Assessment of changes to screening programmes: why randomisation is important
BMJ 2015; 350 doi: https://doi.org/10.1136/bmj.h1566 (Published 30 March 2015) Cite this as: BMJ 2015;350:h1566
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.
We agree with Hazel Thornton’s comments that an inclusive approach should be adopted for evaluating changes within screening programmes with involvement of non-health-professional members of the community. Our proposal for screening evaluations certainly allows for this. We fully agree that patients should be informed of possible benefits and harms of the screening variations and of no screening and that consent is a collaborative process1. We certainly do not see participation rate as an end-point for judging the success of a screening programme, but rather that people who consider the benefits outweigh the harms do go on to attend for screening – without additional barriers that prevent them from putting their intention into practice. We think that screening should be offered with the best possible technical quality, to inform consumer weigh-up of benefits and harms. Technical failure, as we describe for the FOBTs, interferes with the whole screening process and may affect perspectives on decision-making even when the technical failure is fixed.
Deciding on screening options and outcomes might be guided by processes such as Citizen’s juries2, 3. That would also be a good forum in which to assess uncertainties, and the potential of our idea of randomising within population screening. We look forward to initiatives that build on this, including ways of maximising citizen involvement.
1. Hersch J, Barratt A, Jansen J, Irwig L, McGeechan K, Jacklyn G, Thornton H, Dhillon H, Houssami N, McCaffery K. Use of a decision aid including information on overdetection to support informed choice about breast cancer screening: a randomised controlled trial. The Lancet.
2. Rychetnik L, Carter SM, Abelson J, Thornton H, Barratt A, Entwistle VA, Mackenzie G, Salkeld G, Glasziou P. Enhancing Citizen Engagement in Cancer Screening Through Deliberative Democracy. Journal of the National Cancer Institute. 2013.
3. Rychetnik L, Doust J, Thomas R, Gardiner R, MacKenzie G, Glasziou P. A Community Jury on PSA screening: what do well-informed men want the government to do about prostate cancer screening—a qualitative analysis. BMJ Open. 2014;4(4).
Competing interests: No competing interests
Bell et al brilliantly stressed that screening in healthy subjects must be based on the strongest available evidence.(1) They illustrated this with the shift from guaiac to immunochemical tests for bowel cancer screening. Indeed, long term randomization is mandatory to allow comparison of clinically relevant benefits and harms. This is a key issue as marketing claims from the first round are not relevant for implementation of newer tests: a) positive and negative test rates are only surrogate endpoints; b) during first round screening prevalence of adenoma or cancer is maximal in the screened population. The sensitivity and the specificity may not be constant for subsequent rounds where the increased sensitivity could decrease and the specificity worsen.(2)
France wildly shifted from guaiac to immunochemical tests without debate on the need for long term randomization and without plan for the shift. 11,000 tests (estimate) from people that have made the screening with the guaiac test will be destroyed as they sent it after 31st January 2015.(3) No information yet on how people will be informed or who will do this. In the Alsace region 30,000 unused guaiac tests will be destroyed.(3) The new immunochemical test will be available on … April 14th.
1 Bell KJ, Bossuyt P, Glasziou P, Irwig L. Assessment of changes to screening programmes: why randomisation is important. BMJ 2015;350:h1566.
2 Braillon A. Can surrogate end points from a first-round screening be reliable for colorectalcancer screening? Gastroenterology 2012;142(3):e29.
3 Mascret D. [Cancer screening: thousands of tests lost.] Le Figaro 9 March 2015. Available at http://sante.lefigaro.fr/actualite/2015/03/09/23484-depistage-cancer-col... accessed 3 April 2015.
Competing interests: No competing interests
Many might agree, as I do, with Bell et al [1], who argue that the best way to assess the changes that occur within screening programmes as new technologies and treatments become available is through randomised trials within the programmes. Indeed, some have lamented that this was not done earlier so that fewer uncertainties would have remained now!
I am, however, disquieted by their approach in particular with respect to its failure, as far as I can determine, to include as yet any non-health-professional citizens in this preliminary proposal stage; their question about use of information-provision; their view of consent as something provided by the participant rather than collaboratively achieved [2]; and generally to convey the notion that their proposal is one that should be considered solely by health professionals. Surely it is a matter for all citizens to be involved with from the start of deliberation of these questions for research? Overall, if citizens were to be involved, this proposal would provide a wonderful opportunity to educate the public whilst they are well about all the uncertainties that obtain in screening and in the research process generally. Testing Treatments could be used to explain randomisation, for example, or to learn about screening in Chapter 4 `Earlier is not necessarily better`, and to consider uncertainties. [3]
Australia is in the forefront for exploring how Society might become better engaged in such matters. For example, a Screening Symposium in Sydney in 2011 [4], posited on the belief that `epidemiological evidence of benefits and harms is not enough to inform screening policy`, set out to `find ways to engage communities in identifying the gaps in the existing evidence and facilitate evidence-informed deliberations among community representatives about what is important to them`. The BMJ itself has recognised the importance of patient and citizen involvement in the whole research process in its recent policies. Furthermore, screening is a public health matter that concerns all citizens, be they health-professional or not, utilising public funds in its implementation. Provision of adequate information for potential screening participants has been the subject of intense debate in the pages of the BMJ (and elsewhere) over many years, and is now accepted as an essential human right. The strongest available evidence has been used and debated when considering one of the BMJ`s recent campaigns of `Preventing Overdiagnosis`, with input from non-health-professionals.
