This is the fourth in a series of five articles describing the challenge of managing laboratories effectively. Today we focus on the technical side of management with special emphasis on maintaining quality results. As a review other topics are linked below:
- The Nature of the Laboratory
- Human Resource Management in the Laboratory
- Business Management of a Laboratory
- Technical Management of a Laboratory – Maintaining Quality
- Bringing it All Together for Effective Management (Future)
Technical Management/Quality:
Many approach technical and quality management as somehow separable. They are not. The selection, validation and application of testing technology to achieve reliable results is inseparable from managing laboratory quality. This can be seen in any of a number of quality management guides, but none are clearer than ISO 17025. In the Introduction we read that the standard supersedes ISO/IEC Guide 25 which:
....contained all of the requirements that testing and calibration laboratories have to meet if they wish to demonstrate that they operate a management system, are technically competent, and are able to generate technically valid results....
In the Scope, ISO 17025 states:
1.1 This International Standard specifies the general requirements for the competence to carry out tests and/or calibrations, including sampling. It cover testing and calibration performed using standard methods, non-standard methods, and laboratory-developed methods.
1.2 This International Standard is applicable to all organizations performing tests and/or calibrations. These include, for example, first-, second- and their-party laboratories, and laboratories where testing and/or calibration forms part of inspection and product certification.
It is actually hard to imagine a testing or research laboratory that does not fall under this very broad international scope. Hence, a review of ISO 17025 guidelines is indispensable in establishing, correcting and improving technical and quality management systems in a laboratory. Managers of all types engaged in laboratory work (front line to highest executives) are highly encouraged to obtain, read, understand and follow this extensive standard regardless of the specific field of technology. It is virtually impossible to have consistently reliable testing results unless the recommendations of ISO 17025 are followed. In this blog we will review just of few of these key elements.
Management System:
One of the most important features of ISO 17025 is recognition that there must be a management system in place. It does not do to hire the “best and the brightest,” buy them “very expensive toys” and turn them loose. Nothing good will ever come of that. The data generated, even if valid, will have no credibility and for that reason will have no value in the marketplace. Quite to the contrary, a management systems must be deliberately design and maintained that touches all facets of the laboratory. This includes systems that:
- Provide a planned and functioning quality management organization,
- Provide for written quality standards of performance – including Standard Operation Procedures (SOP’s),
- Provide for the management of all pertinent records – including results and documents supporting those results,
- Provide a stated quality policy that can supersede and critique specific procedures,
- Provide methods for identifying and maintaining quality objectives and
- Provide means for management review of performance.
Unless an organization has the systems in place to implement these basic functions, quality performance will eventually suffer with dramatic consequences. That “system” includes a human resource element. A management system cannot function if the managers are do not have the proper training, experience and commitment. Furthermore, the management system must be designed such that it naturally hires, trains and utilizes appropriate technical personnel in the testing processes.
Document Control:
ISO 17025 relies heavily on documentation to ensure that quality systems are in place and operational. This grates on many technical personnel. Unfortunately, failure to document almost always results in later confusion and doubt. This destroys the value of any test result. Even more insidious is the failure to plan document control and retrieval. Much too often, good initial documentation is frustrated by the inability to retrieve the data and the supporting information needed to verify and validate those data. Without the ability to retrieve critical information, the value of testing efforts evaporates.
A simple “record and keep everything for X years” is not a document control plan. Quite to the contrary, in the case of document control, less is often better. Aggressive destruction of redundant and irrelevant records quite often helps focus attention and efforts onto the critical few rather than the irrelevant many. Retention of key data elements “forever” may be necessary to preserve the value of some experimental efforts.
