CHAPTER 3

Quality aspects of laboratory

medicine
Helen Bruce  •  Marta Lapsley

CHAPTER OUTLINE

INTRODUCTION  21 CLINICAL QUALITY INDICATORS  24

WHAT IS QUALITY?  21 Clinical effectiveness  24
Key performance indicators  24
QUALITY STANDARDS  21 Demand management  25
Quality assurance  21
Regulation of laboratories  22 EVIDENCE-BASED CLINICAL
BIOCHEMISTRY  25
Quality management systems  22
Personnel  22 POINT-OF-CARE TESTING  25
Premises and environment  22 CONCLUSION  26
Information systems  23
Evaluation and audit  23

INTRODUCTION that they produce can be relied upon to support clinical

decision-making. They include internal quality control
It was emphasized in Chapter 2 that every step in (IQC) schemes, for example involving replicate analysis
the process, from the ordering of the investigation, of clinical samples and the repeated analysis of reference
through collection of the specimen(s) required, their samples of identical composition, and external quality as-
transport to the laboratory and analysis, to the de- surance (EQA) schemes ­involving the analysis of identical
livery of a report to the clinician, is vital in ensuring samples by a large number of laboratories.
that results of ­
investigations are used appropriately. The measurement of IQC samples allows a prospec-
Interwoven throughout each step is the requirement to tive estimate of the precision of the analyses that are be-
ensure the quality of the whole process. Chapter 2 has ing performed. The frequency of measurement of such
already discussed the need to minimize analytical varia- controls may depend upon each laboratory’s workload
tion through the use of processes to reduce imprecision and the type of analysis undertaken. For example, it is not
and bias. The use of standard operating procedures has unusual to run quality control material at the start and
also been alluded to, but this chapter will expand on the the end of a sample batch measured by enzyme-linked
use of laboratory standards to maintain quality. immunosorbent assay (ELISA).
The measurement of EQA samples provides labora-
tories with a retrospective view of the accuracy of their
WHAT IS QUALITY? measurements and the consistency of any bias that may
be present. The data can be invaluable when trouble-
Quality is difficult to define, and may be considered to shooting possible assay problems, by allowing the labo-
be ‘doing the right test, for the right patient, at the right ratory to compare their result with both the reference
time for the right reason’, but may equally be defined as value (if available) and the method-specific mean. The
‘adhering to stringent processes in order to complete the review and interpretation of EQA reports is an impor-
task’. Ultimately, perhaps the question to ask is how to tant skill for laboratory personnel to develop because
assess the quality of a laboratory service. of the variety of ways in which different EQA schemes
choose to present their data. (Examples may be accessed
via the website addresses of the various EQA providers
QUALITY STANDARDS included in Further reading, below.)
Not only the analysis itself, but also the interpretation
Quality assurance of the result should be subjected to quality assurance:
Clinical biochemistry laboratories operate extensive there exists a peer-reviewed EQA scheme for interpreta-
quality assurance procedures to ensure that the results tive comments.

