Quality-management system holds the key to an error-free lab

Laboratory quality can be defined as accuracy, reliability, and timeliness of reported test results. Quality assurance is an important aspect of in-vitro diagnostics (IVDs), which encompasses all aspects of the quality management of diagnostic testing, including quality control of the testing process as well as monitoring and control of the outcomes of that process. These are designed to ensure that test results are as accurate as possible.

Laboratories produce test results that are widely used in clinical and public health settings, and health outcomes depend on the accuracy of the testing and reporting. Inaccurate IVD testing can have significant adverse impacts for patients. For an individual patient, it means not only a delay in receiving the correct diagnosis, but can also result in improper or unnecessary treatment, which may lead to long-term complications, delay in correct diagnosis, additional and unnecessary diagnostic testing, additional clinic visits and additional cost. For public health programs, failure to identify pathogens correctly in the event of an outbreak can lead, among other things, to a delay in determining the true cause and extent of an outbreak, and inadequate or incorrect control measures being put into place. In general, poor diagnostic quality leads to poor patient outcomes, increasing morbidity and mortality, and poses a barrier to effective public health interventions. Although it should be self-evident that IVD testing is only valuable if it is accurate, quality assurance has often lagged behind scale-up of testing in resource-limited settings.

To achieve the highest level of accuracy and reliability, it is essential to perform all processes and procedures in the laboratory in the best possible way. In order to have a functioning quality-management system, the structure and management of the laboratory must be organized so that quality policies can be established and implemented. Therefore, the quality-management system model, which looks at the entire system, is very important for achieving good laboratory performance. There must be a strong supporting organizational structure – management commitment is crucial – and there must be a mechanism for implementation and monitoring.

The laboratory is a complex system, involving many steps of activity and many people. The most important laboratory resource is competent, motivated staff. The quality-management system addresses many elements of personnel management and oversight, and reminds us of the importance of encouragement and motivation. Many kinds of equipment are used in the laboratory, and each piece of equipment must be functioning properly. Choosing the right equipment, installing it correctly, ensuring that new equipment works properly, and having a system for maintenance are all part of the equipment-management program in a quality-management system. The management of reagents and supplies in the laboratory is often a challenging task. However, proper management of purchasing and inventory can produce cost savings in addition to ensuring supplies and reagents are available when needed. The procedures that are a part of management of purchasing and inventory are designed to ensure that all reagents and supplies are of good quality, and that they are used and stored in a manner that preserves integrity and reliability. SOPs need to be developed for the shipping and reception of samples for QA. An investigation request form should accompany samples that are sent, and every effort should be made to ensure that transport conditions are appropriate and the same for all samples (for example, route, packing conditions, and type of container). Attainment of levels of proficiency by the laboratory and its staff may be a prerequisite.

Process control is comprised of several factors that are important in ensuring the quality of the laboratory testing processes. These factors include quality control for testing, appropriate management of the sample, including collection and handling, and method verification and validation. The process of assessment is a tool for examining laboratory performance and comparing it to standards, benchmarks, or the performance of other laboratories. Assessment may be internal (performed within the laboratory using its own staff) or it may be external (conducted by a group or agency outside the laboratory). Laboratory quality standards are an important part of the assessment process, serving as benchmarks for the laboratory.

Internal QC comprises procedures that are introduced within the field laboratory. External QC involves external monitoring, such as the duplicate testing of samples in another reference laboratory to serve as a gold standard or blind measurement in the field laboratory of a set of samples provided by an external reference laboratory. The essence of internal QC lies in a tight circle of checks, reporting, evaluation, and action. It is essential to have detailed manuals of every procedure, with a checklist to be consulted each time the procedure is run. Well-kept records, with regular review of these by the supervisors, are key elements in QC. Laboratory QC procedures must be an integral part of the work plan.

External QA programs help to check the accuracy of test results. Reproducibility can be assessed adequately by internal QC procedures, but checks on accuracy are best done, for many tests, in collaboration with other laboratories. The results from a laboratory may be highly reproducible within that laboratory but might be consistently incorrect. There are a range of external QA programs, which offer both testing of site-generated samples and/or the provision of a panel of samples with known characteristics that are specific for each assay (for example, biochemistry and hematology analyzers). If specimens are selected for QA checks after they have been analyzed locally, and in such a way that the laboratory staff will not know which specimen will be selected, the use of site-generated QA systems are to be preferred to QA that depends on specimens provided by the external laboratory, since the laboratory staff will know which these QA panel specimens are and may take particular care with them.

The reliability of laboratory results should be tested by regular checks on their reproducibility. The level of acceptable variation will depend both on the test and the study. This information is normally predefined in SOPs, test manuals, and the study protocol. Many test systems have inbuilt controls for this purpose, using standardized reagents of known concentration or quantity. The use of such standard controls is important, but not necessarily sufficient, to monitor the quality of test procedures. Depending on the procedure, samples should be tested in duplicate or re-read by a second technician. The frequency with which such repeats are performed depends upon how well the laboratory is running and how long it has been doing the test. Typically, when a test is first introduced or a new staff member is conducting the test, a high frequency of such checking is appropriate, with a decreased frequency as the procedures become more familiar, assuming the re-tests are showing negligible differences to the original results. In many circumstances, it will be appropriate to ensure that duplicate analyses are done on between 5 percent and 10 percent of samples on a routine basis. It is sometimes possible and advisable to seed known positive or known negative samples into the test runs, which are labelled in such a way that the laboratory staff running the test cannot spot them. This is particularly important if it is expected that the great majority of samples will either test negative or test positive (for example, seeding a positive result if a long run of negatives is expected).

Reproducibility should be checked within batches, between batches, and from day to day or week to week by the use of appropriate controls. Intra-observer variation can be determined by having duplicate samples processed by the same observer at different times, and inter-observer variation measured by having the same samples processed independently by two different staff members. Inter-product variation is tested by comparing new versus old batches of staining solutions, media, reagents, and so on, on a group of the same samples.

The product of the laboratory is information, primarily in the form of test reporting. Information (data) needs to be carefully managed to ensure accuracy and confidentiality, as well as accessibility to the laboratory staff and to the healthcare providers. Information may be managed and conveyed with either paper systems or with computers. The primary goal in a quality-management system is continuous improvement of the laboratory processes, and this must be done in a systematic manner. There are a number of tools that are useful for process improvement. The concept of customer service has often been overlooked in laboratory practice. However, it is important to note that the laboratory is a service organization; therefore, it is essential that clients of the laboratory receive what they need. The laboratory should understand who the customers are, and should assess their needs and use customer feedback for making improvements.

Quality assurance is often viewed as an added cost of an IVD system. This is particularly evident in resource-limited settings, where countries must make difficult decisions each year with respect to how to apportion budgets for diagnostics. To date, most testing in these countries, especially device-based testing, has taken place in centralized laboratories, while the majority of patients live, and present for treatment, in rural or peri-urban areas far from those facilities. Therefore, effective scale-up of IVDs requires not only increasing capacity for testing centrally, but also decentralizing testing to take diagnostics closer to the point of care (POC), with measures to ensure the quality of testing, wherever it is performed.

Laboratories not implementing a good quality-management system are guaranteed that there will be many errors and problems occurring that may go undetected. Implementing a quality-management system may not guarantee an error-free laboratory, but it does yield a high-quality laboratory that detects errors and prevents them from recurring.