Abstract
The latest laboratory workload figures have been the subject of many a corridor discussion, management meeting or national forum. Most clinical laboratories continue to see 8–10% increases in workload year on year, usually without increases in budgets to match. With the global ‘credit crunch’, it becomes even more relevant that national health service (NHS) organizations deliver evidence-based practice, provide value for money and reduce waste.
Why are there so many requests?
The reasons for this ever-increasing workload are manifold. Certainly technology has had an impact, though it seems that the faster we turn the results around, the more quickly they are needed. It is not just the analytical technology that has changed. Information technology now has bidirectional links and electronic patient requesting/reporting and the speed of data transfer have increased exponentially. Despite this, it feels as though it is always a step behind where it needs to be and can be a double-edged sword. When it works, rapid result availability can give the impression that tests are easy to perform, with consequent effects on increased requesting. When it fails, previous results are unavailable, resulting in increased duplicate testing.
Changes to medical training and proliferation of non-medical advanced multiprofessional practice also have a direct impact on the way in which pathology requests are made, and by whom. There is more protocol and guideline-based requesting, sometimes by people who have relatively little pathology training. This can lead to unnecessary requests for batteries of tests for rare conditions or repeat requests for tests at ridiculous frequencies.
Workload problems can get worse when there is money involved. The albumin-to-creatinine ratio (ACR) is a prime example. In our laboratory, ACR requests were running at around 30 per month in early 2008. When the Scottish Intercollegiate Guidelines Network and National Institute for Health and Clinical Excellence guidance on chronic kidney disease came out in mid- 2008, requests increased to 150 per month. However, following the announcement of the Quality and Outcome Framework indicator that involved ACR in January 2009, request numbers increased rapidly to 1700 per month by August 2009 and are still rising.
Another driver for workload increases is patient power; specifically the increase in patients' ‘knowledge’ and expectations through Internet access, and indirectly due to the fear of litigation and consequent risk avoidance culture. We have all heard examples of doctors requesting tests on patients, just to reassure them that they are doing something. In some cases, practice culture is based on the desire to intervene at all costs and to be medicolegally defensive. If we are to manage demand, we will need to address these underlying causes. This will require close liaison with clinical teams. In my experience, they are equally shocked by some examples of requesting and are only too willing to join forces.
Inappropriate testing
The Carter Report (phase 1), 1 a Department of Health-commissioned independent review of NHS pathology services, states: ‘70–80% of all health-care decisions affecting diagnosis or treatment involve a pathology investigation’. Over 500 million clinical biochemistry tests are performed each year in England alone. However, the Carter Report (phase 2) indicated that unnecessary requesting of pathology tests is widespread. 2 Indeed, Rao et al. 3 stated that 25–40% of such tests are unnecessary. In clinical biochemistry, this proportion is greater (up to 98% in one study), 4,5 indicating that many patients may be suffering needless discomfort, anxiety and inconvenience.
The scale of unnecessary testing can be seen in differences in requesting patterns between GP practices. Even in diseases such as diabetes mellitus where NICE 6,7 and QOF 8 provide guidance on testing intervals, National Pathology Benchmarking Scheme (NPBS) data for glycated haemoglobin (HbA1c) testing show a 15-fold variation in requests per patient on the primary care diabetes register (David Holland, NPBS, personal communication), suggesting that requesting outside this guidance is common. Unlocking the key to this variation and identifying the prevalence of unnecessary requesting would have major implications for reducing patient inconvenience and allow targeting of NHS resources to areas of known patient benefit.
Why should we manage demand?
The drivers for managing demand boil down to finance, quality and patient issues (unfortunately for most laboratories, in that order). The cost of inappropriate requesting includes not only test consumables and reagents, but also additional consultations, treatment and investigations to rule-in or -out further intervention. As pathology services cost the NHS ∼£2.5 bn pa, 1 eliminating inappropriate testing could save the NHS up to £1 bn pa in test costs alone. From an individual trust perspective, reducing internal testing, or the number of referred tests, saves money. Controlling demand (rather than managing it) is therefore very attractive to hospital managers and finance directors. Given that reducing inappropriate testing generally improves quality and patient experience, perhaps it should not just be seen as a management ploy to persuade us to save money.
