For healthcare professionals only


“Citizens’ jury” disagrees over whether screening leaflet should put reassurance before accuracy

BMJ 2012; 345 doi: (Published 26 November 2012) Cite this as: BMJ 2012;345:e8047

Re: “Citizens’ jury” disagrees over whether screening leaflet should put reassurance before accuracy

If the jury prefers the word "overtreatment" to "overdiagnosis" and "disadvantage" to "risk" does that mean women are to be told "There is a disadvantage of being overtreated"? which is meaningless, when the fact of the matter is "There is a risk of being overdiagnosed"?

A cause-specific death prevented is not a life saved, so the fact that the jury preferred what is not the case cannot be allowed to affect what women are told. The facts are that there is no evidence screening prolongs lives and the Marmot figure is an overestimate according to the authoritative research of the more experienced Nordic Cochrane Centre. To omit these facts will be to mislead.

To prefer talking of lives saved as “upbeat” shows a dangerous misunderstanding of the task. It would be upbeat if we could say that breast screening is really a game of tiddlywinks, only it wouldn’t be true, and that’s why it can’t be said. The jury should have been told that it is no part of the task of health information to “spin”.

The jury agreed to start by citing “the number of lives saved by breast screening” – in view of the above they must say that this may be 0. They agreed to follow this “by the caveat that a small number of women would suffer overtreatment”. If 4000 is a small number then 1300 is tiny. To be “balanced” (their declared aspiration) they should cite both numbers or neither, in which case on their own terms, “a tiny number of lives is saved and a small number of women are overtreated”, but that would be seriously misleading, since 4000, a quarter of all diagnoses, an underestimate according to Nordic Cochrane, is a lot of women. It is noteworthy that they agreed not to give the number. Why would they do that, unless influenced by those conducting the process and perhaps some witnesses, who have an interest in protecting the programme and who have so often said that the facts would harm it?

Health information aims to inform, not to reassure. The facts must speak for themselves. People wanting reassurance see their doctors. Screeners go to women who are not otherwise worried, offering an intervention, which implies they will be worse off without it. This gives women something to worry about. In their desperation at the latest results they have now resorted to brainwashing techniques, trying to pretend overdiagnosis both isn’t happening and is good news for women.

When invited to witness to the jury I was told they weren’t going to decide content. But these “recommendations” are about content. They make a material difference to the information. One of my many reasons for declining to participate is that content and “presentation” are inseparable – by using different words you say something else, which means something different. This report of the proceedings shows that, and that the very suggestion that the jury was to consider only “presentation” seems to be sophistry.

Competing interests: Diagnosed through screening

27 November 2012
Miriam Pryke
PhD student
King's College London
Strand London WC2R 2LS