- Our online table of contents is updated at least twice each day. Read all articles published in the last 7 days.
- Keep up to date with cardiology: Access the latest cardiovascular medicine resources from across BMJ Group.
- The BMJ iPad app brings you the best of print and online, including live links to the latest news, blogs, video, and podcasts. Get the BMJ iPad app.
-
OPEN ACCESS: All research articles are freely available online, with no word limit. Find out more about the BMJ's open access policy. Submit your paper.
Recent rapid responses
Rapid responses are electronic letters to the editor. They enable our users to debate issues raised in articles published on bmj.com. Although a selection of rapid responses will be included as edited readers' letters in the weekly print issue of the BMJ, their first appearance online means that they are published articles. If you need the url (web address) of an individual response, perhaps for citation purposes, simply click on the response headline and copy the url from the browser window.
Displaying 1-10 out of 10 published
3 July 2012
Des Spence’s claim that attempts to diagnose disease earlier amount to “bad screening and the worst science” is a sweeping generalisation that is false in numerous situations. Surely he has completely lost the plot! Early diagnosis is what the public expects of its doctors and is the basis of the practice of medicine, with the clinical identification of features of putative disease leading to special investigations. Who would ever congratulate a doctor for late diagnosis? There are innumerable instances when late diagnosis is associated with adverse outcomes, to which our defence organisations regularly testify. HIV infection, tuberculosis, skin cancers, type 1 diabetes are just a few cases that spring to mind. Even in regard to the recent national campaign to improve early diagnosis of lung cancer, which seems to have stimulated his article, the statement is premature and grossly nihilistic. Several statements in support of his argument are factually incorrect: having symptoms does not imply incurable disease in lung cancer; there is no doubt that cough is a symptom of this disease -and it is a key one because of its particular association with better outcome;1 Hamilton has studied and reported on the predictive value of symptoms in lung cancer.2
More seriously, Spence has conflated screening with early diagnosis in order to damn the latter. Early identification of symptomatic disease is an entirely different proposition to the screening of asymptomatic patients. In the case of lung cancer, screening (both with plain radiography and computed tomography) is bedevilled by the serious problem of overdiagnosis, i.e. the identification of clinically silent cases of disease that would not cause problems in the subject’s lifetime. Therefore, diagnosis currently rests on the identification of symptoms, which do serve to imply “real” disease.
Moving on to address lung cancer specifically, is Dr Spence is suggesting we ought not to investigate someone who has haemoptysis, or perhaps wait for even more specific features to emerge, with the patient becoming frail and ill, when the diagnosis is easy, but its value to the patient so much less? The question at hand is how hard we should try. For example, persistent haemoptysis and stridor in a cachectic older smoker will very likely reveal lung cancer that is incurable. By investigating more patients with lower probabilities of cancer being present, when patients remain fit and well, with less specific, common but potentially sinister symptoms that might not be severely troubling to them, such as a new or altered cough, the hit rate will be lower but the chance finding curable disease that much higher. This message has been accepted by the public and doctors in relation to other cancers. For example, every woman knows that she should have a lump in the breast investigated even if she is otherwise well. Why is it that almost uniquely, clinicians seem only to think of lung cancer only when the patient appears to be unwell? In this way, the poor outcomes of lung cancer become a self-fulfilling prophecy.
UK lung cancer patients have a substantially poorer outcome than in most other developed economies in Western Europe and North America,3 which implies that the outcomes we see are not inherent in the disease biology, but relate to other factors, that may be modifiable. Because UK patients present with more advanced disease, an important issue appears to be the late identification of patients in primary care. This is may reflect public ignorance of the symptoms of lung cancer and a reluctance by UK GPs to comply with NICE guidance on when to refer patients with red flag symptoms for a chest X-ray. This is to say we should try harder.
Whilst the primary objective of early diagnosis initiatives is to achieve a stage shift, thereby increasing cure rates, there are other benefits to early diagnosis. Depressingly frequently, patients present for the first time to secondary care with extensive disease and poor performance status, often as medical emergencies, and have very poor outcomes. These patients frequently die in hospital, robbing them and their relatives of the opportunity properly to make preparations for their death. When patients are diagnosed only when they have become ill and frail, opportunities for treatments with proven palliative and life-prolonging value, including palliative care, chemotherapy, radiotherapy and targeted therapies, are missed. Whilst Des Spence is arguing against attempting early diagnosis, is he really arguing for this?
