statistics 21

A.3 Evaluation of performance 23

A.4 Preliminary determination of test item homogeneity 25

Annexure B 26-27

Quality Management of Proficiency Testing Programs

Part-2:

Selection and use of proficiency testing programs bylaboratory accreditation bodies.

1. Scope 29

2. Selection of Proficiency Testing Programs 29

3. Policies on Participation in Proficiency Testing Programs 30

4. Use of Results by NABL 30

5. Action and Feedback by Laboratories 31


5/36


6/36



1. Introduction________________________________

1. Proficiency Testing through Interlaboratory comparisons may be used, for example, to:

a) determine the performance of individual laboratories for specific tests ormeasurements and to monitor laboratories continuing performance;

b) identify problems in laboratories and initiate remedial actions which may be related

to, for example, individual staff performance or calibration of instrumentation;c) establish the effectiveness and comparability of new test or measurement methods

and similarly to monitor established methods;

d) provide additional confidence to laboratory clients;

e) identify interlaboratory differences;

f) determine the performance characteristics of a method often known ascollaborative trials;

g) assign values to reference materials (RMs) and assess their suitability for use inspecific test or measurement procedures.

2. Testing and Calibration Laboratories that wish to obtain and maintain NABL

accreditation must comply with the requirements as laid down by NABL.

3. Currently accreditation is granted in fields of: testing (biological, chemical,electrical, electronics, fluid-flow, forensic, mechanical, non-destructive,optical/photometry, radiological & thermal); calibration (electrotechnical, fluid-flow,mechanical, radiological, and thermal & optical measurements) and medical(clinical biochemistry, clinical pathology, haematology & immunohaematology,microbiology & serology, histopathology, cytopathology, genetics, nuclearmedicine-invitro test only).

4. This publication sets out the criteria NABL uses for the development & operationand selection of proficiency testing program formulated by NABL for itsapplicant/accredited testing and calibration laboratories.

5. This publication would regularly be reviewed and the changes may have to beincorporated because of:

Any modification/changes of ISO/IEC Guide 43-1 & Guide 43-2 (ProficiencyTesting by interlaboratory comparisons)

The deliberation and decisions of ILAC. The deliberation and decisions of APLAC. The feed back from accredited laboratories. The feed back from assessors/experts. The decisions of the board and various committees of NABL.


7/36



2. Definitions__________________________________

2.1 Test Technical operation that consists of the determination of one or morecharacteristics of a given product, process or service according to specifiedprocedure.

2.2 Calibration The set of operations which establish, under specified

conditions, the relationship between values indicated by a measuringinstrument or system or values of a measure.

2.3 Test Method Specified technical procedure for performing a test.

2.4 Test result The value of a characteristic, obtained by completely carrying outa specified measurement method.

2.5 Test item Material or artifact presented to the participating laboratory for thepurpose of proficiency testing.

2.6 Testing/Calibration Laboratory A party seeking or holding NABLaccreditation for testing and calibration. This party may be an individual, anorganisation or a part of an organsiation.

2.7 Laboratory accreditation A formal recognition that a testing/calibrationlaboratory is competent to carry out specific test or specific types of tests ormeasurements.

2.8 Laboratory assessment Examination of a testing/calibration to evaluate itscompliance with specific laboratory accreditation criteria.

2.9 Accreditation body A Governmental or non-governmental body whichconducts and administers a laboratory accreditation system and grantsaccreditation.

2.10 Accredited laboratory A laboratory to which accreditation has been granted.

2.11 Accreditation criteria A set of requirements used by an accreditation body

which a testing/calibration laboratory must meet to be accredited.

2.12 Assessor An expert who carries out some or all functions related tolaboratory assessment.

2.13 Accredited laboratory test report A test report which includes a statementby the testing laboratory that it is accredited for the test reported and that thetest has been performed in accordance with the conditions prescribed by theaccreditation body.

2.14 Reference material (RM) A material or substance one or more properties ofwhich are sufficiently well established to be used for the calibration of anapparatus, the assessment of the measurement method, or for assigning thevalues to materials.


8/36



2.15 Certified Reference Material (CRM) Reference material (RM), accompaniedby a certificate, one or more of whose property values are certified by aprocedure which establishes its traceability to an accurate realisation of the unitin which the property values are expressed, and for which each certified valueis accompanied by an uncertainty at a stated level of confidence.

2.16 Interlaboratory comparison Organisation, performance and evaluation of

testing/calibration (competence) on the same or similar items by two or morelaboratories in accordance with pre-determined conditions.

2.17 Proficiency Testing methods of checking laboratory testing/calibrationperformances by means of interlaboratory comparison.

2.18 Reference Laboratory Laboratory that provides reference values on a testitem with a known uncertainty (usually a National Calibration Laboratory).

2.19 Accepted Reference Value

a) A theoretical or established value based on scientific principles

b) An assigned value, based on experimental work of some national orinternational organisation.

c) A consensus value, based on collaborative experimental work under theauspices of a scientific or engineering group.