Corrective Action:
Corrective Action is another key tenant of ISO 17025. It appears in many ways. Clearly when procedures fail, there needs to be a recognition that the system has failed and corrective action should be taken. Most organization formalize procedures around failures that are likely and maybe even anticipated. The danger here is that these “failures” can become “routine” and not considered failures at all, but just the way business is done. When “corrective action” becomes routine, it is no longer “corrective.” It has become a part of the standard way of doing business. In this case, Corrective Action must go deeper and investigate the entire quality system. It is at this point that Corrective Action becomes a management review to ferret out where policies or even managers themselves have gone awry.
Selection and Validation of Methods:
ISO 17025 specifically acknowledges a common error in testing – using the wrong method. Applying inappropriate measurement technique on a sample happens far too often. The results of the measurement may be horribly biased and that bias may not be at all obvious. Careful selection of sampling, sub-sampling and measurement techniques are all crucial for valid results. Horror stories in this area abound. Hence, the Standard requires that SOP’s clearly state the appropriate scope for sampling, analysis and data reporting. This selection process cannot be left up to chance, but must be an overt part of the management system so that even accidental mismatches are quickly identified and corrected.
The Standard goes on to say that all methods must be validated for the types of samples being processed. Where practical “standard methods” should be used. These are methods that have been validated by multiple organizations and documented as being appropriate for specific sample types. The Standard describes some of the techniques used for validating non-standard method. The technique used “shall be as extensive as is necessary to meet the needs of the given application or field of application.” This is a tall order in many cases and can rarely be done without some equivocation. The limitations of the validation should be disclosed to data users.
Statistical Controls:
ISO 17025 relies heavily on statistical validity of data. Data that are devoid of any estimates of precision and accuracy are of little value. In all cases, data should have associated estimates of uncertainty. Where this can be done in a statistically valid fashion the actual uncertainty should be reported. Where this is not practical, comments on the probable error and potential error should be given. The degree of rigor in estimating error should be appropriate for the use of the data. The management system should be capable of ensuring that the precision and accuracy of the results are appropriate for the use of the data and that, where this may be questionable, the data user is appropriately notified.
These same statistical control principles must be applied to all equipment, standards and reagents. Calibration procedures and checks should be documented in SOP’s and the acquiring, use and retiring of standards and reagents be appropriately controlled.
Sampling:
Unlike many standards, ISO 17025 overtly address the appropriateness of sampling. This is one of the Standard’s most significant contributions to quality management. Inappropriate sampling usually results in bias and often contributes to imprecision. Too often the laboratory ignores or is unaware of the sampling challenge. This is especially problematic when the laboratory must take a sub-sample in the laboratory to do an analysis. If unaware of innate inhomogeneity or instability, the laboratory can inadvertently ruin an otherwise valid assay. The Standard strongly suggests cooperation between the data user, field personnel and the laboratory.
Traceability:
The Standard emphasize in several places the need to be able to trace calibration materials to the International System of Units (SI). This includes masses, lengths, volumes, temperatures, pressures, frequencies, voltages, energies, etc. Pertinent units must be clearly and accurately reported in all results and calculations. Traceability means that all important calibration standards and materials are certified as to their value and can be traced back to SI, usually by the manufacturer of the standard or material. Where this is not practical, the laboratory must make its own disclosure on how a calibration standard or material was prepared by using SI traceable calibration standards or materials. The Standard specifically recognizes the potential for inappropriate calibrations that can be difficult to discover.
Appropriate Data Reporting:
The Standard specifically acknowledges that some test results will be deficient in some ways, but may still be useable for the intended purpose. The Standard requires that all deficiencies be acknowledge to data users so that they may be the final judge on the utility of the data. The Standard encourages “full disclosure” regardless of the source of the deficiency.
Conclusions:
ISO 17025 is a very detailed listing of quality management principles that should be used by all laboratories. We have listed most of the general principles. Managers should be familiar with and embracing of these principles. They are a very good start toward consistently reliable and relevant laboratory data.
Stites & Associates, LLC, is a group of technical professionals who work with clients to improve laboratory performance and evaluate and improve technology by applying good management judgment based on objective evidence and sound scientific thinking. For more information see: www.tek-dev.net.
Comments