Should we abandon routine blood tests?
BMJ 2017; 357 doi: https://doi.org/10.1136/bmj.j2091 (Published 03 May 2017) Cite this as: BMJ 2017;357:j2091All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
BMJ Response
I applaud Faulkner and colleagues for a well-written article which is relevant to various specialties within the NHS and has generated immense debate. The authors suggest that the status quo of ordering routine blood tests in acute departments should be challenged, citing these as being akin to a clinician’s “psychological support blanket”, and potential for cost savings. I acknowledge that the acute profession has geared towards “knee-jerk” diagnostics, for instance, it has become standard practice for patients with medical ailments to be referred from the ED with the usual gamut of blood tests, ECG, chest X-ray and urinalysis.
As a medical registrar leading a busy acute take who is faced with the responsibility to admit and discharge patients overnight, I disagree with the concept of not having this information at our fingertips. As clinical assessments and impressions can vary between junior doctors, investigations are often the only objective information available. Using these tests in conjunction with the clinical situation supports our Bayesian model for decision making, e.g. the negative predictive value of troponins and D-Dimer for atypical chest pain, and normal blood panel in an elderly patient who is “generally unwell”, affects our clinical impression. In a previous study of routine blood tests conducted in the general medical setting, 12% were abnormal, but only 0.5% led to changes to patient treatment.[1] However, this did not explore the impact of the tests on discharge-related decision making.
Spurred perhaps by an increasingly internet savvy and litigious society, it is healthy and insightful for clinicians to doubt their diagnosis based on history and examination alone. After all, this is the rationale for having differential diagnoses, especially when managing a progressively older demographic with increasing co-morbidity, polypharmacy, and with atypical presentations of serious conditions. It is arguable that in an acute medical setting, a standard panel of blood tests can be helpful, as most medical presentations can theoretically be associated with haematological or biochemical abnormalities. This concept was recently demonstrated in a study of over 4000 Danish medical patients aged 65,[2] which correlated the proportion of abnormal routine blood tests on admission with poorer survival, after adjusting for age and concomitant diagnoses. The authors went as far as proposing the use of routine blood tests as a biomarker of frailty.
In defence of acute physicians, using terms such as “routine bloods” detract from their real purpose or acting as a basic screening tool for potentially serious pathology. In a hospital environment where resources are available at our fingertips, failing to detect malignancy-related anaemia, renal impairment, or liver disease after a medical admission, regardless of cause, may incur longstanding repercussions. Although, I do agree that attention to pathology informatics is required to avoid unnecessary repetition of blood tests, I believe that a routine blood panel should be considered part of holistic medicine and not just as a “psychological support blanket”.
References
1. Hubbell FA, Frye EB, Akin BV et al. Routine admission laboratory testing for general medical patients. Med Care. 1988 Jun;26(6):619-30.
2. Klausen HH, Petersen J, Bandholm T et al. Association between routine laboratory tests and long-term mortality among acutely admitted older medical patients: a cohort study, BMC Geriatrics 2014; 17:67
Competing interests: No competing interests
I completely agree with Faulkner & colleagues. As lead of our preoperative assessment unit I am often shown abnormal blood results nobody knows what to do with, with the plaintive request "can we pass the patient fit?" I try to reinforce the principle that we shouldn't order tests unless we understand the limitations of the test itself (i.e. Sensitivity and specificity) as well as something about the condition we are testing for, particularly prevalence. Otherwise we won't be able to interpret the results in any meaningful way and patients are subject to long and unnecessary delays.
This is well illustrated by an old but still relevant paper from the NEJM (1) that I sometimes quote to trainees. Firstly they looked at 25 routine tests done on healthy volunteers- the chance of all coming back "normal" was only 28%. So doing tests assuming they will be normal is asking for trouble. They then went on to ask 60 medics at Harvard medical schools what the chance was of a particular positive test actually meaning the patient has the disease, given a 5% false positive rate and a known disease prevalence of 1 in 1000. Only 8 got it right. The answer is just under 2% but most people immediately assume 95% - what we are really looking at here of course is positive predictive value. As medics we often assume that if a patient has an abnormal test result there is something wrong with them but that is relatively rarely the case.
The golden rule of preop assessment, where we are usually testing well patients, is NOT to do routine tests unless we know what an abnormal result means ( i.e. something about its PPV) and it will influence management. An abnormal clotting screen ( my bete noire) in an asymptotic patient has a Positive Predictive value of at best 1 in 1000. So don't do it.
1 Interpretation by physicians of laboratory results. Casscells W et al N Eng J Med 1978 Nov 2;299(18):999-1001
Competing interests: No competing interests
I agree with Faulkner, Reidy and McGowan that reducing cost and unwarranted variability in care are priorities for any modern healthcare system. However, their argument against routine blood tests seems to assume that all such tests fall into two categories: those that of themselves change management, and those that do not. Indeed, they argue against tests that serve only to “give…more information”.