The Doncaster study4 sought to address the public ignorance and the poor observance of NICE guidance on when to do a chest X-ray in primary care, head on. Public awareness was raised and there was a significant 26% increase in the number of chest X-rays done. Due to small numbers, the positive effects seen on number of lung cancers diagnosed failed to reach statistical significance and the study was not powered to show an improvement in the number of patients identified with curable disease. Dr Spence is quite incorrect though to say the study showed “no impact on the presentation of treatable lung cancer.” In a follow up study, we have again found beneficial trends, including a 10% increase in the number of cases found (implying catch-up and that earlier diagnosis was occurring) and a 50% increase in the number of patients diagnosed with stage 1 disease.
I believe the national early diagnosis campaign therefore represents a bold and exciting experiment to test the hypothesis that increased awareness by the public and their doctors can lead to earlier diagnosis and improve disease outcomes, at least up to international standards.
References
1. Buccheri G,.Ferrigno D. Lung cancer: clinical presentation and specialist referral time. European Respiratory Journal 2004;24:898-904.
2. Hamilton W, Peters TJ, Round A, Sharp D. What are the clinical features of lung cancer before the diagnosis is made? A population based case-control study. Thorax 2005;60:1059-65
3. Coleman MP, Forman D, Bryant H, Butler J, Rachet B, Maringe C et al. Cancer survival in Australia, Canada, Denmark, Norway, Sweden, and the UK, 1995-2007 (the International Cancer Benchmarking Partnership): an analysis of population-based cancer registry data. Lancet 2011;377:127-38.
4. Athey VL, Suckling RJ, Tod AM, Walters SJ, Rogers TK. Early diagnosis of lung cancer: evaluation of a community-based social marketing intervention. Thorax 2012;67:412-417.
Competing interests: None declared
Doncaster Royal Infirmary, Armthorpe Road Doncaster DN2 5LT
Click to like:
+2 July 2012
Dear Dr Spence,
Thanks for your reply. I’m quite sure you try to get your facts right and read widely. Even so, you wrote that the early cancer diagnosis field ‘generally has no intellectual framework, rigour, research or systematic reviews’. Debbie Sharp and I pointed out several systematic reviews and strands of research in our response. It might have been a better start to begin with acknowledgement that – at least on this point – you were wrong.
But we don’t want a fight: we enjoy your taking on the ‘received wisdoms’ in medicine – some of which truly are lacking in intellectual framework, rigour, research or systematic reviews.
So, let’s agree where we disagree. We made no claims for the efficacy or value of awareness campaigns or of screening (check our letter – we didn’t) so your adding them to your rejoinder is irrelevant.
Let me put my credo down so we can agree the terms of the debate at least: here it is
I believe expediting the diagnosis of cancer in symptomatic patients is beneficial overall.
You disagree. That’s fine. Of course all scientific research has flaws: that’s part of life, though I think you rather misread my ovarian paper and the Abdel-Rahman one. To me the body of scientific evidence for expediting symptomatic diagnosis paints is clearly in favour, and I hope to continue producing research aimed entirely at helping us GPs with our difficult choices in whom to investigate (and equally importantly whom not to investigate).
Actually, Des, shall we ask the BMJ to put up an e-poll on that one sentence credo?
Willie Hamilton
PS Debbie Sharp is away
Competing interests: As above
peninsula College of Medicine & Dentistry, Veysey Building, Exeter EX2 4SG
Click to like:
+28 June 2012
A strategy of 'I am not going to strive for early diagnosis because I do not believe in it' is unlikely to be popular with patients. Moreover it is likely to be unsafe as the commonest source of error and litigation in primary is missed or delayed diagnosis (1)
Patients expect their doctor to evaluate symptoms and strive for a diagnosis where possible, or to define problems where diagnosis is not possible (which is often in general practice) and to arrive at a shared understanding of the problem. Managing clinical risk in the consultation is an essential feature of good general practice
1. http://www.medicine.manchester.ac.uk/primarycare/npcrdc-archive/archive/... (accessed June 27th 2012)
Competing interests: I was part of the NPSA work on cancer safety and early diagnosis from 2007-2009 I have presented at NAEDI conferences As RCGP Chair, I was a member of the National Cancer strategy developement and proposed the national audit in primary care I am a supporter the programme of work in early diagnosis of cancer Past Chairman of the RCGP 2004-2007 Chair of The national council for palliative care
Highgate Medical Centre, 5 Storer Close, Sileby. LE12 7UD
Click to like:
+25 June 2012
Dear Editor,
Des Spence asks, ‘does early diagnosis save lives?’ in his typically provocative column. His irritation in seeing patients with three weeks of cough mixes up two separate concepts: public awareness of symptoms (particularly cancer symptoms) and early diagnosis. He purports that the early cancer diagnosis field ‘generally has no intellectual framework, rigour, research or systematic reviews’ and decries the lack of research giving positive predictive values of symptoms in primary care.
