Clinical quality indicators in laboratory medicine: a survey of current practice in the UK

Abstract

Background

Clinical outcomes are the current paradigm for assessing the performance of health institutions. In contrast, clinical laboratories are assessed on largely operational criteria.

Methods

An invitation to an online questionnaire was sent to all members of the Association for Clinical Biochemistry in the UK and Eire. All data were normalized to the total number of respondents irrespective of how many respondents answered each question.

Results

A total of 335 individuals responded to the survey. A further six responses were discarded as the respondents worked outside clinical laboratories. Overall responses showed that of the selected quality indicators, responses ranged from 7% to 84%. In general, although most laboratories provided an advisory and interpretative service, there was little evaluation and audit of this component of the service. The majority of respondents reported a written critical (alert) list and a hospital point-of-care committee.

Conclusion

This survey has explored clinical quality indicators in the preanalytical, analytical and postanalytical phases of laboratory practice. This should act as a template for laboratory assessment and clinical quality improvement.

Introduction

Clinical laboratories in the UK are regularly inspected by Clinical Pathology Accreditation (CPA; www.cpa-uk.co.uk/) and judged according to its standards. These standards are largely based around the operational delivery of laboratory analyses with relatively few measures of clinical quality such as appropriateness of the service and outcome. This may be due to the difficulty with which such quality indicators can be defined. However, despite this, there has been debate about the appropriateness of the present accreditation system in the current health-care arena, which is now focused on quality outcome measures. ^1,2

Over the past few years, there has been a paradigm shift in the delivery of laboratory medicine to a clinical service measured by outcomes. Laboratory economics demand that testing must be focused on clinical benefits. Patient safety is compromised by tests that are missed or delayed and by tests that are inappropriately requested and then misinterpreted. These areas have been referred to as the ‘pre–pre’ and ‘post–post’ analytical phases and are clearly in the domain of clinical decision makers. ³

A programme to improve quality in laboratory medicine is being driven in the USA by the Division of Laboratory Systems in the Centre for Disease Control. They have established a group to examine clinical quality issues with an aim of producing best practice guidelines. ⁴ To date they have pilot-tested new methods to conduct laboratory medicine quality improvement reviews and are seeking collaborative help to identify good practices. A recent literature review of quality indicators in laboratory medicine has identified an evidence base for only 14 criteria. ⁵

The practice of laboratory medicine in the UK and many other countries includes a more proactive clinical service. We have therefore explored a list of possible quality indicators and surveyed members of the Association for Clinical Biochemistry in the UK and Eire to determine whether these quality indicators are appropriate and to what level they are currently being met.

Methods

A series of focus groups of laboratory professionals was held to try to identify clinical practices in laboratory medicine that could be used as quality indicators. These were incorporated into a questionnaire that was converted into an online format. An invitation to complete this questionnaire was sent in January 2010 to all members of the Association for Clinical Biochemistry in the UK and Eire. A four-week time window for completion was available with closure on 22 February 2010.

Data were analysed using Analyse-it^® version 2.08 add-in package for Microsoft Excel (Leeds, UK; www.analyse-it.com). All data were normalized to the total number of respondents irrespective of how many respondents answered each question.

Results

A total of 335 individuals responded to the survey. A further six responses were discarded as the respondents worked outside clinical laboratories. The results of the survey are presented in Tables 1–5. The data are largely qualitative as it has not been possible to ensure that only one response per laboratory has been included nor what grade each respondent holds. The response cited in Tables 1–5 represents the number of respondents to each question as a percentage of the total number of respondents (335).

Table 1

Preanalytical

Number of laboratories that have surveyed patients about phlebotomy	27.6%
Number of laboratories that have patient information sheets	77.3%
There is a dedicated person/rota available for 24/7 advice	78.5% (94% response)
Proportion on your 24/7 on call rota have FRCPath	68% (79% response)
Number of laboratories who measure response time for 24/7 enquiries	7.1%
Laboratory has an electronic handbook	83.9%
There is advice on test interpretation	66%
There is information regarding critical differences (reference change values) between consecutive results	22%
Number of laboratories that comply with NICE recommendations	39%
There is written evidence of an annual review of the regular development and maintenance of your test repertoire	51%
Number of laboratories that are compliant with any national guidance, e.g. IFCC, NICE, Royal Colleges, National Societies, if yes, please list	50%

NICE, National Institute for Health and Clinical Excellence; IFCC, International Federation of Clinical Chemistry

Table 2

Analytical

Turnaround times are discussed with your clinicians	75%
Internal evaluations of new methods made prior to implementation	84%
Assay precision determined at critical concentrations for any assays	55%
Reference ranges are determined in your laboratory on locally sourced samples	42%
There is a trust point-of-care committee	67%

Table 3

Postanalytical

There is a process for demand management	43%
Laboratory provides help and advice in interpreting clinical laboratory data	80%
There is an audit of the effect of added interpretative comments	16%
There is a written critical limits (alert) list	58%
There is a record of the number of calls/emails/letters received for clinical advice	18%
Proportion of requests with additional tests added by laboratory	13% response
There are automatic reflex tests via reporting rules	57%