2.20 Assigned Value Estimate of true value used in the assessment ofproficiency.

2.21 Co-ordinator Person or body which co-ordinates all the activities associatedwith a proficiency testing program.

2.22 Accuracy The closeness of agreement between test/calibration results andthe accepted reference value.

2.23 Trueness The closeness of agreement between the average value obtainedfrom large series of test/calibration results and an accepted reference value.

2.24 Bias The difference between the test results and an accepted referencevalue.

2.25 Precision The closeness of agreement between independent test/calibrationresults obtained under prescribed conditions.

2.26 Traceability Property of the result of a measurement or the value of astandard whereby it can be related to stated references, usually national orinternational standards, through an unbroken chain of comparisons all havingstated uncertainties.

2.27 Nodal laboratory/body The approved laboratory/organisation by NABLwhich may be responsible for organising specific type of proficiency testing

programs on testing/calibration.


9/36



2.28 Laboratory Co-ordinator The individual who will be responsible to organisethe proficiency testing program from the approved nodal laboratory/organisation.

2.29 Outlier Member of a set of values which is inconsistent with the othermembers of that set.

2.30 Extreme results Outliers and other values which are grossly inconsistent

with other members of the data set.

2.31 Robust statistical techniques Techniques to minimize the influence thatextreme results can have on estimates of the mean and standard deviation.

2.32 Uncertainty of measurement Parameter associated with the results of ameasurement, that characterizes the dispersion of the values that couldreasonably be attributed to the measurand.


10/36



3. Scope__________________________________

3.1 The aim of the publication is to set out the criteria for the operation ofproficiency testing of NABL accredited laboratories.

3.2 This publication consists of two parts viz. Part-I Development and operation ofproficiency testing programs and Part-II Selection and use of proficiencytesting programs by laboratory accreditation bodies.

3.3 Additional criteria may be specified by NABL depending upon the specifictesting/calibration that is to be evaluated.

3.4 This guide describes how NABL the accrediting body should select andoperate proficiency testing program.


11/36



4. Types of Proficiency Testing___________________

Proficiency testing techniques vary depending on the nature of the item or materialunder test, test method in use and the number of testing laboratories participating. Mostpossess the common feature of comparison of test results obtained by one testing laboratorywith those obtained by one or more other testing laboratories. In some programs, one of theparticipating laboratories may be a controlling, coordinating, or reference function. The

following are the major types of proficiency testing programs.

4.1 Interlaboratory Testing

In this program participating laboratories are provided with sub-samples from asource of a suitable degree of homogeneity which they are expected to test atcomparable levels of competence. Feature of such programs usually are:

a) It is essential that the (sub) samples provided to each participant aresufficiently homogenous so that any results later identified as extremeshould not be attributed to any significant sample variability.

b) After completion of the testing the results are returned to the coordinating

body and compared with the assigned values to give an indication of theperformance of the individual laboratories and the group as a whole.

c) This is the type commonly used by accrediting bodies, regulatory bodiesand other organisations, when they utilize schemes in the testing fields.

d) Examples of test items used in this type of proficiency testing include food,body fluid, soil, water and other environmental materials. In some casesseparate portions of previously established (certified) reference materialsare circulated.

e) Occasionally this technique is also used for interlaboratory measurementschemes.

4.2 Split Sample Testing

This program involves samples of a product or a material being divided, intotwo or more parts with each participating laboratory testing one part of eachsample

a) This program differs from Inter-laboratory proficiency testing as there isusually limited control of, or preliminary data on, the homogeneity of thesample being divided.

b) This technique is sometimes used by clients of laboratory services,including regulatory authorities.

c) Such programs often need retention of sufficient material to resolve anyperceived differences between the limited number of laboratories involvedby further analysis by additional laboratories.


12/36



d) Similar intercomparisons are regularly conducted in commercialtransactions when samples representing a traded commodity are splitbetween a laboratory representing the supplier and another laboratoryrepresenting the purchaser.

e) A similar technique of split-sample testing is also used in the monitoring ofclinical and environmental laboratories. Typically, these schemes involve inthe results from several split samples over a wide concentration interval

being compared between an individual laboratory and one or more otherlaboratories. Under such schemes, one of the laboratories may beconsidered to operate at a higher metrological level (i.e, lower level ofuncertainty) due to the use of reference methodology and more advancedequipment, etc. Its results are considered to be the reference values insuch intercomparisons and it may act as an advisory or mentor laboratoryto the other laboratories comparing split-sample data with it.

