- Email: [email protected]
Lessons from the controversy over statins – Authors’ reply
4
5
Centres for Disease Control and Prevention. Laboratory testing for the diagnosis of HIV infection. June 27, 2014. http://www.cdc.gov/ hiv/pdf/hivtestingalgorithmrecommendationfinal.pdf (accessed Aug 9, 2016). Suligoi B, Pupella S, Regine V, Raimondo M, Velati C, Grazzini G. Changing blood donor screening criteria from permanent deferral for men who have sex with men to individual sexual risk assessment: no evidence of a significant impact on the human immunodeficiency virus epidemic in Italy. Blood Transfusion 2013; 11: 441–48.
Lessons from the controversy over statins As the chair of the Committee on Publication Ethics (COPE), I would like to respond to the remarks of Richard Horton in his Offline Comment1 (Sept 10, p 1040) about the role and actions of COPE, in which he highlighted the statins Review by Collins and colleagues.2 COPE is an international inter disciplinary organisation, not just a UK one, whose remit is the provision of education and advice to members with questions about publication ethics. We do have a process whereby an individual can bring to our attention complaints about journal processes, but we cannot interfere in editorial decisions, or investigate the underlying issues of a complaint because we have neither the resources nor—more importantly—the appropriate level of subject-specific expertise. Horton states that “COPE declined to act further”; however, this is incorrect. COPE did request details of processes at the British Medical Journal (BMJ), in accordance with our remit. The guidance issued from COPE’s review (I was not part of this having recused myself because of a potential conflict of interest) offered constructive criticism about how BMJ had managed the peer review process.3 BMJ had already addressed those issues following their own independent review and COPE was satisfied with the procedural changes that were implemented. Because it is not appropriate for COPE to make any specific judgment about effects on public health, www.thelancet.com Vol 388 November 5, 2016
COPE also recommended that Collins and colleagues engaged in open dialogue on the specific issues in the medical literature. We note this has now happened with the publication of their Review in The Lancet. Putting the correction of Richard Horton’s Comment to one side, and instead looking for useful lessons, COPE would be willing to discuss his suggestion for an independent tribunal. This tribunal would probably need public funding and the ability to apply sanctions and, to a degree, the ability to become a regulator for the research community. This is not COPE’s remit, but we are interested in being part of the discussion on such an approach. I am the Chair of COPE. At PLOS I worked previously with one of the BMJ staff who subsequently handled the complaint from Rory Collins and colleagues to the BMJ. I have co-authored a paper with Fiona Godlee, the BMJ Editor in Chief.
Virginia Barbour [email protected] Committee on Publication Ethics, Brisbane, QLD 4000, Australia 1 2
3
Horton R. Offline: Lessons from the controversy over statins. Lancet 2016; 388: 1040. Collins R, Reith C, Emberson J, et al. Interpretation of the evidence for the efficacy and safety of statin therapy. Lancet 2016; published online Sept 8. http://dx.doi. org/10.1016/S0140-6736(16)31357-5. BMJ. Statins—a call for transparent data. 2016. http://www.bmj.com/campaign/statins-opendata (accessed Oct 14, 2016).
the responses of COPE, are available online for public scrutiny. These documents show that Barbour only recused herself from this investigation some months after adjudicating on the submission and liaising with the subject of the complaint. COPE has an ambiguous role and mandate. Barbour suggests that COPE does not interfere in editorial decisions or investigate the underlying issues of a complaint. But these statements are untrue. COPE provides advice to complainants and actively investigates and follows up with journals to question their decisions. It is therefore unfortunate that COPE’s procedures seem to have become so arbitrary. As a co-founder of COPE, The Lancet hoped to support an organisation that, after appropriate discussion and investigation, provided impartial recommendations to journals about allegations of publication or research misconduct. It is regrettable that COPE has drifted away from its original mission.
Anne Cusack/Contributor
Correspondence
For information about COPE see http://publicationethics.org/
I am the Editor of The Lancet.
Richard Horton [email protected] The Lancet, 125 London Wall, London EC2Y 5AS, UK 1
Horton R. Offline: Lessons from the controversy over statins. Lancet 2016; 388: 1040.