21
22 Clinical biochemistry

Reviewing quality assurance procedures is an es- their own discipline or as part of a multidisciplinary
sential part of the audit of laboratory performance (the team. The development of web-based document man-
assessment of overall performance with regard to the agement systems (e.g. QPulse, i-Passport) has enabled
appropriateness of the use of tests, the interpretation laboratories to greatly improve document control and to
of their results, adherence to standard procedures, cost- track review dates, changes and the number of printed
effectiveness etc.). Laboratory data should also be in- copies in circulation. Complete written procedures
cluded in the medical audit, in which the effectiveness should be in place for all tasks performed in a laboratory.
of all aspects of clinical care is examined. The ultimate Manufacturers are obliged to publish ‘Instructions for
test of the performance of laboratory data is provided by Use’ with each reagent supplied that contains the ma-
the clinical outcome when this has been determined, in jority of worthwhile data. These procedures, however,
whole or in part, by those data. form only part of the process, as the essence of a quality
management system is that it documents what should
Regulation of laboratories take place within the laboratory in its entirety. All mem-
bers of laboratory staff should be familiar with the qual-
External regulation of laboratories has evolved over some ity management system, understand the requirement for
time to include every aspect of the laboratory service. such a system and the need to maintain the documents
Independent regulatory bodies are empowered to award within it. Indeed, it is the effective implementation of
accredited status to laboratories meeting stringent stan- the system, and adherence to it, that is likely to lead to
dards and to fail to award or remove accreditation from quality improvement. Monitoring the effectiveness of
laboratories which are found to have serious failings at the system is achieved by audit (see below) and by re-
inspection. Throughout the world, there are several or- viewing processes in a regular basis.
ganizations that fulfil this role: International Standards Many laboratories try to minimize the amount of
Organization (ISO), United Kingdom Accreditation printed material generated, owing to ever greater space
Service (UKAS) and the Division of Laboratory Services and storage constraints. Improved reliability of informa-
and Standards in the USA, to name but a few. Their tion technology and the need to minimize waste are also
main aim is to ensure a laboratory’s quality and compe- important driving factors towards operating a paperless
tence to perform specified tasks. It is important to note Quality Management system.
that all parts of the laboratory are inspected, including
the environment in which the laboratory is based. Very
dilapidated buildings (in which pathology departments
Personnel
are sometimes located) will be a cause of concern for When contemplating quality, staffing issues may not im-
the inspectors, as it may reflect the low priority given to mediately spring to mind. Staff are, however, the most
laboratories within a wider healthcare setting, such as in valuable assets of any laboratory. Staff morale and con-
a hospital. tinuing professional development play a vital role in a
Within the UK, for example, the current list of stan- ‘good’ quality laboratory. Education and training of all
dards against which laboratories are judged comprises levels of staff within the laboratory ensures that their
almost 60 pages. The broad areas covered are listed in knowledge and understanding of laboratory science is
Box 3.1. It is worth noting that the analytical process it- current and up-to-date. The competence of staff under-
self is just one out of eight categories examined by such taking laboratory tasks should be documented and regu-
an inspection. larly reviewed. Formal individual review meetings are
recommended to assess training needs and to promote
further development.
Quality management systems
A system of ‘Quality Management’ is an all-encompassing
term used to describe the steps a laboratory may take
Premises and environment
to maintain current procedures and to establish continual The control of access to laboratories and the efficient use
service improvement via quality objectives. of heating, lighting and space are important consider-
The majority of laboratories in the UK will have a ations when evaluating a suitable working environment.
designated Quality Manager, either exclusively within Health and safety aspects of the premises and environ-
ment need to be reviewed, both in terms of staff safety
(e.g. lone working, risk of slips and trips, adequate fire
BOX 3.1 The areas examined by the UK plan) and general laboratory safety such as electrical is-
Accreditation Service sues (e.g. overloaded sockets, regular testing of electrical
appliances, use of extension cables) and chemical safety
• Quality management system (e.g. appropriate use of material safety data sheets and
• Personnel adequate storage facilities for acids, controlled drugs
• Premises and environment and toxins). The latter category may also include aware-
• Equipment, IT and materials
• Pre-examination processes
ness of laboratory waste disposal, which will encompass
• Examination processes both biological and chemical waste. In the UK, for ex-
• Post-examination phase ample, NHS Trusts have financial incentives related to
• Evaluation and quality assurance recycling and alternative methods of disposal other than
incineration.
 3 
Quality aspects of laboratory medicine 23

