96 e-Letters

  • More than 20,000 systematic reviews were published in 2017

    The number of systematic reviews being published each year is actually much larger than the already impressive number of 10,000 quoted. In 2017 over 20,000 were published; 20,661 are cited in the KSR Evidence database in early August 2018. KSR Evidence, includes all systematic reviews and meta-analyses published since 2015 and for many reviews provides a critical appraisal and a short, accessible bottom line. www.ksrevidence.com

  • Problems associated with not using a true placebo in clinical trials

    The problems associated with NOT using a true placebo in clinical trials - in this case GSK's trial for Cervarix - were discussed way back in 2009:



    September 9, 2009


    DR. DEBOLD: I just think in the absence of having a true placebo in the entire study, it is very, very difficult to sort out what the effects are here, what is it that we are really dealing with, what is the baseline, versus what potentially is caused by the so-called controls.

    In this particular study, not only was there two different strengths of Havrix used, there was alum used. In the pooled analysis there were a number of other vaccines used as the control, which makes it scientifically very difficult to sort things out.

  • Retraction notices in PubMed

    I searched PubMed for the retraction notices. The 3 retracted articles published by Elsevier have no retraction notices in PubMed. Retraction notices for the 2 articles in the Journal of the American Society of Hypertension were published in vol. 9, issue 10, this issue is indexed in PubMed and articles citations are in PubMed but the two retraction notices are missing. "Publishers of journals in PubMed must submit citation and abstract" (PubMed FAQ for publishers), these two notices should normally have been submitted by the publisher. As for the article published in the European Journal of Pharmaceutical Sciences, no retraction notice was found, information about the article being retracted is available in the original article but, as no retraction notice seems to be available, there is no information about the retracted status in PubMed. As PubMed is often the source used by researchers doing a study on retractions of articles (e.g. PMID: 28683764, PMID 26797347, PMID: 24928194) it is important to find this information in the database. Publishers should always publish retraction notices for retracted articles and submit the citations to PubMed when the journal is indexed in this database.

  • Response to Brodersen et al’s ’Overdiagnosis: what it is and what it isn’t'

    In their editorial, Brodersen et al. present two types of overdiagnosis. Their two types appear to be identical to two types of overdiagnosis we identified in research published in 2016 (Rogers WA and Mintzker Y. Getting clearer on overdiagnosis. J Eval Clin Prac 2016;22: 580-587). In that paper, we provide a detailed account of maldetection overdiagnosis and misclassification overdiagnosis, together with an analysis of the relevance of these two different types.
    This matter, including our request for a correction, is fully explained in our letter published in this journal: Response to Brodersen et al’s ’Overdiagnosis: what it is and what it isn’t' (http://ebm.bmj.com/content/early/2018/03/29/bmjebm-2018-110948).

  • Re:DPP4is are safe
    Joshua J. Fenton

    Dear Editor,

    Andrea Giaccari asserts that I wrote in my editorial that sitagliptin in the TECOS trial "caused" 20% increase in the secondary outcome of congestive heart failure. That is not what I wrote. I wrote that "the study data remain consistent with" a 20% increase in this adverse outcome. While the wide confidence interval is also consistent with a reduction in heart failure risk, the primary goal of the TECOS tr...

    Show More
  • DPP4is are safe
    Andrea Giaccari

    Dear Editor,

    I read with interest the comments of Dr. Fenton. In his editorial (Evid Based Med. 2016 Jun;21(3):81-2) Dr Fenton stated that sitagliptin caused in TECOS "a 20% increase in the secondary outcome of congestive heart failure (intention-to-treat HR 1.00, 95% CI 0.82 to 1.20, p=0.98)". This is really misleading. With exactly the same numbers Dr. Fenton could state that sitagliptin caused an 18% reduction in hos...

    Show More
  • Judging evidence in postoperative analgesia
    Andrew Moore

    Dear Editor,

    First, thank you for highlighting our paper. However, I do want to take issue with this commentary.

    Systematic reviews in postop analgesia have been done now for over 20 years, and there is considerable methodological research to substantiate what is done. The results are robust and trustworthy.

    Single trials, however well done, are not trustworthy because while they may be powered t...

    Show More
  • 'Cognitive biases plus' and healthcare evidence
    Shashi S. Seshia

    Dear Editor,

    Mazar and Ariely's recent paper [1] reinforces the concepts and suggestions discussed in our recent publications: dishonesty is a human universal, and there is no one-size-fits-all solution [2,3]. Education, moral reminders and changing how researchers are rewarded are important tools [1]. Most importantly, we need to reclaim the integrity, dedication and code of honor Sir Austin Bradford Hill consider...

    Show More
  • Re:Inaccurate radiation exposure calculation
    William D. Rogers

    Dear Editor,

    You are correct that protocols and improved technology have led to reductions in radiation exposure from CT scanning at some hospitals. I would suggest though that the resultant reduction in the risk of fatal cancer due to imaging does not affect the conclsion of the paper. If a laparotomy on a healthy young patient carries no risk of death and CT scanning imposes a risk of death the decision to perfor...

    Show More
  • Inaccurate radiation exposure calculation
    Nigel D'Souza

    Dear Editor,

    Dr Rogers et al have astutely pointed out the dangers of routine CT assessment of right iliac fossa pain in the paediatric population. I agree wholeheartedly that the role of clinical judgement, alongside observation and serial examination remain critical. Ultrasonography and MRI are additional valuable diagnostic adjuncts that do not incur a radiation dose to patients.

    I would question the da...

    Show More