4.3 Measurement Comparison

Measurement comparison schemes involve the test item to be measured orcalibrated being circulated successively from one participating laboratory to the

next. Features of such schemes usually are as follows:a) Assigned values for the test item are provided by a Reference Laboratory,

which might be a countrys highest authority for the measurementconcerned. It may be necessary for the test item to be checked at specificstages during the conduct of the proficiency test. This is to ensure thatthere are no significant changes in the assigned value throughout thecourse of the proficiency test.

b) Schemes involving sequential participation take time (in some casesyears) to complete. This causes a number of difficulties such as: ensuringthe stability of the item; the strict monitoring of its circulation and the timeallowed for measurement by individual participants; and the need to supply

feedback on individual performance to laboratories during the schemesimplementation, rather than waiting until it finishes. In addition, it may bedifficult to compare results on a group basis as there may be relatively fewlaboratories whose measurement capabilities closely match each other.

c) The individual measurement results are compared with the referencevalue established by the Reference Laboratory. The coordinator shouldtake into account the claimed measurement uncertainty of eachparticipating laboratory.

d) Examples of items (measurement artefacts) used in this type ofproficiency testing include reference standards (e.g. resistors, gauges andinstruments).


13/36



4.4 Qualitative Schemes

Evaluation of laboratory testing performance will not always involveinterlaboratory comparisons. For example, some schemes are designed toevaluate the capabilities of laboratories to characterize specific entities (e.g.type of asbestos, identity of a specific pathogenic organism, etc.)

Such schemes may involve the special preparation of test items with addition

of the subject component by the scheme coordinator. As such, the schemesare qualitative in nature, and do no need the involvement of multiplelaboratories or interlaboratory comparisons to evaluate a laboratorys testingperformance.

4.5 Known-Value Schemes

Other special types of proficiency testing schemes may involve the preparationof test items with known amounts of the measurand under test. It is thenpossible to evaluate the capability of an individual laboratory to test the itemand provide numerical results for comparison with the assigned value. Onceagain, such proficiency schemes do not need the involvement of multiple

laboratories.

4.6 Partial process Schemes

Special types of proficiency testing involve the evaluation of laboratoriesabilities to perform parts of the overall testing or measurement process. Forexample, some existing proficiency schemes evaluate laboratories abilities totransform and report a given set of data (rather than conduct the actual test ormeasurement) or to take and prepare samples or specimens in accordancewith a specification.


14/36



5. Organisation and Design______________________

5.1 Framework

Proper design for proficiency testing program is required to ensure its successand smooth operation. The coordinator in consultation with other expertsshould develop an appropriate plan for the particular proficiency test. The planshould include the following information:-

a) the name and address of the organisation conducting the proficiencyprogram;

b) the name and address of the coordinator and other personnel involved inplanning and operation;

c) the nature and the purpose of the program;

d) a procedure for the manner in which the participants are selected.e) the name and address of the laboratory or laboratories performing the

program, sampling, sample processing and list of potential participants.

f) the nature of the test item(s) and test(s) selected as well as shortdescription of the consideration underlying these choice.

g) a description of the manner in which the test items are obtained,processed, checked and transported;

h) a description of the information that is supplied to participants in thisnotification phase and of the time schedule for the various phases of theproficiency testing.

i) the expected initial and target dates or deadlines of the proficiency schemeincluding the date(s) for the testing to be carried out by the participants;

j) the basis of performance evaluation techniques;

k) information on methods or procedures which participants may need to useto perform the tests or measurements (commonly their routine procedures);

l) a description of the extent to which the test results, and the conclusions

that will be based on the outcome of the proficiency tests, are to be madepublic.

m) the expected initial and target dates or deadlines of the proficiency schemeincluding the date(s) for the testing to be carried out by the participants.

5.2 Staff

5.2.1 The staff involved in providing the program organised by NABL must haveadequate qualifications and experience to design implementation and reportingof interlaboratory tests.

5.2.2 There should be an advisory panel for various disciplines of testing and

calibration and functions of this panel may include;a) the development and review of procedures for the planning, execution,

analysis and report of the proficiency testing program;


15/36



b) the identification and evaluation of interlaboratory test comparisonorganised by bodies;

c) providing advice to a body assessing the technical competence of testingcalibration laboratories on the use of proficiency testing as an element of itslaboratory evaluation.

5.3 Test and Data-Processing equipments

The equipments and facilities must be adequate to conduct the tests. Thestorage and security of data file is also to be ensured. The use of computerbased system is recommended.

5.4 Statistical design

5.4.1 The statistical model and data analysis techniques to be used should bedocumented, together with a short description of the background to theirselection. Further details of common statistical procedures and treatment ofproficiency testing data are discussed in Annex-A.

5.4.2 Appropriate statistical design of a proficiency testing scheme is essential.Careful consideration should be given to the following matters and theirinteraction:a) the precision and trueness of the test(s) involved;b) the smallest differences to be detected between participating laboratories

at a desired confidence level;c) the number of participating laboratories;d) the number of samples to be tested and the number of repeat tests or

measurements to be carried out on each sample;e) the procedures to be used to estimate the assigned value;f) procedures to be used to identify outliers.

5.4.3 In the absence of reliable information concerning a) it may be necessary insome cases to organise a pilot interlaboratory comparison (collaborative trial)to obtain it.

5.5 Test Item Preparation

5.5.1 Preparation of test items may either be contracted out or undertaken by thecoordinator.

5.5.2 The test items or materials to be distributed in the program should generally besimilar in type to those routinely tested by the accredited laboratory.