Table 4

Evaluation, research and development

Department attends multidisciplinary team meetings	56%
There is an audit of erroneous reports	43%
Collaborative clinical audits were made with clinical colleagues in last 12 months	45% response

Table 5

Education and training

Undergraduate lectures/tutorials delivered by laboratory staff	39%
Postgraduate lectures/tutorials delivered by laboratory staff	43%
GP surgery lectures/tutorials delivered by laboratory staff	24%
Public lectures/visits to your laboratory arranged by laboratory staff	33%

GP, general practitioner

Discussion

The external quality of clinical laboratories is currently assessed by a combination of CPA (UK) that assesses largely operational metrics and by external quality assessment schemes that assess analytical performance. The assessment of the clinical appropriateness and advisory components of a laboratory service would represent a third component of quality assessment, but this aspect is currently not systematically evaluated. This is largely because the range and definition of appropriate quality metrics has not yet been developed.

There is a paucity of literature on quality indicators. This has been recently reviewed by Shahangian and Snyder. ⁵ They recognized that there was a ‘…considerable challenge in identifying, defining, and, ultimately, implementing indicators that cover the … laboratory testing process in general and specific to different diseases and conditions’. This may explain why they found an evidence base for only 14 criteria across the whole range of pathology. These were: test order appropriateness, wristband id errors, patient satisfaction with phlebotomy, specimen quality (adequacy, contamination, information error), proficiency performance, cervical cytology–histopathology mismatch, availability of inpatient results, corrected laboratory reports, critical value reporting, turnaround time, clinician satisfaction and follow-up of abnormal cervical cytology.

What did we find

Our survey shows a high proportion of laboratories providing an advisory and interpretative service to their users both in and out of hours. This was provided by telephone and written reports. There was, however, little evaluation and audit of this component of the service. The majority of respondents reported a written critical (alert) list of results that need urgent communication to responsible clinicians and that their hospital had a point-of-care committee. Only half of the respondents reported compliance with national guidelines. However, since there are many guidelines and the laboratory components are buried in long reports, it is possible that a different proportion is actually delivering on those guidelines. This findings contrast with a survey performed on US hospital laboratories. This noted that US laboratories were good at incident investigations and performing patient and physician satisfaction surveys but poor at clinical guidelines, studying test utilization and assessing test interpretation by physicians. ⁶ This exposes a curious paradox where much of the day-to-day time is spent focusing on internal activities almost to the exclusion of the exterior of the laboratories except when a clinician calls in for advice, whereas the external health-care community would benefit immensely from more time and attention from laboratory professionals.

It could be argued that most of the indicators surveyed in this report are still process indicators that have a relationship to the quality of the clinical service provided. These are important to health-care outcomes rather than being indicators of clinical quality.

What should we do next

One of the problems of performing an exploratory survey such as this is its essentially qualitative nature. This is demonstrated by the response to the question regarding the provision of interpretative help. Clinical biochemistry prides itself on this aspect of the service. It is surprising, therefore, that there was not a 100% response regarding the provision of interpretative support; this may be due to answers by trainees and technical staff who are ACB members. Should quality indicators become part of the measure of clinical laboratories, it would be necessary to ensure a single respondent and many of the questions used would need to be converted to require a numerical rather than quantitative answer. Moreover, the quality indicators would need to be interpreted in the light of clinical requirement. It might be appropriate to assess turnaround times only on those analytes that are required for rapid clinical decisions. Moreover, different emphasis might be appropriate for specialist laboratories where the degree of interpretative advice is likely to be greater.

Guidelines are now being produced by all national societies and professional bodies. Many have laboratory tests as a component of the management of patients. The provision of these tests in a timely manner clearly has the potential to permit clinical decisions and therefore clinical outcomes. These tests and their interpretation could be developed into a list of quality indicators.

A quality clinical laboratory service might be simply described as performing the right test on the right person at the right time and interpreting that test correctly. This would be reflected as a service that provides quality in the preanalytical, analytical and postanalytical phases. The preanalytical phase might include the provision of advice on the right test and timing for that test as well as the correct sample and tube for collection and provision of the appropriate range of tests. Indicators in the analytical phase might include choice of the most appropriate methods, evaluation of their performance at critical decision-making values and protocols for evaluating analytical errors. The postanalytical phase might include interpretative advice, ^7,8 reflective testing ⁹ and telephoned critical results. ¹⁰

In summary, we have surveyed clinical biochemists to determine whether quality indicators can be used as a measure of clinical quality. This could then be used in addition to the current accreditation system that is largely a measure of operational performance.

DECLARATIONS

Competing interests: None.

Funding: None.

Ethical approval: Not applicable.

Guarantor: JHB.

Contributorship: JHB conceived, performed the study and wrote the paper.

Acknowledgements: I am extremely grateful to all those colleagues who have taken part in focus groups that have helped to develop the quality indicators measures and to pilot the questionnaire and to Dr David Burnett for his helpful comments. I also thank Nic Law for building the online questionnaire.