Information systems
The development of sophisticated information systems
has had a huge impact on numerous aspects of laboratory Identify
Review
area for
medicine. Pre-analytical improvements include the ability standards
audit
to display relevant test information and sample require-
ments via electronic requesting. Real-time availability of
results and the possibility of report acknowledgement are
also areas revolutionized by information systems. Set criteria
Re-audit
and standards
Information systems affect the way in which laborato-
ries communicate with their users. Many laboratories now
maintain an electronic laboratory handbook, as well as a
paper copy. The advantages of an electronic version are
that it reduces the cost of production (and paper waste) and Implement Collect data
it is easier to maintain and keep up-to-date. Patient infor- changes
mation leaflets and testing protocols such as dynamic func-
tion tests can easily be accessed via hyperlinks, if available. Identify Data
The increasing use of e-mail has allowed laboratories areas for analysis and
to communicate electronically with end-users, although improvement comparison
and change with standards
security must be assured before sending patient identifi-
able data. Future developments will increasingly allow pa-
tients to use this or other electronic means to access their
own results directly. In the UK, some patient groups, such FIGURE 3.1 ■ The complete audit cycle. The process can be
as those with kidney disorders, already have electronic ac- started at almost any point of the cycle, but often the first task
cess to selected results via a nationally organized database. is to identify an area of interest and then determine standards
The integration of laboratory information systems with against which to conduct an audit.
electronic patient records in both primary and secondary
care has revolutionized the delivery of pathology results over in trying to assess the impact of laboratory testing on pa-
the past decades. The ability to view test results remotely tient pathways and outcomes. Ultimately, the role of the
and securely has improved the safe delivery of healthcare, laboratory is to improve the healthcare of patients and,
especially in non-traditional settings, and has expedited the although difficult to measure, the contribution of labora-
move of care from hospitals to within the community. tory testing should be documented. Collaboration with
other clinical disciplines allows the laboratory to forge
vital relationships with clinicians and other staff within
Evaluation and audit the healthcare environment.
The use of audit tools in laboratory medicine is wide- The third type of evaluation that the laboratory
spread. During formal assessment of the laboratory’s should undertake is an assessment of user satisfaction.
service, a combination of vertical, horizontal and examina- It is important to engage all users of the laboratory ser-
tion audits are used to assess compliance with the quality vice, as each different group may have different needs
standards set. Vertical audits generally look at the sample and requirements. Clinicians, nurses, other healthcare
journey throughout the ­entire testing process, from test professionals and other laboratories should be asked
requesting and phlebotomy (if possible), to report gen- to contribute. All aspects of the service can be covered
eration and receipt by the requestor. Actual events are from the users’ opinions on the judicious use of inter-
compared with written procedures to check adherence. pretative comments to turnaround times and phlebot-
A failure to follow any stated procedure may trigger a omy services. User satisfaction surveys may eventually
horizontal audit of that particular part of the testing pro- be circulated to patients (who may be the true end-users
cess. A rather extreme example would be that during the of the service), especially those with long-term chronic
course of a vertical audit, it is discovered that no quality disease states (e.g. diabetes mellitus, chronic kidney dis-
control material was run during analysis of a particular ease), as they take greater ownership of their health and
analyte such as glycated haem­oglobin; examiners may act treatment targets. Box 3.2 gives a list of the other areas
upon this information and perform a horizontal audit of covered by questions commonly seen in user surveys.
all recent analytical batches of glycated haemoglobin to
see if this was an isolated occurrence or if there was a
systematic failure to follow the stated procedure. BOX 3.2  xamples of topics covered within
E
Ensuring that the audit cycle is complete is critical for user satisfaction surveys
continued quality improvement. Re-auditing previous
processes, in which highlighted deficiencies have forced a • Availability of clinical advice
• Ease of requesting
change in practice, is vital to ensure that the changes in
• Availability of test information
practice have resulted in the anticipated improvements. • Frequency of phlebotomy collections (primary care)
Figure 3.1 shows the key steps to completing an audit cycle. • Online access to results (availability of)
Clinical audit is another key aspect of evaluating labo- • Out of hours service provision
ratory effectiveness. Clinical audit plays an important role
24 Clinical biochemistry