5.5.3 The assigned value should not be disclosed to the participants until after the

results have been collated.


16/36



5.6 Test Item Management

5.6.1 Procedures for sampling, randomising, transporting, sorting and handling oftest items or materials to be documented.

5.6.2 The bulk material prepared for the proficiency test must be sufficiently

homogenous for each test parameter so that all laboratories will receive testsamples that do not differ significantly.

5.6.3 Coordinator should consider any hazards that the test materials might poseand appropriate to advise any party that might be at risk of the potential hazardinvolved.

5.6.4 Where possible the coordinator should also provide evidence that the testitems are sufficiently stable to ensure that they will not undergo any significantchange throughout the conduct of the proficiency test. When unstablemeasurands need to be assessed, it may be necessary for the coordinatingorganisation to specify a date by which the testing should be completed and

any required special pre-testing procedures.

5.7 Choice of Method/Procedure

5.7.1 Participants will normally be able to use the method of their choice, which isconsistent with routine procedures used in their laboratories. However, incertain circumstance, the coordinator may instruct participants to use aspecified method. Such methods are usually nationally or internationallyaccepted standard methods, and will have been validated by an appropriateprocedure (e.g. collaborative trial).

5.7.2 Where a calibration procedure is used, the assigned value will often be a

reference value obtained from measurements obtained by a high-echeloncalibration laboratory (often a National Standards Laboratory) which should usea well defined and accepted procedure.

5.7.3 Where participants are free to use a method of their own choice, coordinatorsshould, where appropriate, request details of the methods used to allow theuse of participants results to compare and comment on the methods.

5.8 Evolution of Proficiency Testing Programs

To ensure that proficiency testing programs are able to adapt to technical andscientific developments, they may need to include new types of samples ornew methods or procedures. Early conclusions from the results of suchprograms on the performance of individual laboratories should be drawn withdue care.


17/36



6. Operation and Reporting_____________________

6.1 Coordination & Documentation

The day-to-day operation of a program should be the responsibility of thecoordinator of the program. All practices and procedures must bedocumented. These may be included in or supplemented by a Quality Manual.

(see Annex-B).

6.2 Instructions

6.2.1 Chairman/Director, NABL would appoint the coordinator from NABL Secretariator any outside agency competent to conduct the proficiency testing.

6.2.2 Detailed instructions covering all aspects of the programs which must beadhered to by the participating laboratories should be provided by thecoordinator.

6.2.3 These will include details concerning factors, which may influence the testing/

calibration of the supplied proficiency testing items or materials. These factorsmay include operators, nature of items or materials, equipment status,selection of test/calibration procedure and timing of testing/calibration. Thematerials are to be prepared by an accredited laboratory with a code number.

6.2.4 NABL should organise proficiency testing in various disciplines. Alllaboratories holding NABL accreditation for those tests would be required toparticipate. At least one proficiency testing in major field of testing/calibrationwould be organised by NABL annually. These rounds of testings should beorganised at various locations of the country.

6.2.5 Proper notice (at least 6 weeks) well in advance would be issued by NABL/

Coordinating laboratory to all probable participating laboratories informing themregarding the proficiency test, name of the coordinator, his address, probabledate of supply of the samples and type of proficiency testing to be conducted.

6.2.6 Participating laboratories would be asked to submit the test report within fourweeks from the date of the receipt of the sample to the coordinator of theprogram.

6.3 Packaging and Transportation

The packaging has to be adequate and able to protect the stability andcharacteristics of the test items. There may be certain restrictions ontransportation such as dangerous goods, regulations or customs requirement.The laboratories themselves must also take responsibility for the transport ofthe items, particularly in sequential measurement comparison programs.NABL/coordinating laboratory should ensure proper packaging of the samples.


18/36



Appropriate customs declaration forms wherever applicable should becompleted by the coordinator to ensure those delays in custom clearance areminimized.

6.4 Testing/Calibration

6.4.1 The coordinator shall ensure that the participating laboratories have used the

standard process of testing/calibration for which the program has been fixed.

6.4.2 The laboratory shall be required to provide a list of equipments used in testing/calibration.

6.4.3 It is the responsibility of the participants to use calibrated equipment, followgood laboratory upkeep, to be diligent in following the method and to treat testspecimens/calibration artifacts in the normal laboratory priority. Specimen/artifacts identification must be maintained during all work.

6.4.4 The participants are responsible for ensuring that test results are submittedwithin due date prescribed by NABL. If a participant suspects that the test will

not be completed by the deadline, the laboratory should contact the programcoordinator of NABL for instructions on how to proceed.

6.5 Data Analysis & Records

6.5.1 The data received from the participating laboratories must be entered andanalysed and then reported back as soon as practicable. It is essential thatproper procedures are to be adopted to check the validity of data entry andtransfers and subsequent statistical analysis. Retention of data capturesheets, computer back-up files or print-outs, graphs etc. are to be kept for areasonable period of time i.e, at least for 2 years.