He’s sadly missed the systematic review of awareness campaigns1 as well as several systematic reviews of cancer symptoms in primary care,2 based on a series of papers with positive predictive values, including one in the BMJ on ovarian cancer.3 We are restricted to five references, but can happily provide more.
His broader point, however, is that early diagnosis doesn’t work. It’s true there are no randomised controlled trials of earlier cancer diagnosis. That is not from lack of an intellectual framework: quite simply no funder or ethics committee will allow us to design a trial of cancer diagnosis with one arm slower than the other. Even so, there is much observational evidence pointing towards disadvantages from delays in diagnosis.4 Furthermore this probably explains a considerable part of the UK’s poor cancer record.5
Spence provides a useful service as an iconoclast, but needs to be a little more careful in doing his background reading.
Yours sincerely,
William Hamilton, GP and professor of primary care diagnostics, Peninsula College of medicine and Dentistry, Exeter
Debbie Sharp, GP and professor, University of Bristol.
References
1 Austoker, J. et al. Interventions to promote cancer awareness and early presentation: systematic review. Br J Cancer 101, S31-S39, doi:http://www.nature.com/bjc/journal/v101/n2s/suppinfo/6605388s1.html (2009).
2 Astin, M., Griffin, T., Neal, R. D., Rose, P. & Hamilton, W. The diagnostic value of symptoms for colorectal cancer in primary care: a systematic review. Br J Gen Pract 61, e231-e243 (2011).
3 Hamilton, W., Peters, T. J., Bankhead, C. & Sharp, D. Risk of ovarian cancer in women with symptoms in primary care: population based case-control study. BMJ 2009, 339:b2998 (2009).
4 Torring, M. et al. Time to diagnosis and mortality in colorectal cancer: a cohort study in primary care. British Journal of Cancer 104, 934-940 (2011).
5 Abdel-Rahman, M., Stockton, D., Rachet, B., Hakulinen, T. & Coleman, M. P. What if cancer survival in Britain were the same as in Europe: how many deaths are avoidable? Br J Cancer 101, S115-S124, doi:http://www.nature.com/bjc/journal/v101/n2s/suppinfo/6605401s1.html (2009).
Competing interests: Both authors hold several research grants in the field of cancer diagnosis, and have multiple papers on the subject. WH is the clinical lead on the ongoing revision of NICE guidance on referral for suspected cancer (CG27).
peninsula College of Medicine & Dentistry, Veysey Building, Exeter EX2 4SG
Click to like:
+25 June 2012
Author wrote: "diabetes, we have no evidence that promoting early diagnosis through symptoms affects any hard clinical outcomes". The reality is worse. As USPSTF summarized
(http://www.uspreventiveservicestaskforce.org/uspstf/uspsdiab.htm), in asymptomatic people any screening probably doesn't work and in asymptomatic hypertensives it probably works. I bet many diabetologists may be surprised to learn this.
Competing interests: None declared
First State Moscow Medical University, pob 13 Moscow 109451 Russia
Click to like:
+25 June 2012
In reply to Dr Mackenzie, I can think of a few problems with the prima facie Good Idea that early diagnosis makes things better:
It assumes we can identify a disease at its early stage
It assumes we know how common the abnormalities identified are and what their progression will be
It assumes we can identify all and only those cases of the disease which will progress
It assumes we have a proven intervention that can prevent progression
It assumes early intervention produces better outcomes
It assumes absence of uncertainty or disagreement amongst qualified professionals as to the nature of the disease and the success and value of the proposed approach
It assumes that early intervention cannot do harm
It assumes the commitment of an informed eligible public to accepting the intrusion
It assumes medical practice proceeds only on the basis of good quality evidence
It assumes the absence of nonclinical pressures and conflicts of interest on the part of professionals
It assumes detection and intervention have a cost in personal terms the individual will deem worth paying
Some or all of these things are often not the case. Mammography screening is a case in point. The point of no return of some cancers is so critical it is highly unlikely to be discovered by an early detection programme, while many cancers have such a long gestation period or will never progress that an early detection programme in practice misses the very cases he is concerned about while making many other people cancer patients needlessly resulting in net harm on a grand scale. This is just one example of the iatrogenic harm done by inadequately grounded attempts to diagnose diseases early.
Dr Mackenzie wants to believe that at least something he does is worthwhile. I hope it is, but none of that can offset the potential for harm of uncritical enthusiasm for early detection.