CLINICAL QUALITY INDICATORS Clinical effectiveness

In the context of laboratory medicine, this may be per-
Laboratories are generally adept at maintaining qual-
ceived as the laboratory’s ability to introduce new tests
ity standards within their own environment; however,
and to dissuade clinicians from using tests deemed to be
the need to determine and measure objectively their
obsolete. Clinical biochemists may be involved in the
clinical effectiveness is growing, owing to increasing
development of specific testing protocols in Emergency
pressure on limited resources. Clinical quality in-
Departments, for example, or in the interpretation of
dicators have recently been developed to aid in this
data in complex cases. Another way in which laborato-
process. The key to the use of these indicators is that
ries contribute to effective clinical care, is the identifica-
they must be easily measureable and that the results
tion of artefactual causes of altered biochemical analytes,
can been collected in a timely manner (a yearly review
such as the effect of thrombocytosis on potassium con-
would be of limited use in highlighting errors and im-
centration. The implementation of national guidelines
proving performance promptly). Several quality indi-
and the adoption of agreed testing strategies or more
cators can be brought to mind, but if the mechanism
precise analytical methods (such as the measurement of
for capturing the necessary data is too demanding or
creatinine by enzymatic methods), are all examples of
difficult to achieve, then the whole process becomes
ways in which the clinical effectiveness of laboratories
meaningless.
has been improved.
Quality indicators can fall into two broad categories:
those that are designed to highlight improvement and those
that are used to capture a possible deterioration in stan- Key performance indicators
dards. An example of each may include the use of exter-
nal quality assurance to show improving assay precision In an attempt to objectively assess the quality and ef-
and the monitoring of turnaround times from emer- fectiveness of predominantly pre- and postanalytical
gency departments to detect peaks and troughs in service processes within the laboratory, the Royal College of
­provision. Turnaround times in particular, can be an area Pathologists in the UK has established Key Performance
where the laboratory and requesting clinicians can have Indicators. These suggested standards cover a wide range
differing views, with clinicians considering turnaround of areas, from the percentage of staff in training to the
time to be ‘vein-to-brain’ time rather than from receipt speed at which a laboratory can respond to a request for
of sample in the laboratory to a clinically validated re- clinical advice (what defines clinical advice may be a con-
sult. Table 3.1 lists other potential quality indicators that tentious issue). Table 3.2 lists some of the other additional
laboratories may be able to use to monitor their own performance standards suggested by the Royal College
performance. of Pathologists. These are still in their infancy and some

TABLE 3.1  Examples of potential quality indicators

Category Example quality indicator

Preanalytical Investigation strategies and protocols

Introduction and maintenance of reference (critical) change values
Analytical Matching analytical performance to clinical requirements
Postanalytical Data interpretation
Addition of appropriate further tests
Direct clinical services Patient/ward interventions, e.g. review of patients with metabolic disease,
electrolyte or fluid balance problems
Participation in multidisciplinary team meetings
Education Undergraduate and postgraduate teaching; participation at hospital-wide
events such as Grand Round meetings; newsletters

TABLE 3.2  Some key performance indicators recommended by the UK Royal

College of Pathologists
Category Performance standard and method of measurement
Training 15–30% of the biomedical scientist, clinical scientist and medical staff
should be in training posts
User surveys The laboratory should include patients in the annual user satisfaction
survey
Turnaround times The percentage of investigations from A&E to be completed within 1 h.
Target: 90% from April 2014
Results communication The percentage of critical results phoned to the requesting clinician within
2 h of availability. Target: 97% by April 2014
 3 
Quality aspects of laboratory medicine 25