6.5.2 Data analysis should generate summary measures and performance statisticsand associated information consistent with the schemes statistical model andthe objectives of the scheme. The influence of extreme results on summarystatistics should be minimized by the use of outlier detection tests to identifyand then omit them or, preferably, by the use of robust statistics. Annex Acontains some broad suggestions for statistical evaluations.

6.5.3 Program coordinators should have documented criteria for dealing with testresults that may be inappropriate for proficiency testing evaluations. Forexample, it is recommended that for measurands for which the test materialhas been shown not to be sufficiently homogenous or stable for the purposesof a proficiency test, no grading or scoring should be given.


19/36



6.6 Program Reports

6.6.1 The content of program reports will vary depending on the purpose of aparticular program should be clear and comprehensive and include data on thedistribution of results from laboratories together with an indication of individualparticipants performance.

6.6.2 The following information should normally be included in reports of proficiencytesting programs.

Name and address of organisation providing the program. Names and affiliations of the persons involved in the design and conduct of

the program. Date of issue of report Report number and clear identification of program. Clear description of items or materials used including details of sample

preparation, homogeneity testing. Laboratory participation codes and test results. Statistical data and summaries including assigned values and range of

accepted results. Procedures used to establish any assigned value. Details of the traceability and uncertainty of any assigned value. Assigned values and summary statistics for test methods/procedures used

by other participating laboratories (if different methods are used by differentlaboratories).

Comments on laboratory performances by the coordinator and technicaladvisers.

Procedures used to design and implement the program (which may includereference to a scheme protocol).

Procedures used to statistically analyse the data (see AnnexA). Advice where appropriate, on the interpretation of the statistical analysis.

6.6.3 For schemes operated on a regular basis, it may be sufficient to have simplerreports such that many of the recommended elements in 6.6.2 could beexcluded from routine reports, but inclined in periodic summary reports and onrequest from participants.

6.6.4 Reports should be made available quickly within specified time-tables.Although, ideally, all original data supplied should be reported to participants, itmay not be possible to achieve this in some very extensive schemes.Participants should receive a summary sheet of their result from NABL/coordinating laboratory. In some programs such as long period measurementcomparison schemes, interim reports should be issued to individualparticipants.


20/36



7. Evaluation of Proficiency Test Results___________

7.1 Where an evaluation of performance is needed the coordinator should beresponsible for ensuring that the method of evaluation is appropriate tomaintain the credibility of the scheme.

7.2 The coordinator with the assistance of the technical advisors wherever

necessary may provide comments on performance with respect to :

Overall performance versus prior expectations (taking uncertainties intoaccount)

Variation within and between laboratories Variation between methods or procedures, if applicable Possible sources of error and suggestions for improving performances Any other suggestions, recommendations or general comments Conclusions

7.3 It may be necessary to provide individual sheets for participants after aparticular program and these may include updated summaries of performance

of individual laboratories over various rounds of an ongoing program. Suchsummaries can be further analysed and trends highlighted if required.

7.4 A variety of procedures and treatments (statistical methods) are available forproficiency testing program. Coordinator may choose most appropriate one forthe ongoing program.


21/36



8. Communication With Participants___________

8.1 Participants should be provided with a detailed set of information of joining aproficiency testing program. Subsequent communication with the participantscan be by a report. Participants should be advised immediately of anychanges in program design or operation.

8.2 Participants should be able to refer to the coordinator if they consider thatassessment of their performance in a proficiency testing is in error.

8.3 Feedback from laboratories are encouraged so that they can also contribute tothe development of a program.

8.4 Participating laboratories should be advised by NABL to maintain their ownrecords of performance in proficiency testing including the results ofinvestigation of any unsatisfactory results and subsequent corrective actionstaken by them.

8.5 For laboratories reporting unsatisfactory results, NABL should ask to

investigate and comment on its performance within three months. Wherever necessary, the laboratory should undertake any subsequent

proficiency test to confirm that corrective actions taken by the laboratoryare effective.

If required, NABL should depute expert to confirm that corrective actionsare effective.

8.6 NABL should ensure that the records of performance of the proficiency testingare maintained by the participating laboratories and are made available to theAssessment team whenever necessary.


22/36


23/36



Annex A(Informative)

Examples of statistical methods for treatment of proficiency test data

Proficiency test results can appear in many forms, spanning a wide range of data types andunderlying statistical distributions. The statistical techniques used to analyse the results need tobe appropriate for each situation, and so are too varied to specify.

There are, however, three steps common to all proficiency tests, when participants results areto be evaluated:

a) determination of the assigned value;b) calculation of performance statistics;c) evaluation of performance;and, in some cases,d) preliminary determination of test item homogeneity and stability.

This annex gives general criteria for statistical techniques that can be applied as needed toguide specific applications.

With new interlaboratory comparison schemes, agreement initially is often poor due to newquestions, new forms, artificial test items, poor agreement of methods, or variable laboratoryprocedures. Coordinators may have to use robust measures of relative performance (such aspercentiles) until agreement improves. Statistical techniques may need to be refined onceinterlaboratory agreement has improved and proficiency testing is well established.