In their opening paragraph, Bell et al, when advocating randomised controlled trials, suggest that one of the specific components where uncertainties may remain is: “Should potential participants be given an information booklet?” And when considering any obstacles that need to be overcome, the authors suggest that
“It may be argued that randomisation within screening programmes does not have sufficient benefit for participants and may confuse invitees, leading to low participation and limited health benefits.”
Why no mention of harms avoided? Might not `low properly informed participation` be preferable to achieving `high under- or ill-informed participation`? Is `high participation` a proper goal? Would not all citizens be beneficiaries of a well conducted screening trial? Currently, under-informed citizens can be subjected unknowingly to harm. Consenting invitees might legitimately give their consent for a variety of reasons, aside from their degree of possible confusion. Participants agree to participate in trials for many different reasons, including altruism, with varying degrees of `understanding`/`misunderstanding` (confusion). [5]
I find it disturbing that there seems to be little recognition of the current inclusive approach that respects citizens and their rights, or gives them opportunity to exercise their responsibilities.
[1] Bell KJL, Bossuyt P, Glasziou P, Irwig L. Assessment of changes to screening programmes: why randomisation is important. BMJ 2015;350:h1566
[2] Sutton AG. New rules of consent. BJM rapid response 1st April 2015. http://www.bmj.com/content/350/bmj.h1534/rr-8
[3] Evans I, Thornton H, Chalmers I, Glasziou P. Testing treatments: better research for better healthcare – Second Edition. Pinter and Martin, London. 2011. ISBN 978-1-905177-48-6 Free download from www.testingtreatments.org/the-book/
[4] Screening Symposium, School of Public Health, University of Sydney, 8 December 2011 “What do the community want from screening?”
[5] Dixon-Woods, M, Ashcroft RE, Jackson CJ, Tobin MD, Kivits J, et al. Beyond “misunderstanding”: Written information and decision about taking part in a genetic epidemiology study. Social Science and Medicine. 2007;65:11:2212-2222
Competing interests: No competing interests
Re: Assessment of changes to screening programmes: why randomisation is important
It is good to know that Bell et al [1] agree with me [2] that “an inclusive approach should be adopted for evaluating changes within screening programmes with involvement of non-health-professional members of the community.” [1] Their endorsement of deliberative democratic methods of assessing uncertainties, as well as the potential of their idea of randomising within population screening, by means of Citizens Deliberations, is another area where we are in agreement, and is heartening. [3]
But it is evident that I did not express myself clearly enough, since these areas of agreement do not cover my primary concern that they had not included (as far as I could determine) “… as yet any non-health-professional citizens in this preliminary proposal stage” [2] – i.e. in their discussions, either by reference to their having drawn on perhaps a local citizens` advisory group, or some-such, or by inclusion of a non-health professional in their deliberations for the work-up of this paper, or even also perhaps by inclusion in their authorship. It is a tenet of citizen and patient involvement that they should “be involved [with] from the start of deliberation” [2], i.e. the `back-of-an-envelope-stage`, if they want to get the very best out of full collaboration. It is also worth re-iterating, I believe, that “this proposal would provide a wonderful opportunity to educate the public whilst they are well about all the uncertainties that obtain in screening and in the research process generally.” [2]
The well-articulated author`s rapid response by Alexandra Barratt, [4] responding constructively to ideas and criticisms in the varied rapid responses to her paper about screening controversies, is a model for encouraging all-inclusive collaboration of anyone seeking to improve the current status quo. The public need to be brought up to speed by every available means, and given opportunity to contribute wherever and whenever possible. I think that you have missed providing a good opportunity at the outset of your work: `public involvement` could have beneficially helped to shape and inform your work.
[1] Authors` response to Thornton H. BMJ rapid response to Bell KJL, Bossuyt P, Glasziou P, Irwig L. Assessment of changes to screening programmes: why randomisation is important. BMJ 2015;350:h1566 http://www.bmj.com/content/350/bmj.h1566/rr-1
[2] Thornton H. BMJ rapid response to Bell KJL, Bossuyt P, Glasziou P, Irwig L. Assessment of changes to screening programmes: why randomisation is important. BMJ 2015;350:h1566 http://www.bmj.com/content/350/bmj.h1566/rr
[3] Rychetnik L, Carter SM, Abelson J, Thornton H, Barratt A, Entwistle VA, Mackenzie G, Salkeld G, Glasziou P. Enhancing Citizen Engagement in Cancer Screening Through Deliberative Democracy. Journal of the National Cancer Institute. 2013.
[4] Barratt A. Author`s rapid response to previous responses to Barratt A, Overdiagnosis in mammography screening: a 45 year journey from shadowy idea to acknowledged reality
BMJ 2015;350:h867 http://www.bmj.com/content/350/bmj.h867/rr-8
Competing interests: No competing interests