will require further clarification. However, they do pro- there are none for many well-established biochemical
vide an excellent basis on which a laboratory can assess its tests, and there is no requirement to conduct studies
current performance. In order for these key performance of the efficacy and validity of new tests before they are
indicators to have an impact on commissioning patterns introduced to the market, as there is with therapeutic
and thereby possibly reward the best-performing labora- agents. For example, numerous new tumour mark-
tories, each participating laboratory should be expected ers have been developed in recent years, but for many,
to publish some of their results on a regular basis, thus there is only limited evidence (and only in specific con-
allowing comparison with their peers. texts) of their clinical value. The context in which an
investigation is used is an essential consideration. For
Demand management example, there is no doubt that the measurement of
prostate-specific antigen (PSA) is of value in monitoring
In the past, clinicians were often allowed unrestricted ac- the response of patients with prostatic cancer to treat-
cess to all available assays within the laboratory, but as costs ment, but basal values have little prognostic significance
increase, test repertoires expand and undergraduate train- and there remains no definite evidence that screening
ing in laboratory medicine decreases, laboratories now find older men for prostatic cancer by measuring PSA has
themselves attempting to manage the increasing demand any ­effect on outcome.
placed on them. Considerable work has been done to as- Evidence-based clinical biochemistry is gathering mo-
certain sensible repeat testing intervals for analytes which mentum. It starts with the technical evaluation of a test
are too frequently requested in secondary care. Some ex- and encompasses the assessment of diagnostic perfor-
amples include bone and liver profiles, vitamin D and thy- mance and clinical utility, as discussed in Chapter 2. One
roid function testing. Electronic order requesting enables of its important products is the development of guide-
the laboratory to communicate warning messages to the lines for the investigation of patients with specific condi-
clinician prior to further repeat requesting and also to edu- tions. But just as there is continued clinical surveillance
cate the users about the most efficient use of laboratory ser- of drugs after they have been introduced to the market
vices. Another form of demand management may involve (resulting, in some cases, in their later being withdrawn
the manual vetting of requests from within the laboratory. or having their licences altered), so the impact of guide-
This may be for specialist tests that are referred to other lines and other products of evidence-based laboratory
laboratories or for analytes that have a specific clinical role, medicine must be reviewed to ensure that they are ap-
such as tumour markers. Vetting performs a dual purpose; propriate and effective, and remain so with the passage of
first, it enables the lab to control costs and second, it allows time. The process of clinical audit – defining standards;
the lab to open a dialogue with the requestor regarding the examining processes against these standards; identifying
need and suitability of the test. deficiencies; modifying the processes to correct them,
and repeating the cycle to check for improved perfor-
mance – is fundamental to this.
EVIDENCE-BASED CLINICAL BIOCHEMISTRY
The later years of the 20th century saw the establish- POINT-OF-CARE TESTING
ment and development of the concept of evidence-based
medicine, in essence the use of the best available data There is a current trend to deliver a more patient-centric
from robust clinical studies to inform the management service with some laboratory testing taking place within
of patients, including diagnosis, prognosis and treat- the community, at point-of-care. This desirability, under
ment. This has led to the promulgation of the concept of some circumstances, to provide biochemical services at
evidence-based clinical biochemistry (and evidence-based the bedside or in the clinic or doctor’s surgery, has led
laboratory medicine in general) in which, for example, to the development of techniques and analytical instru-
the relationship between investigations and clinical ments that can be used reliably by individuals who do not
outcomes is rigorously analysed, using tools such as likeli- necessarily have rigorous laboratory training. Examples
hood ratios (see p. 19), to determine which investigations include hydrogen ion and blood gas analysis in inten-
are the most appropriate (in terms of being valid and in- sive therapy units, principally as an aid to the manage-
formative) in individual clinical situations and evaluating ment of patients on ventilators; dipstick analysis of urine
their subsequent use. samples; blood glucose measurements on hospital wards
In evidence-based medicine, the gold standard for and in patients’ homes, and bilirubin analysis in neonatal
providing evidence is the systematic review of evidence intensive therapy units. The advantages of such testing
adduced from clinical trials, preferably randomized and include the rapid availability of results at the location
double blinded. Such reviews should be designed to min- where management decisions are made and an ability to
imize bias, random errors and confounding. Systematic perform the test frequently; both these facilitate rapid
reviews may include meta-analysis – a statistical analy- responses to changes in the patient’s condition. Certain
sis of the results from a series of independent studies sources of error may be eliminated (although care must
designed to produce a single statement about outcome, be taken not to introduce others) and there may be
whether it be the efficacy of a particular treatment or the ­financial savings, for example in relation to the transport
value of a particular diagnostic investigation. of specimens or of patients; however, the cost per test is
Although the number of systematic reviews of diag- frequently much higher for point-of-care tests (PoCTs)
nostic investigations published annually is increasing, than for the central laboratory.
26 Clinical biochemistry