Editor’s reply I’m grateful to Virginia Barbour for responding to the criticisms of the Committee on Publication Ethics (COPE). And I very much welcome her endorsement of the idea for an independent tribunal to consider allegations of research or publication malpractice. However, if she had quoted my words fully and fairly, I think she would have to agree that the claim that “COPE declined to act further” in response to “a direct request [to] conduct an independent investigation”1 was correct. COPE did decline to act further, and this refusal has led to the current exchange. The questions raised by a group of respected scientists, together with
Authors’ reply Virginia Barbour states in her letter that, with respect to the submission to the Committee on Publication Ethics (COPE) in October, 2014, by a group of senior doctors and scientists, she had “recused myself because of a potential conflict of interest”. However, that is not strictly accurate; Barbour only recused herself in September, 2015, 8 months after she had adjudicated on the submission in January, 2015. In that adjudication, Barbour had not addressed any of the specific concerns that had been raised about failures of editorial integrity at the British Medical Journal (BMJ), and had not reviewed any of the supporting material that was 2237
Correspondence
provided as links within the submission, so she was asked in February, 2015, to address these issues specifically. Barbour was then also responsible for sharing the submission with the BMJ several months before recusing herself. In June, 2015, a response from the BMJ was addressed in a second substantive submission from the group of senior doctors and scientists, which reiterated their concerns and again asked that COPE address each of them specifically. It was only in September, 2015, that Barbour stated that she was recusing herself because she had a potential conflict of interest. The nature of that competing interest was not made clear (and it is not disclosed in her letter), but what is clear is that Barbour did not recuse herself before having determined the outcome of COPE’s review. Indeed, conclusions from Barbour’s report are being used as the basis for asserting that COPE has conducted a properly independent review of the issues raised.1 After Barbour did recuse herself, it took until April, 2016, before COPE eventually stated that it felt the BMJ had taken steps to correct its mistakes. However, COPE still did not address any of the specific concerns raised about failures of editorial integrity at the BMJ or the adverse effect on public health of the misleading claims that the BMJ had published about side-effect rates with statins. COPE was again asked to address these points specifically. However, it refused to do so and also refused to allow the matter to be considered by its Ombudsman, writing in July, 2016, that it regarded “this matter as closed as there is nothing further that we can usefully bring to this debate”. Therefore, it is difficult to understand Barbour’s other assertion that Richard Horton was incorrect in stating that COPE declined to act further.2 In parallel with Barbour, the Editor of the BMJ has suggested in a Letter3 that Horton’s Offline Comment2 was incorrect because he had not had access 2238
to the material submitted to COPE (including by the BMJ) and its responses. However, that is not correct: a detailed description of this sequence of events, along with links to the submissions to COPE and its responses, was made available at the Cholesterol Treatment Trialists’ Collaboration website4 in June, 2016, and other related material was also provided to The Lancet. Our Lancet Review 5 provides a detailed explanation of the reasons why the available evidence from the randomised controlled trials provides a more reliable basis for assessing the safety and efficacy of statin therapy than does observational evidence from case reports or health-care databases. Its intention is to help ensure that doctors and their patients are properly informed about the safety and efficacy of statin therapy (and not, as stated by the BMJ’s editor,6 to shut down debate). As Barbour’s letter demonstrates, Horton’s proposal for a tribunal that could address such matters of public health importance independently of the journals does need to be explored further. However, in the meantime, the failure of the BMJ to deal properly with the seriously misleading claims about the effects of statin therapy that it published, and then the failure of COPE to deal properly with editorial failures at the BMJ, is continuing to have an adverse impact on public health worldwide. Along with several other scientists and doctors, we were cosignatories to the submission to COPE and coauthors of the recent Lancet Review of the safety and efficacy of statin therapy. The CTSU has received grant funding from the pharmaceutical industry for independent research; our approach to maintaining the independence of that research is described at https://www.ctsu.ox.ac.uk/research/ctsuindependent-research_27june14.pdf and to maintaining our own independence by not taking honoraria or consultancy payments from industry is described at https://www.ctsu.ox.ac.uk/about/ctsu_ honoraria_25june14-1.pdf.
Jane Armitage, Colin Baigent, *Rory Collins [email protected] Clinical Trial Service Unit & Epidemiological Studies Unit (CTSU), Nuffield Department of Population Health, University of Oxford, Oxford OX3 7LF, UK
1
2 3
4
5
6
Husten L. The Lancet versus BMJ: a dispatch from the statin wars. New York: MedPage Today. Sept 15, 2016. http://www. medpagetoday.com/Cardiology/ CardioBrief/60251 (accessed Oct 11, 2016). Horton R. Offline: Lessons from the controversy over statins. Lancet 2016; 388: 1040. Godlee F. Rapid response re COPE complaint. London: BMJ. Sept 14, 2016. http://www.bmj. com/content/351/bmj.h3908/rr-8 (accessed Oct 3, 2016). Cholesterol Treatment Trialists’ Collaboration. Renewed call to retract BMJ articles that mislead on statin effects. https://www. cttcollaboration.org/news/renewed-call-toretract-bmj-articles-that-mislead-on-statineffects (accessed Oct 3, 2016). Collins R, Reith C, Emberson J, et al. Interpretation of the evidence for the efficacy and safety of statin therapy. Lancet 2016; published online Sept 8. http://dx.doi. org/10.1016/S0140-6736(16)31357-5. Godlee F. Statins: we need an independent review. London: BMJ. Sept 15, 2016. http://www.bmj.com/content/354/bmj.i4992 (accessed Oct 11, 2016).