So how do we manage demand?
The major challenge is how we go about managing demand. Many studies have investigated this, with variable degrees of success. Approaches used fall into the following three categories:
(1) Pre-laboratory. This comprises approaches to educate and engage with requestors regarding testing. It may include something as simple as a regular memo or presentation to the clinical team, formal contributions to training or induction curricular, contributing to guideline/pathway development through to provision of performance data (number of inappropriate requests, total test numbers, costs per test, total costs, and a range of more sophisticated indicators). 9,10 These, however, require continuous monitoring in an ever-changing environment of new guidelines and clinical staff turnover.
A further approach is to limit the test repertoire available to the requestor (or at least to more junior requestors), for example, by redesigning the request form (or electronic patient request screen). 11 Indeed, one of the Healthcare Commission Acute Hospital Portfolio assessment criteria for diagnostic services in 2005 emphasized the importance of this by asking the total number of tests that could be requested if the requestor ticked all the boxes on the form. Use of disease-specific profiles can also limit testing, 12,13 though it must be carefully monitored and, like most demand management approaches, agreed with the clinical team. We have used this method in an admission unit setting with some success. 13 Finally, the withdrawal of outdated tests, an essential component of ongoing service development, can have a similar effect. For example, in line with recent suggestions, 14 we and others have withdrawn faecal occult blood testing.
(2) Within-laboratory. This is probably the easiest approach for the laboratory, but does not stop the initial request. It largely revolves around use of minimum retest interval data to prevent duplicate testing. Minimum retest intervals adopted by many laboratories are based on criteria such as analyte half-lives and analytical variability. For example, for HbA1c, retesting within two months is not warranted since biological changes are dependent on the red cell half-life (∼120 days) and current analytical methods are insufficiently precise to detect the resultant small change in this marker expected in an individual patient within this time period. However, these laboratory-based assessments have often not been validated in clinical situations. It is possible, therefore, that there is a case for more frequent testing based on effects on patient outcomes, quality-of-life and prescribing patterns. Certainly, patient views have not been included in assessments of testing frequency.
Other within-laboratory examples include vetting requests as they arrive. However, these approaches have no impact on preventing patient discomfort, anxiety and inconvenience. It certainly works, but is a poor reflection on the laboratory if used in isolation.
(3) Post-laboratory. A further consideration is to work with clinical teams to review the impact of the test results on the patient pathway. This can comprise: (i) assessment of results not reviewed by the requestor or clinical team, (ii) results reviewed but which had no impact on clinical management and (iii) results reviewed and changed management, but did not affect patient outcome. Assessing these aspects can be practically challenging, as there may be many non-clinical reasons for results not impacting on patient management, but they are nevertheless important.
National drivers
With the publication of the Carter review, phase 2, 2 it is clear that managing demand is a major focus, with much emphasis laid on the potential of benchmarking data. Indeed, the NPBS is running a pilot project to examine ways of providing individual GP practices with data on requesting patterns relative to peer results. Future options include providing duplicate testing frequency data. With the Department of Health backing this project, it will provide a strong impetus for primary care to provide value for money. After all, if one practice can achieve their QOF targets by doing half the number of tests of a neighbouring practice, they may indeed be able to have their cake and eat it.
We are therefore left in no doubt that demand management is an important component of clinical laboratory activity. Indeed, we have a duty of care to ensure the best service for patients and this will include reducing inappropriate testing. The only question that remains is around how this fits with the business model. Not many successful businesses strive to regulate demand for their services. It would be interesting to know how many laboratories are under pressure to manage demand within their trust, but find somewhat less enthusiastic support in managing external requesting. This is perhaps a true test of whether we all work together in a health economy for the benefit of patients?
DECLARATIONS