New diagnostic tests: more harm than good
BMJ 2017; 358 doi: https://doi.org/10.1136/bmj.j3314 (Published 18 July 2017) Cite this as: BMJ 2017;358:j3314
All 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.
In only 10 minutes, bees can be trained to detect early several forms of cancer in human patients.
New biomarker diagnostic tests do not necessarily have to be extremely complex and expensive, and thus not cost-effective, as this analysis implies.
References
https://www.dezeen.com/2013/11/20/honey-bees-can-be-trained-to-detect-ca...
https://www.theguardian.com/science/2013/nov/12/honey-bees-trained-detec...
Competing interests: No competing interests
It is good to read the very interesting and relevant article "New diagnostic tests: more harm than good".
When existing diagnostic tests are
1. Not precise and their results are not accurate
2. Associated with diagnostic test related adverse effects in patients
3. Not cost effective
then the need for "New Diagnostic tests" arises to overcome the limitations associated with existing diagnostic tests.
New diagnostic tests should be of good efficacy (accurate and precise), safe in patients (not associated with adverse effects) and provide good quality of life, as well as being cost effective [1, 2].
There must be logical and rational use of new diagnostic tests, otherwise there is a waste of time, effort and money [3].
To optimize the use of "New diagnostic tests", clinicians' responsibility cannot be ignored [4].
Regards,
References:
1. http://www.bmj.com/content/351/bmj.h5527
2. https://www.ncbi.nlm.nih.gov/pubmed/7490184
3. https://www.ncbi.nlm.nih.gov/pubmed/7490184
4. http://accessmedicine.mhmedical.com/content.aspx?bookid=496§ionid=41...
Competing interests: No competing interests
I am not sure that all is bad or good. Consider the case of a noninvasive colon cancer at-home test, for instance. Is it good or bad ? I guess it depends on who you ask. A website for such a test lists that 13% of individuals without cancer or a precancerous state can test positive. That does not mean that the false positive rate is high but rather that a precancerous lesion could be picked up on colonoscopy, for example. But what of those whose lives were saved by this simple test ? Is it a worthless test ?
Secondly, it is not cut and dry in medicine, either. For instance, according to the CDC, colorectal screening tests are recommended, and this is seconded by the US Preventive Services Task Force, that adults between 50 to75 ought to be screened. However, while the data could be all-encompassing is this true for all ethnic groups as in Asians, or vegetarians, for example ? I am not sure they have been followed as diligently, as Caucasians or blacks have been. Therefore, the debate is open-ended.
Finally, smartphone based apps have made a difference in the lives of the man on the street. An iphone app touts atrial fibrillation detection at 0.975 SPECIFICITY. Do some tests miss the mark and it is all hype? Probably so. But time and field testing by those who eventually get to use them will either weed them out or give them a thumbs up. Doctors are lagging behind, might I add, in this digital and mobile revolution. Whoever could imagine a few decades ago that Facebook would have links to organ donor sites and one could find a match for a kidney or liver scouring the Facebook pages ? That's technology for you. And it will only increase whether we like it or not.
Jagannadha R Avasarala, MD, PhD
Greenville, SC, USA
Competing interests: No competing interests
A major issue with many new (and some old) diagnostic tests is what they purport to diagnose. In cancer diagnostics, a major issue is our understanding of the term, “Cancer”. For laymen (and unfortunately even many doctors), cancer is by definition a life-threatening disease but a significant proportion of the cancers diagnosed by screening tests are likely to be intrinsically harmless and would not have caused any problems for that particular patient in his/her lifetime. Hence, the only risk to the patient would be related to diagnosis and treatment.
Cancer is not a single disease but a group of diseases with very different risks of adverse outcome. Even phrases such as “breast cancer” and “thyroid cancer” as each includes a number of very different diseases ranging from the relatively harmless to the deadly. The spectrum of thyroid cancer ranges from “papillary microcarcinoma”, which generally has little effect on life expectancy to “anaplastic thyroid cancer”, which has a dismal prognosis with most patients surviving no more than a few weeks. Most newer diagnostic tests do not clearly indicate which cancers they diagnose. Early diagnosis of clinically insignificant cancers would certainly do more harm than good.