This annex does not consider statistical techniques for analytical studies other than fortreatment of proficiency test data. Different techniques may be needed to implement the other

uses of interlaboratory comparison data listed in the Introduction.

NOTE ISO/TC 69 is currently preparing a document providing detailed information onstatistical methods contained in this annex.

A.1 Determination of the assigned value and its uncertainty

A.1.1 There are various procedures available for the establishment of assignedvalues. The most common procedures are listed below in an order that, inmost cases, will result in increasing uncertainty for the assigned value. Theseprocedures involve use of:


24/36



a) Known values with results determined by specific test item formulation(e.g. manufacture or dilution).

b) Certified reference values as determined by definitive methods (forquantitative tests).

c) Reference values as determined by analysis, measurement orcomparison of the test item alongside a reference material or standard,

traceable to a national or international standard.

d) Consensus values from expert laboratories expert laboratoriesshould have demonstrable competence in the determination of themeasurand(s) under test, using validated methods known to be highlyprecise and accurate, and comparable to methods in general use. Thelaboratories may, in some situations, be Reference Laboratories.

e) Consensus values from participant laboratories using statisticsdescribed in A.1.3 with consideration of the effects of extreme values.

A.1.2 Assigned values should be determined to evaluate participants fairly, yet toencourage interlaboratory and intermethod agreement. This is accomplishedthrough selection of common comparison groups, wherever possible, and the

use of common assigned values.

A.1.3 The following statistics may be appropriate when assigned values aredetermined by consensus techniques:

a) qualitative value consensus of a predetermined majority percentage(usually expressed on a nominal or ordinal scale);

b) quantitative value average for an appropriate comparison groupsuch as

i) mean, which may be weighted or transformed (e.g. trimmed orgeometric mean),

ii) median, mode or other robust measure.A.1.4 Where appropriate, the uncertainty of assigned values should be determined

using procedures described in Guide to the Expression of Uncertainty inMeasurement.

A.1.5 Extreme results are treated as follows.

When participants results are used to determine assigned values, techniquesshould be in place to minimise the influence of extreme results. This can beaccomplished with robust statistical methods or by removing outliers prior tocalculation (see ISO 5725-2). In larger or routine schemes, it may be possibleto have automated outliers screens.

a) If results are removed as outliers, they should be removed only forcalculation of summary statistics. These results should still be evaluatedwithin the proficiency scheme and be given the appropriate performancerating.


25/36



A.1.6 Other considerations are as follows.

a) Ideally, if assigned values are determined by reference or participantconsensus, the coordinator should have a procedure to establish thetrueness of the assigned values and for reviewing the distribution of thedata.

b) The coordinator should have criteria for the acceptability of an assignedvalue in terms of its uncertainty.

A.2 Calculation of performance statistics

A.2.1 Performance on single test items

A.2.1.1 Proficiency test results often need to be transformed into a performancestatistic, to aid interpretation and to allow comparison with defined goals. Theobjective is to measure the deviation from the assigned value in a manner thatallows comparison with performance criteria. Techniques may range from noprocessing required to complex statistical transformations.

A.2.1.2 Performance measures should be meaningful to scheme participants.

Therefore, measures should relate to the application needs for the test and bewell understood or traditional within a particular field.

A.2.1.3 Variability measures are often used for calculation of performance statisticsand in summary reports of proficiency testing schemes. Common examples ofsuch variability measures for an appropriate comparison group include:

a) standard deviation (SD)

b) coefficient of variation (CV) or relative standard deviation (RSD)

c) percentiles, median absolute deviation or other robust measures.

A.2.1.4 For qualitative results, no calculation is usually necessary.

Commonly used statistics for quantitative results are listed below in order ofincreasing degree of transformation of participants results

a) Difference (x X), where x is the participants result and X is theassigned value.

b) Percent difference

(x X)

Xx 100

c) Percentile or rank.


26/36



d) z scores, where

x Xz = s

and s is an appropriate estimate/measure of variability which is selectedto meet the requirements of the scheme. This model can be used both inthe situation where X and s are derived from participants results or when

X and s are not derived from (all) the participant results. [For example,when assigned values and variability are specified; refer to 4.2 ofInternational Harmonized Protocol for Proficiency Testing of (Chemical)Analytical Laboratories.]

e) En numbers (typically used in measurement comparison schemes), where

x XEn =

U2lab + U2ref

and Ulab is the uncertainty of a participants result and Uref is theuncertainty of the reference laboratorys assigned value.

A.2.1.5 Considerations are as follows

a) The simple difference between the participants result and the assignedvalue may be adequate to determine performance, and is most easilyunderstood by participants. The quantity (x X) is called the estimate oflaboratory bias in ISO 5725-4

b) The percent difference adjusts for concentration, and is well understoodby participants.

c) Percentiles or ranks are useful for highly disperse or skewed, results,ordinal responses, or when there are a limited number of differentresponses. This technique should be used with caution.

d) Transformed results may be preferred, or necessary, depending on thenature of the test. For example, dilution-based results are a form ofgeometric scale, transformable by logarithms.

e) If statistical criteria are used (e.g. z scores), the estimates of variabilityshould be reliable; that is; based on enough observations to reduce theinfluence of extreme results and achieve low uncertainty.