Department of Error Brinkman SA, Johnson SE, Codde JP, et al. Efficacy of infant simulator programmes to prevent teenage pregnancy: a school-based cluster randomised controlled trial in Western Australia. Lancet 2016; 388: 2164–75—In this Article, the name of the funding organisation Healthway should have been written as “Western Australian Health Promotion Foundation (Healthway)” in the Summary Funding section. In figure 1, both percentages written as 45% in the control group should be 50%. In table 1, in the first row, the numbers in brackets are the mean number of participants per school, not percentages, and the table footnote has been modified accordingly. In the second paragraph of the Discussion, the participation rate at the individual level in the control schools should be 50%. These corrections have been made to the online version as of Nov 3, 2016, and the printed Article is correct. Shaw D, Guise J-M, Shah N, et al. Drivers of maternity care in high-income countries: can health systems support woman-centred care? Lancet 2016; 388: 2282–95—In figure 1A of this Review, the label for the y-axis should have read “MMR (per 100 000 livebirths)”. This correction has been made to the online version as of Nov 3, 2016, and the printed Article is correct.
www.thelancet.com Vol 388 November 5, 2016
5
Centres for Disease Control and Prevention. Laboratory testing for the diagnosis of HIV infection. June 27, 2014. http://www.cdc.gov/ hiv/pdf/hivtestingalgorithmrecommendationfinal.pdf (accessed Aug 9, 2016). Suligoi B, Pupella S, Regine V, Raimondo M, Velati C, Grazzini G. Changing blood donor screening criteria from permanent deferral for men who have sex with men to individual sexual risk assessment: no evidence of a significant impact on the human immunodeficiency virus epidemic in Italy. Blood Transfusion 2013; 11: 441–48.
Lessons from the controversy over statins As the chair of the Committee on Publication Ethics (COPE), I would like to respond to the remarks of Richard Horton in his Offline Comment1 (Sept 10, p 1040) about the role and actions of COPE, in which he highlighted the statins Review by Collins and colleagues.2 COPE is an international inter disciplinary organisation, not just a UK one, whose remit is the provision of education and advice to members with questions about publication ethics. We do have a process whereby an individual can bring to our attention complaints about journal processes, but we cannot interfere in editorial decisions, or investigate the underlying issues of a complaint because we have neither the resources nor—more importantly—the appropriate level of subject-specific expertise. Horton states that “COPE declined to act further”; however, this is incorrect. COPE did request details of processes at the British Medical Journal (BMJ), in accordance with our remit. The guidance issued from COPE’s review (I was not part of this having recused myself because of a potential conflict of interest) offered constructive criticism about how BMJ had managed the peer review process.3 BMJ had already addressed those issues following their own independent review and COPE was satisfied with the procedural changes that were implemented. Because it is not appropriate for COPE to make any specific judgment about effects on public health, www.thelancet.com Vol 388 November 5, 2016
COPE also recommended that Collins and colleagues engaged in open dialogue on the specific issues in the medical literature. We note this has now happened with the publication of their Review in The Lancet. Putting the correction of Richard Horton’s Comment to one side, and instead looking for useful lessons, COPE would be willing to discuss his suggestion for an independent tribunal. This tribunal would probably need public funding and the ability to apply sanctions and, to a degree, the ability to become a regulator for the research community. This is not COPE’s remit, but we are interested in being part of the discussion on such an approach. I am the Chair of COPE. At PLOS I worked previously with one of the BMJ staff who subsequently handled the complaint from Rory Collins and colleagues to the BMJ. I have co-authored a paper with Fiona Godlee, the BMJ Editor in Chief.
Virginia Barbour [email protected] Committee on Publication Ethics, Brisbane, QLD 4000, Australia 1 2
3
Horton R. Offline: Lessons from the controversy over statins. Lancet 2016; 388: 1040. Collins R, Reith C, Emberson J, et al. Interpretation of the evidence for the efficacy and safety of statin therapy. Lancet 2016; published online Sept 8. http://dx.doi. org/10.1016/S0140-6736(16)31357-5. BMJ. Statins—a call for transparent data. 2016. http://www.bmj.com/campaign/statins-opendata (accessed Oct 14, 2016).
the responses of COPE, are available online for public scrutiny. These documents show that Barbour only recused herself from this investigation some months after adjudicating on the submission and liaising with the subject of the complaint. COPE has an ambiguous role and mandate. Barbour suggests that COPE does not interfere in editorial decisions or investigate the underlying issues of a complaint. But these statements are untrue. COPE provides advice to complainants and actively investigates and follows up with journals to question their decisions. It is therefore unfortunate that COPE’s procedures seem to have become so arbitrary. As a co-founder of COPE, The Lancet hoped to support an organisation that, after appropriate discussion and investigation, provided impartial recommendations to journals about allegations of publication or research misconduct. It is regrettable that COPE has drifted away from its original mission.