An irony of early cancer diagnosis (and cancer screening programmes) is that worse the test (or screening programme), the higher the satisfaction level among all involved! A test with significant overdiagnosis of clinically insignificant cancers would result in large numbers of satisfied patients (who think they have had their lives saved), doctors (who think they have saved lives) and politicians, who crow about increased cancer survival rates that in reality are only the result of the excellent prognosis of tumours that would never have caused any harm to the patient. However, the sad reality is that these patients would have suffered psychological and physical harm related to unnecessary therapy, while waste of precious healthcare resources would have adversely affected the care of other patients (cancer and non-cancer).
Competing interests: No competing interests
The problem is that tests are assessed at present merely by using the indices of sensitivity and specificity, which were developed during WWII for radar detection. The STARD and GRADE guidelines only recommend these indices of performance for screening tests; they are also the only indices used by regulatory bodies (e.g. NICE). This very limited assessment makes it is easy for new diagnostic tests to be introduced and used inappropriately. Sensitivities and specificities may be easy for epidemiologists to understand but they are not appropriate for diagnostic reasoning and cannot address the problems of over-investigation, over-diagnosis and over-treatment.
It is important to know the differential diagnosis of each of a new test’s possible results and their estimated probabilities. We also need to know ratios of sensitivities and false negative rates with respect to relevant differential diagnoses [1]. It is failure to provide this information that leads to wasteful over-investigation. It is also important to assess a new test’s ability to form diagnostic and treatment indication criteria [1]. It is failure to do this that leads to wasteful and damaging over-diagnosis and over-treatment.
Some years ago a BMJ editor commented that evidence-based diagnosis was in the dark ages [2]. It still remains in the dark ages as evidenced by widespread over-investigation, over-diagnosis and over-treatment. Another BMJ editor laments this again this week over a decade later [3]. The only ethical defence against such harm to patients is research based on a detailed understanding of the diagnostic process [1].
References
1. Llewelyn H, Ang AH, Lewis K, Abdullah A. The Oxford Handbook of Clinical Diagnosis, 3rd edition. Oxford University Press, Oxford, 2014, pp 615 – 664. http://oxfordmedicine.com/view/10.1093/med/9780199679867.001.0001/med-97...
2. Delamothe T. Editor's Choice - Diagnosis—the next frontier. BMJ 2006; 333:0-f doi: https://doi.org/10.1136/bmj.333.7565.0-f
3. Hurley R. Editor's Choice - Defenders against over-diagnosis. BMJ 2017; 358:j3487. https://doi.org/10.1136/bmj.j3487
Competing interests: No competing interests
In neurological diseases it may be act as a double edged weapon. Sometimes knowing or unknowing the facts about neurological diseases may be stressful or harmful to the patients.
Competing interests: No competing interests
I agree with the previous respondent.
I doubt that the new wave of highly specific and often very expensive diagnostic testing and targeted treatment (such as *zumabs), using newly acquired knowledge of the human genome, will give as much benefit as acceptance by society of a need to change our behaviour.
Alcohol, Tobacco, and Sugar (including HFCS) in large overdose cause monstrous levels of morbidity and mortality . Society is just now beginning to wake up to the paradigm that Sugar is "Pure, White, and Deadly" as Professor John Yudkin proposed in his book over 40 years ago!
Foods that are labelled "Lite" may well contain 10% less fat, but if the protein content is unchanged then, according to the "Law of Conservation of Mass", the carbohydrate content has increased 10%. The cost to modern society is enormous in every respect.
"If you don't change direction, it may come to pass that, you may end up there, where you are heading" - ancient Chinese proverb, Lao Tzu
Competing interests: No competing interests
Microbiology is another area where there have been huge shifts recently. Where much of the technology was effectively a 19th Century process of microscopy and growing organisms on different media, many labs have shifted, for a vast amount of the work, to molecular technologies such as polymerase chain reaction (PCR) testing. With the economies of scale now available, this is often cheaper.
Eventually it may be a vast improvement. However we don't fully understand what to make of the results at present. The probes used to identify toxigenic Escherichia coli, for example, may also identify other organisms; and the yield of positive results is much higher. The test is more sensitive? Does this mean that we were previously missing many results of real clinical and public health signficance? Or have we traded off greater sensitivity for a cheaper test with less specificity?
Particularly when taking into account the public health actions that may follow - work for PHE and local authority environmental health departments, children excluded from school or nursery etc. - this can mean considerable cost (albeit not all of it directly to the NHS).