27/36



A.2.2 Combined performance scores

A.2.2.1 Performance may be evaluated on the basis of more than one result in asingle proficiency test round. This occurs when there is more than one testitem for a particular measurand, or a family of related measurands. This wouldbe done to provide a more comprehensive evaluation of performance.

Graphical Methods such as the Youden Plot or a plot showing Mandels

h-statistics are effective techniques for interpreting performance (see ISO5725-2).

Examples are as follows.

a) Composite score for the same measurand:

- Number of satisfactory results;

- average or summed z score;

- average absolute difference (in units or percent);

- summed absolute difference (or square difference);

b) Composite score for different measurands:- number (or percent) of satisfactory results;

- average absolute z scores;

- average absolute difference relative to the evaluation limits.

A.2.2.2 Considerations are as follows.

a) Scores may be transformed (if necessary) so that they all follow the sameassumed distribution (e.g. Gaussian for z scores or chi square forsquared differences).

b) There should be a check for extreme values that could heavily influence a

quantitative composite score.

A.3 Evaluation of performance

A.3.1 Initial performance

Criteria for performance evaluation should be established after taking intoaccount whether the performance measure involves certain features.


28/36



A.3.1.1 These features are the following.

a) Expert consensus: where the advisory group, or other qualified experts,directly determine whether reported results are fit for the purpose. Expertconsensus is the typical way to assess results for qualitative tests.

b) Fitness for purpose: considering, for example, method performancespecifications and participants recognized level of operation.

c) Statistical determination for scores: where criteria should beappropriate for each scores. Common examples of application of scoresare:

i) for z scores:

|z| 2 = satisfactory

2 < |z| < 3 = questionable

|z| 3 = unsatisfactory

ii) for En numbers:

|En| 1 = satisfactory|En| > 1 = unsatisfactory

d) Consensus of participants: the range of scores or results used by somepercentage of participants, or from a reference group, such as:

- Central percentage (80%, 90% or 95%) satisfactory, or

- One-sided percentage (lowest 90%) satisfactory.

A.3.1.2 For split-sample designs, an objective may be to identify inadequatecalibration and/or large random fluctuation in results. In these cases,evaluations should be based on an adequate number of results and across awide range of concentrations. Graphical techniques are useful for identifying

and describing these problems, in particular, graphs showing the differencesbetween the laboratories plotted against the corresponding average values.Results can be compared using regression analysis and analysis of residualswith appropriate parametric or non-parametric techniques.

A.3.1.3 Graphs should be used whenever possible to show performance (e.g.histograms, error bar charts, ordered z score charts). These charts can beused to show:

a) distributions of participant values;

b) relationship between results on multiple test items; and

c) comparative distributions for different methods.


29/36



A.3.2 Monitoring performance over time

A.3.2.1 A proficiency test scheme can include techniques to monitor performance overtime. The statistical techniques should allow participants to see the variabilityin their performance; whether there are general trends or inconsistencies, andwhere the performance varies randomly.

A.3.2.2 Graphical methods should be used to facilitate interpretation by a wider variety

of readers. Traditional Shewhart control charts are useful, particularly forself-improvement purposes. Data listings and summary statistics allow moredetailed review Statistics used to evaluate performance should be used forthese graphs and tables.

A.4 Preliminary determination of test item homogeneity

Appropriate statistical techniques should be used for the evaluation of datafrom homogeneity testing of test items. One suitable approach is described inThe International Harmonized Protocol for the Proficiency testing of(Chemical) Analytical Laboratories. See Appendix II: A RecommendedProcedure for Testing Material for Sufficient Homogeneity.


30/36



Annex B(informative)

Quality management of proficiency testing programs

It is recommended that a quality system should be established and maintained. This systemshould be documented, for example, in a quality manual. It should outline the policies andprocedures which exist to ensure the quality of the proficiency testing schemes, to giveconfidence to both participants and users of participants data. It is recommended that theorganisation conducting a scheme should meet the requirements of quality management andtechnical competence based on the appropriate clauses of ISO/ IEC 17025: 2005.

The following topics are recommended for inclusion in such quality documentation.

a) Quality policy

b) Organisation of coordinating body

c) Staff training and competence

d) Staff roles and responsibilities

e) Documentation control

f) Audit and review procedures

g) Aims, scope, statistical design and format of proficiency testing schemes

h) Operational procedures

- sample preparation

- homogeneity testing of samples

- equipment

- procedures for establishing assigned values

- suppliers, including sub-contractors

- logistics

- analysis of data

i) Preparation and issuing of reports

j) Action and feedback by participants


31/36



k) Documentation and archiving of records

l) Complaint-handling procedures

m) Policies on confidentiality and ethical procedures

n) Computing information

o) Safety and other environmental factors

p) Sub-contracting

q) Fees for participation

r) Scope of availability of schemes

s) General policies on participation and on use of scheme results.