Competing interests: Diagnosed through breast cancer screening
King's College London, Strand, WC2R 2LS
Click to like:
+25 June 2012
Des Spence is correct to feel sceptical about some campaigns for "screening" and case finding which may not be supported by robust evidence for patient outcomes.
However, in doing so, he has omitted to add that some campaigns for early diagnosis really do reduce morbidity and mortality, and are supported by NICE.
I could equally summarise this in two acronyms: HIV QED.
Competing interests: None declared
London Sexual Health Programme, The Village Practice, 115 Isledon Road, London N7 7JJ
Click to like:
+25 June 2012
Dear Graeme - nobody likes happy news ! But I promise to write an article called "Good Medicine" - I might choke while I write it ! Thanks des
Competing interests: None declared
General Practice, Maryhill Health Centre G20 9DR
Click to like:
+Dear Prof Hamilton
I try hard to get my facts right and read widely. The evidence for “early symptoms” do come from case control studies that are open to confounding. Firstly your paper on symptoms of Ovarian Cancer is interesting but it is not without it critics. [1] Any conclusion from retrospective case note studies is potentially confounded, because GPs would consider “abdominal distension” a red flag, refer and record this in the record. What is written may not reflect what is reported. Also perhaps you could explain how to distinguish the difference between “bloating” and “distension” clinically? How will the general public make such a distinction? So I question the reported positive predictive value of 2.5%, but even if correct, it is neither sensitive nor specific. Consider the low prevalence: I have seen only one case of ovarian cancer in 20 years. Could it really be 1 in 40 women I see with symptoms of bloating/distension has ovarian cancer?
More importantly, have you considered the potential harms involved? Incidental cysts, the potential risk of surgery and anxiety? Also “Distension” would also suggest late stage disease. You have no evidence that promoting early awareness impacts on mortality, but worse still systematic screening demonstrates no impact on survival for ovarian cancer.[2] If you have other evidence that the promotion of disease awareness of “early symptoms” actually saves lives, I would be interested to read these.
Also, you quote a paper on UK cancer survival comparing to the European average. As you are aware these are fraught with danger, especially data recording [3] . But accepting this, you should read the detail, the excess for mortality seems to largely for Breast and Prostate cancer. Firstly we have a national screening programme for breast cancer, so not sure how these deaths are “preventable”. Secondarily detecting early prostate cancer does not impact on survival [4] I do not accept that it is simply an issue of delayed referral for the variation in cancer survival, this is disingenuous to GPs, and just too simplistic to be plausible.
I am happy to support the promotion “early symptoms” but we should be sure these have value and some hard facts. If systematic screening doesn’t work then why would poorly conceived, expensive 2-3 month public awareness campaign possible have any impact? These campaigns have the potential for real harm in the same way as formal screening does; believing is not the same as knowing and we need facts, not fiction of benefit.
[1] Hamilton, W., Peters, T. J., Bankhead, C. & Sharp, D. Risk of ovarian cancer in women with symptoms in primary care: population based case-control study. BMJ 2009, 339:b2998 (2009).
[2] Buys SS, Partridge E, Black A, Johnson CC, Lamerato L, Isaacs C, et al. Effect of screening on ovarian cancer mortality: The Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Randomized Controlled Trial. JAMA2011;305:2295-303
[3] http://www.nature.com/bjc/journal/v101/n2s/pdf/6605401a.pdf
[4] Djulbegovic M, Beyth RJ, Neuberger MM, Stoffs TL, Vieweg J, Djulbegovic B, et al. Screening for prostate cancer: systematic review and meta-analysis of randomised controlled trials. BMJ 2010;341:c4543
Competing interests: None declared
General Parctice , Maryhill Health Centre G20 9 DR
Click to like:
+22 June 2012
Will Des ever actually write something positive about our job? I am a miserable Scot just like him and love the column for that reason. However even I need a break from my Caledonian pessimism sometime and have the need to feel that some of the things I do are worthwhile.
Some things are just true without further analysis and early diagnsosis of treatable pathological processes must be a good thing.
As a colleague said once, there must be a day when your cancer becomes incurable. The idea is to treat it before that day.
However keep up the good work, Des. We are all dOOOOmed after all.
Competing interests: None declared
Cumbria Partnership, Cumbria
Click to like:
+Latest comments and most commented
Re: Campaigners criticise report into Camelford water poisoning
Published 25 May 2013
DSM-5: hopefully a fatal blow for diagnosis
Published 25 May 2013
Better Deal for Public
Published 25 May 2013
Re: Infections, vaccines, the State, the Doctors and the Drug Firms
Published 25 May 2013