I suspect that, over time, we will learn how to respond most appropriately; but for now - and for some time to come - this change is not always altogether helpful.
Competing interests: No competing interests
Diagnosis is the sine qua non of healthcare. Without a diagnosis, doctors don't know how to help patients. But there are many patients who can't be diagnosed or helped. So let's enhance diagnostics and therapeutics with commonsense holistics by avoiding all toxins and addictions, because both are common, covert causes of multimorbidity and mortality. Intellectual but ineffectual, and oblivious to the obvious, diagnostics and therapeutics are contrived and divisive, but logical holistics are realistic and pragmatic, and should be automatic.
Competing interests: No competing interests
Re: Not just new diagnostic tests, all tests can sometimes cause more harm than good
‘All screening programmes do harm; some also do good as well, and, of these, some do more good than harm at a reasonable cost.’1
This provocative statement is arguably true of any test, not just screening, and not just new diagnostic tests. The challenge for clinicians is to share decisions with patients; choosing the right test, for the right patient, at the right time.
Inflammatory marker tests such as C-reactive protein, are simple, inexpensive and commonly used in both primary and secondary care.2 When we interviewed GPs about their use of inflammatory marker blood tests in primary care, all doctors mentioned potential harms as well as benefits of testing.3 These could be divided into several categories; harms of the test itself, workload and financial costs, medicalisation of a patients’ problem, patient anxiety, downstream costs and harms of overdiagnosis. Harms from the test itself are often overlooked and include cost of patients’ time off work, needle phobia, bruising or vesovegal syncopy. One GP described the psychological cost; “there is always this pass/fail thing you know… because it’s secretly looking inside your body, there might be something wrong there, so yeah I think it’s not something one should do lightly.” In UK Primary Care, rising workload is a significant concern,4 and many GPs in our study perceived that time spent dealing with pathology results was often inefficient or wasteful. Medicalisation may occur when tests cause ‘somatic fixation’5, with patients or clinicians focusing only on biomedical aspects of complex illness. Patient anxiety is perhaps the hardest to capture but most important harm of unnecessary testing. Anecdotal stories can be powerful (eg ‘Fran’s story’ https://fabnhsstuff.net/2016/02/10/pathology-optimisation/) but the overall cost is hard to estimate. Further research into the benefits and harms of diagnostic tests is needed, not just for new diagnostic tests, but also for many commonly used 'routine' tests.
Overuse of diagnostic tests may reflect a societal culture of increasing risk aversion and rising medical litigation, and also a culture within medicine which ‘evinces a deep-rooted unwillingness to acknowledge and embrace uncertainty’.6 Technological advances and new tests provide ever more information, yet without wisdom7 we risk making well people sick, rather than making sick people people well.
References
1. Gray JAM, Patnick J, Blanks RG. Maximising benefit and minimising harm of screening. BMJ : British Medical Journal 2008;336(7642):480-83. doi: 10.1136/bmj.39470.643218.94
2. Watson J, Round A, Hamilton W. Raised inflammatory markers. BMJ 2012;344:e454. doi: 10.1136/bmj.e454 [published Online First: 2012/02/07]
3. Watson J, de Salis I, Hamilton W, et al. 'I'm fishing really' - inflammatory marker testing in primary care: a qualitative study. Br J Gen Pract 2016;66(644):e200-6. doi: 10.3399/bjgp16X683857
4. Hobbs FDR, Bankhead C, Mukhtar T, et al. Clinical workload in UK primary care: a retrospective analysis of 100 million consultations in England, 2007–14. The Lancet;387(10035):2323-30. doi: 10.1016/S0140-6736(16)00620-6
5. Biderman A, Yeheskel A, Herman J. Somatic fixation: the harm of healing. Social Science & Medicine 2003;56(5):1135-38. doi: http://dx.doi.org/10.1016/S0277-9536(02)00108-9
6. Simpkin AL, Schwartzstein RM. Tolerating Uncertainty — The Next Medical Revolution? New England Journal of Medicine 2016;375(18):1713-15. doi: doi:10.1056/NEJMp1606402
7. Heath I. Information without wisdom. BMJ 2017;358 doi: 10.1136/bmj.j3203
Competing interests: No competing interests