Standards and Quality In Research Laboratories. By Geoff Visser SABS Research Fellow And Prof Roy Ramphal - UNISA

Transcription

1 Standards and Quality In Research Laboratories By Geoff Visser SABS Research Fellow And Prof Roy Ramphal - UNISA

2 Introduction University research team Namibian field study Interesting findings about natural venting of methane and its effects on vegetation. could not be published because the methods were not validated and instruments were not properly calibrated Grobler, J (2012) Himba Dragon Myth not just hot air, Mail and Guardian, March 16 to 22, 2012.

3 Could attention to laboratory quality have prevented this?

4 why do we need quality in research labs? Safety Accuracy of test results Efficiency Facilitating training / teaching Commercial pressures Acceptance of results Collaborative studies Trustworthiness / Impartiality eg. sampling and analysis in environmental matters

5 what is lab quality? Many of us have associated lab quality with accreditation

6 lots of things make up lab quality Source:

7 the need for laboratory quality systems Many labs have become formally accredited independently recognised for their ability to deliver good quality [accurate] results Basis for recognition of test results Basis for assessing competence Requirements driven by customers (regulators) Need to rely on results and to hold issuers accountable Not all labs have taken up opportunity Research and teaching labs Production labs

8 why accreditation? Recognition of test results by Customer (and their customers) Regulators Other labs or researchers Certification / inspection bodies Public confidence in lab (negative is very true if you lose accreditation its bad!) Assurance of accuracy and precision of test results we can use the data Assurance of competence commitment to continuous improvement Better management of lab? Better management of uncertainty? Cost control?

9 accreditation of research labs Very few research labs accredited by SANAS Some commercial research facilities Poses some questions: 1. Are the Standards not suitable? 2. Are we doing a poor job of Marketing? 3. Is there enough demand for accreditation? 4. Would research labs benefit from a quality management system? 1. Is accreditation necessary? 2. Is quality management the same as accreditation? 5. Do we need a different standard?

10 are research labs different?

11 reasons for not getting accredited Costs associated with accreditation Regular work that could benefit from accreditation will make up a small proportion of the lab work Large variety low volume testing Using non-standard methods for once-off applications Staff turn over or lack of tenure of research personnel Method validation and accuracy is currently part of peer evaluation Journal review panels or examiners No demand from customers Competition Second party evaluation - customers

12 changing market Acceptance of research data in collaborative studies Human Genome project Distributed computation projects SKA Legal acceptance of results Sampling and testing of environmental data Limited access in SA to instruments and technical skills Research and commercial drive to use equipment but commercial pressure requires accreditation?

13 sometimes research can have commercial consequences

14 possible way forward Are we sure that accreditation adds value to all labs? Research project into why labs seek accreditation Market research on added benefits of accreditation

15 possible way forward Do we need a standard applicable to the needs of research and teaching labs? Basic QMS Does ISO/IEC add any value without accreditation?

16 possible way forward Do we need the ability to certify data coming from labs? What do we need to certify on a report by report basis? Suppliers Declaration of conformity what do I need to tell my customers in order to have them trust my test report?

17

18 Quality Assurance Laboratory Quality Assurance (QA) encompasses a range of activities that enable laboratories to achieve and maintain high levels of accuracy and proficiency despite changes in test methods and the volume of specimens tested. A good QA system does these four things: establishes standard operating procedures (SOPs) for each step of the laboratory testing process, ranging from specimen handling to instrument performance validation; defines administrative requirements, such as mandatory recordkeeping, data evaluation, and internal audits to monitor adherence to SOPs; specifies corrective actions, documentation, and the persons responsible for carrying out corrective actions when problems are identified; and sustains high-quality employee performance.