Medicine involves inductive reasoning, where healthcare professionals gain information from test results and interventions and then infer the causes of patients’ problems from this information. One of the great consolations of the inductive process is that by gaining the information necessary to reason, we learn more about patterns and structures in this information and how these patterns guide our inferences. As a result, reducing the amount of information we gather, by reducing the blood tests we take, may reduce our ability to reason about the problems of our patients.
Competing interests: No competing interests
Does anyone know, is there any evidence to show, what patients think about this issue? Is there perhaps a widespread expectation that complete blood tests help to rule out some conditions, and are therefore worthwhile? Or would patients prefer to have only specific tests for suspected conditions? If the second, then repeated tests to discover one thing is within the normal range, before moving on to the next possibility would not be welcome, I suggest, from personal experience. And I doubt if that is really cheaper. Any attempt to 'ration' has to be rational.
Competing interests: No competing interests
As Dr Faulkner points out, routine blood test are a waste of resources. Not only for the patients attending the hospital but also for those who come into the Primary care office asking for their annual check up. This costume is reinforced by some private health companies who try to sell their products. Public health authorities should make public campaigns informing general public about the utility of those unnecessary tests.
Regards
Competing interests: No competing interests
Dear Editor,
I would like to congratulate Faulkner et al on a superb article that highlights a significant issue. The daily CRP in postoperative elective ICU patients has always baffled me. There is clearly a balance to be struck between considered decision making and efficiency. Much of the culture of ordering a 'panel of everything' has been driven by the 4 hour wait in the ED and a drive to help make faster decisions. In the intensive care unit, nurses are authorised to request blood tests and again this often greatly improves throughput, albeit at the cost of some unnecessary tests.
I would like to pick up on one point made by the authors:
"Like modern day imaging requests, every investigation required justification"
I would argue that this is no longer the case, whilst I occasionally encounter an obstructive radiologist the overwhelming approach seems to be that they have given up arguing and that scans often occur at the drop of a hat. One specific example is in major trauma where an entire body CT the so called 'pan-scan' is now the norm. Although there is reasonable evidence for this practice they still miss injuries (1) and I rarely hear about the radiation risk any more. However my issue is less with the 'pan-scan' in trauma but more with the quest for the subsegmental pulmonary embolus. In an era dominated by defensive practice is this the epitome of over investigation in modern clinical medicine? I'm not sure if there is any UK data on the number of negative CTPAs are done each year but my suspicion is that it is high. There is a published acceptable positive rate of 15-37% from the royal college of radiologists and I accept there is a high pick up rate of other conditions. There is however a very poor use of pre test probability scoring systems (3) and if these simple scoring systems were applied more rigidly unnecessary scans would be avoided.
1. Stengel D1, Ottersbach C, Matthes G, et al. Accuracy of single-pass whole-body computed tomography for detection of injuries in patients with major blunt trauma. CMAJ. 2012 May 15;184(8):869-76.
2.Royal College of Radiologists. Appropriateness of Usage of Computed Tomography Pulmonary Angiography (CTPA) Investigation of Suspected Pulmonary Embolism. [Cited 05 May 2017.] Available from URL: https://www.rcr.ac.uk/audit/appropriateness-usage-computed-tomography-pu....
3. Crichlow A1, Cuker A, Mills AM. Overuse of computed tomography pulmonary angiography in the evaluation of patients with suspected pulmonary embolism in the emergency department. Acad Emerg Med. 2012 Nov;19(11):1219-26.
Competing interests: No competing interests
Safety-netting should be routine practice. Asher's criteria for designing a good safety net not only include the question " will the result change my management? but also " is there a better test/strategy?". Blood tests in UK are generally cost-effective, whilst the nit-picking rejection of lab requests can be greatly improved on a systematic basis.
Competing interests: No competing interests
Re: Should we abandon routine blood tests?
Only ordering blood tests when indicated presupposes that staff assessing the patient have adequate time, experience and resources to make accurate judgements reliably.
Often when faced with complexity, uncertainty, limited time and competing clinical demands, and with a desire to avoid subsequent delay, sub-standard care or blame, the more pragmatic approach is probably to order all 'routine investigations'.
Better thought thorough than 'proved incompetent'.
A detailed clinical history and examination of the patient, carefully documented and followed by diagnostic deliberation leading to a decision regarding the necessity for specific tests might cut the cost of investigations, however, it would almost certainly be much more expensive in terms of clinical expertise & time.
The pragmatic (and prudent) approach in an era of frailty, comorbidity, complexity, clinical urgency, complaints and chronic understaffing may well be to simply tick the boxes and do the tests.
Competing interests: No competing interests