Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
YET ANOTHER INCONCLUSIVE AND VALUELESS ARTICLE
- Clifford G. Miller (22 April 2004)
-
YET ANOTHER INCONCLUSIVE AND VALUELESS ARTICLE
- Sam Richmond (23 April 2004)
-
Re: YET ANOTHER INCONCLUSIVE AND VALUELESS ARTICLE
- Clifford G. Miller (25 April 2004)
|
|
|||
|
Clifford G. Miller, Lawyer, graduate physicist, former examining university lecturer on law standards ethics Beckenham, Kent, BR3, England
Send response to journal:
|
What precisely does this news item demonstrate? As Phil Alderson's BMJ editorial points out "Absence of evidence is not evidence of absence" and "We need to report uncertain results and do it clearly". So is this a report of a certain result? It does not tell us that the study concludes there is no link between vaccination and type 1 diabetes nor does it tell us with what degree of confidence the researchers say their research adds to the evidence for or against the proposition that there is or is not a link. This news item provides no justification for the quote it contains that the reported study is "reassuring," and that the findings "can be communicated to parents and allay unfounded fears of vaccination." Where is any summary of the authors' own discussion of any shortcomings and sources of inaccuracy in the result? Where is the summary of discussion of reasons why these results are evidence for the proposition posed or whether there could be other reasons? Where is the summary of discussion of other work needed to assist in reaching a definitive conclusion or are the authors saying their results are definitive? What were the disclosed conflicts of interest, and sources of funding, or reasons for undertaking the research? Is the reporter saying in his opinion there is no link and this paper establishes that? That is the implication of carrying the quotes of the leader of the research team. How can this news item be justification for saying the findings "can be communicated to parents and allay unfounded fears of vaccination." Are the parents supposed to go to the New England Journal of Medicine 2004;350:1398-404 and check it out themselves? Ordinary press journalists might see a report like this in the BMJ and use it to base a press article for the mass media reporting this information, relying on the reputation of the BMJ for believing in the accuracy of the report. Is there really such a thing as medical science? I know Mickey Mouse exists - I saw him. Competing interests: Close relative with life threatening food allergy. |
|||
|
|
|||
|
Sam Richmond, Neonatologist Sunderland Royal Hospital, SR4 7TP
Send response to journal:
|
I think Mr Miller is being slightly obtuse. My answer to his question 'What precisely does this news item demonstrate?" is that the news item demonstrates that someone has taken the trouble to explore the possibility of a link between vaccination and later development of Type 1 diabetes in children. It also tells one precisely where one can look to find the answers to all of Mr Miller's subsequent questions - namely the New England Journal of Medicine 2004, volume 350, pages 1398-1404. To answer all of Mr Miller's questions in the news item would require the BMJ to reprint the article in question. Phil Anderson's editorial does indeed comment that 'Absence of evidence is not evidence of absence' however the absence of any apparent link when that link is specifically searched for in a large cohort of children (in this case 681 children with Type 1 diabetes during 4,720,517 person years of follow up) can reasonably lead to the conclusion that the study comes to - namely "These results do not support a causal relation between childhood vaccination and type 1 diabetes". Given the size and focus of the study I would also suggest that this absence of evidence of a link is indeed "reassuring". Competing interests: None declared |
|||
|
|
|||
|
Clifford G. Miller, Lawyer, graduate physicist, former examining lecturer in law standards & ethics Beckenham, Kent, England, BR3
Send response to journal:
|
I am grateful to Sam Richmond for his comments, which are clearly genuinely made and seem at first to be pertinent and valid. However, all is not quite as it appears. There is also much more at stake here than what is really a theoretical possibility for most readers of tracking down the original paper in the New England Journal of Medicine (‘NEJM’). This news item is not just inaccurate, but misleading. The reader cannot get past the headline without being misled. The headline inaccurately states "Researchers find no link between vaccination and type 1 diabetes" whereas the main conclusion stated by the authors is that "These results do not support a causal relation between childhood vaccination and type 1 diabetes." The former states the authors claim their results do not support a causal relation whereas the latter, despite its ambiguity, seems to say the paper is scientific proof there is no link, which, even from a reading of the conclusion in the paper’s abstract, is clearly not the case. Moreover, thousands of medical practitioners in a hurry across the world will read the BMJ and will not necessarily have or be able to take time to read one of hundreds of papers published every month. They need accessible, accurate and good information to keep them up to date within and outside their own fields. Most will not even subscribe to the NJEM, so that it is more likely some, if they have time, will read the cricket results in the Kathmandu Post than look up the reference in the NEJM. The readers need critical and accurate reporting. An inaccurate repetition of no more than the contents of the press release from the paper’s authors is more likely to hinder that than help. The BMJ needs to ensure and maintain its high quality service to them as its prime readership in taking care to provide critical analysis and comment for its intelligent, highly trained and qualified, but busy, readership. Further, the authors of the NEJM article are just making claims. The claims are yet to be subject to peer scrutiny after publication. The claims might prove valid or baseless, inaccurate or flawed. Peer review prior to publication is not a guarantee of validity of the claims. Accordingly, it is better to keep a balanced view when reporting these things, even if only to make clear that the conclusions are 'claimed' rather than just assuming they are proven scientific facts. Further, whilst claims made in the paper about the accuracy of the data underlying the reported study might seem (even though stated in a vague way) to give Sam Richmond's reading of the paper and his conclusions validity, until that data and the conclusions based on it in the paper have been subject to peer scrutiny, there is no way of knowing for certain how much credence to give to this paper All of this also needs to be considered in the light of the 'drip, drip' effect numerous reports of this nature in this and other fields might have, such that a misleading or inaccurate overview of the current state of research is inadvertently passed on to busy practitioners. Whilst it is easily done and I do not castigate but thank him, Sam Richmond inadvertently also makes one of my points for me. He had to turn to the paper itself in the NEJM in order to report on and provide in a summary form a key aspect which could and should have been stated in the news item concerned. And this leads to a final and most important point. The busy reader does not obtain from the report in this news item any summary of information about the paper to assess its importance and relevance, such as:- * the authors' own discussion of any shortcomings and sources of inaccuracy in the study, its methods or data; * reasons why these results are evidence for the proposition posed or whether there could be other reasons that may require further investigation; * discussion of other work or research needed to assist in reaching a definitive conclusion; * disclosed conflicts of interest, and sources of funding, or reasons for undertaking the research. Bearing in mind the main conclusion of the study, how can this news item be justification for saying the findings "can be communicated to parents and allay unfounded fears of vaccination." when all it is saying is that this study does not support a link, not that there is none. Further, ordinary press journalists might see a report like this in the BMJ and use for a press article for the mass media reporting this information, relying on the reputation of the BMJ for believing in the accuracy of the report but unknowingly then misinforming public opinion by passing on inaccurate information to the public, public officials and politicians. Competing interests: Close relative with life threatening food allergy. |
|||