This is not a new concept. In Victorian times, phy- management, possibly through service level agreements
sicians would perform simple pathological tests at the with community practitioners.
bedside and it was the increasing complexity of the
tests that led to the setting up of centralized labora-
tory facilities. Point-of-care testing offers consider- CONCLUSION
able advantages over laboratory-based testing under
some circumstances, as is apparent from the examples Laboratories are obliged to provide a good-quality ser-
cited above. However, it is important that the tests vice, which should be independently assessed and regu-
and instruments are reliable; that only staff who have lated. As quality standards have evolved, so has the need
been trained to use them are allowed to do so; that to include inspection of procedures to cover the entire
the results are subject to appraisal through a quality sample journey, from phlebotomy to receipt of the final
assurance scheme (preferably supervised by a clinical report. Laboratories should review the needs of their
biochemist) and that there is adequate skilled tech- end-user, be it clinician, nurse or patient and try to adapt
nical and analytical back-up, should problems arise. their service accordingly.
Laboratories are expected to be at the forefront of The need to monitor the clinical effectiveness of labora-
these developments and work with the PoCT user to tories to demonstrate continuous quality improvement and
establish a set of quality standard analogous to those better patient outcomes has resulted in the development of
used in the main laboratory. The international stan- clinical quality indicators. These indicators should be used
dard, ISO 22870, sets out the specific standards that to provide laboratories with sufficient data to identify areas
should be met when such testing takes place within a of good practice and areas where further improvement is
hospital environment. necessary and to promote a culture where continuous qual-
There are however, ongoing clinical governance issues ity improvement is integrated into routine activity.
regarding the use of PoCT. Two key areas of concern
are the assimilation of the PoCT result into the patient
record and the need to ensure adherence to rigorous FURTHER READING
testing processes (a complete audit trail). As an example, Barth J. Clinical quality indicators in laboratory medicine. Ann Clin
the latest blood glucose meters available on the market Biochem 2012;49:9–16.
Barth J. Selecting clinical quality indicators for laboratory medicine.
have the capability of connecting to both laboratory and Ann Clin Biochem 2012;49:257–61.
hospital information management systems. These me- Moore AR. Evidence-based clinical biochemistry. Ann Clin Biochem
ters control access by requiring unique identification of 1997;34:3–7.
individual users. Additionally, connected meters can be A concise introduction to this topic.
controlled remotely from the laboratory to ensure that National External Quality Assurance Scheme, UK NEQAS. www.
ukneqas.org.uk.
the meter has had quality control checks prior to mea- Oxford Centre for Evidence-based Medicine. http://www.cebm.net/?0=1162.
surement of patient samples. Remote access to these me- The website of the Oxford Centre for Evidence-based Medicine. This page
ters can also ensure that use by untrained individuals is illustrates the use of likelihood ratios to modify diagnostic probabilities, but
prevented. there are many other useful pages.
Price CP, Christensen RH, editors. Evidence-based laboratory
These developments are excellent examples of the medicine: principles, practice and outcomes. 2nd ed. Washington:
progress that has been made by manufacturers in this AACC Press; 2007.
area. The need to establish good working relationships A series of essays discussing the acquisition and use of evidence for diagnostic
with the manufacturer and colleagues that use the system investigations, clinical audit and the development and evaluation of guidelines.
is vital to ensure success. Randox External Quality Assurance Scheme, RIQAS. www.riqas.com.
Royal College of Pathologists. How to assess the quality of a pathology
As the use of PoCT devices becomes more prevalent, service. London: Royal College of Pathologists; 2011.
regulation will become stricter and it is envisaged that the United Kingdom Accreditation Service. www.ukas.com
laboratory will take a greater role in their provision and Welsh External Quality Assurance Scheme, WEQAS. www.weqas.com