32/36



Part 2

Selection and Use

of

Proficiency Testing Programs

by

Laboratory Accreditation Bodies


33/36



Selection And Use of Proficiency Testing Programs

by NABL

1. Scope

The objectives of this part of NABL162 are :

To establish principles for the selection of proficiency testing programs for usein laboratory accreditation program; and

to assist in harmonizing the use of results of proficiency testing programs byNABL.

As results from proficiency testing programs may be used in accreditationdecisions, it is important that both NABL and participating laboratories haveconfidence in the design and operation of the programs.

It is also important for participating laboratories and laboratory accreditationassessors to have a clear understanding of NABL policies for participation in suchprograms, the criteria they use for judging successful performance in proficiencytesting programs, and their policies and procedures for following up anyunsatisfactory results from a proficiency test.

It should be recognised, however, that NABL and their assessors may take intoaccount the suitability of test data produced from other activities apart fromproficiency testing programs. This include results of laboratories own internalquality control procedures with control samples, comparison with split-sample datafrom other laboratories, performance of audit tests with certified referencematerials etc.

2. Selection of Proficiency Testing Programs

2.1 To assist in the evaluation of competence of laboratories for laboratoryaccreditation purposes, during assessment and surveillance NABL usesproficiency testing programs complying with the guidelines described in Part-I ofthis document.

2.2 For proficiency testing program run by NABL approved coordinating agencies,NABL asks for documentary evidence that the programs comply with the Part-I ofthis document.


34/36



2.3 In selecting a proficiency testing program, the following factors are considered byNABL:

the tests, measurements or calibrations involved should match the types oftests, measurements or calibrations performed by the applicant or accreditedlaboratories proposed for participation;

with the agreement of their accredited laboratories, NABL should have accessto accredited laboratories results, together with details of the programs design,procedures for establishment of assigned values, instructions to participants,statistical treatment of data and the final report from each selected proficiencytest;

the frequency at which the program is run;

the suitability of the organisational logistics for the program, such as timing,location, sample stability considerations, distribution arrangements, etc.relevant to the group of accredited laboratories proposed for the scheme;

the availability of acceptance criteria for the participating laboratories (i.e, forjudging successful performance in the proficiency test);

the costs of the selected programs;

the programs policy on maintaining participants confidentiality;

confidence in the suitability of test materials, measurement artifacts, etc usedin the program for characteristics such as homogeneity, stability and whereappropriate, traceability to national or international standards.

Note Some proficiency testing programs may offer tests which are not exactmatch for the tests performed by an accredited laboratory (for example, the use ofdifferent national standard for the same determination) but it may still betechnically justified to include the laboratories in the program if the treatment of thedata allows for consideration of any significant differences in test methodology orother factors.

2.4 The selection of a specific proficiency testing program by NABL should beauthorised by, and supervised by, suitably qualified personnel of NABL.

3. Policies on Participation in Proficiency Testing Programs

NABL 163 document is available to laboratories and other interested parties.Issues pertaining to proficiency testing program would be guided by this document.

4. Use of Results by NABL

4.1 The results of proficiency testing programs would be utilised by NABL and

participating laboratories as stated in this document.


35/36



4.2 The results that fall outside the acceptance criteria for a specific program will behandled by NABL as per the documented procedures.

4.3 For laboratories reporting unsatisfactory results, NABL should have followingpolicies to :

have the laboratory investigate and comment on its performance within anagreed time-frame;

where necessary, have the laboratory undertake any subsequent proficiencytest which may be available, to confirm that any corrective actions taken by thelaboratory are effective; and

where necessary, have on-site evaluation of the laboratory by appropriatetechnical assessors/experts to confirm that corrective actions are effective.

4.4 NABL should advise participating laboratories of the possible outcomes ofunsatisfactory performance in proficiency testing program such as i)Continuation of accreditation subject to appropriate corrective action. ii)Withdrawal of accreditation for the relevant tests.

4.5 NABL should have procedures to ensure that the records of performance oflaboratories in proficiency testing programs are maintained (in accreditation files orrecords) for the participating laboratories and are made available to technical

assessors for on-site assessments.4.6 NABL should obtain feedback from accredited laboratories of action taken from

results of proficiency testing programs, particularly for unsatisfactory performance.

5 Action and Feedback by Laboratories

5.1 Accredited laboratories should be required to maintain their own records ofperformance in proficiency testing, including the outcomes of investigations of anyunsatisfactory results and any subsequent corrective or preventative actions.

5.2 The laboratories should draw their own conclusions about their performance. Theinformation that should be taken into consideration includes :

the origin and character of test samples ; the test methods used and, where possible, the assignment of the results to

particular methods.


36/36

National Accreditation Board for Testing and Calibration Laboratories

3rd Floor, NISCAIR14, Satsang Vihar Marg

New Mehrauli RoadNew Delhi 110 067

Tel.: 91-11 26529718 20, 26526864Fax: 91-11 26529716

Website: www.nabl-india.org

biology lab guidelines

Documents