Anne Cusack/Contributor
Correspondence
For information about COPE see http://publicationethics.org/
I am the Editor of The Lancet.
Richard Horton [email protected] The Lancet, 125 London Wall, London EC2Y 5AS, UK 1
Horton R. Offline: Lessons from the controversy over statins. Lancet 2016; 388: 1040.
Editor’s reply I’m grateful to Virginia Barbour for responding to the criticisms of the Committee on Publication Ethics (COPE). And I very much welcome her endorsement of the idea for an independent tribunal to consider allegations of research or publication malpractice. However, if she had quoted my words fully and fairly, I think she would have to agree that the claim that “COPE declined to act further” in response to “a direct request [to] conduct an independent investigation”1 was correct. COPE did decline to act further, and this refusal has led to the current exchange. The questions raised by a group of respected scientists, together with
Authors’ reply Virginia Barbour states in her letter that, with respect to the submission to the Committee on Publication Ethics (COPE) in October, 2014, by a group of senior doctors and scientists, she had “recused myself because of a potential conflict of interest”. However, that is not strictly accurate; Barbour only recused herself in September, 2015, 8 months after she had adjudicated on the submission in January, 2015. In that adjudication, Barbour had not addressed any of the specific concerns that had been raised about failures of editorial integrity at the British Medical Journal (BMJ), and had not reviewed any of the supporting material that was 2237
Correspondence
provided as links within the submission, so she was asked in February, 2015, to address these issues specifically. Barbour was then also responsible for sharing the submission with the BMJ several months before recusing herself. In June, 2015, a response from the BMJ was addressed in a second substantive submission from the group of senior doctors and scientists, which reiterated their concerns and again asked that COPE address each of them specifically. It was only in September, 2015, that Barbour stated that she was recusing herself because she had a potential conflict of interest. The nature of that competing interest was not made clear (and it is not disclosed in her letter), but what is clear is that Barbour did not recuse herself before having determined the outcome of COPE’s review. Indeed, conclusions from Barbour’s report are being used as the basis for asserting that COPE has conducted a properly independent review of the issues raised.1 After Barbour did recuse herself, it took until April, 2016, before COPE eventually stated that it felt the BMJ had taken steps to correct its mistakes. However, COPE still did not address any of the specific concerns raised about failures of editorial integrity at the BMJ or the adverse effect on public health of the misleading claims that the BMJ had published about side-effect rates with statins. COPE was again asked to address these points specifically. However, it refused to do so and also refused to allow the matter to be considered by its Ombudsman, writing in July, 2016, that it regarded “this matter as closed as there is nothing further that we can usefully bring to this debate”. Therefore, it is difficult to understand Barbour’s other assertion that Richard Horton was incorrect in stating that COPE declined to act further.2 In parallel with Barbour, the Editor of the BMJ has suggested in a Letter3 that Horton’s Offline Comment2 was incorrect because he had not had access 2238
to the material submitted to COPE (including by the BMJ) and its responses. However, that is not correct: a detailed description of this sequence of events, along with links to the submissions to COPE and its responses, was made available at the Cholesterol Treatment Trialists’ Collaboration website4 in June, 2016, and other related material was also provided to The Lancet. Our Lancet Review 5 provides a detailed explanation of the reasons why the available evidence from the randomised controlled trials provides a more reliable basis for assessing the safety and efficacy of statin therapy than does observational evidence from case reports or health-care databases. Its intention is to help ensure that doctors and their patients are properly informed about the safety and efficacy of statin therapy (and not, as stated by the BMJ’s editor,6 to shut down debate). As Barbour’s letter demonstrates, Horton’s proposal for a tribunal that could address such matters of public health importance independently of the journals does need to be explored further. However, in the meantime, the failure of the BMJ to deal properly with the seriously misleading claims about the effects of statin therapy that it published, and then the failure of COPE to deal properly with editorial failures at the BMJ, is continuing to have an adverse impact on public health worldwide. Along with several other scientists and doctors, we were cosignatories to the submission to COPE and coauthors of the recent Lancet Review of the safety and efficacy of statin therapy. The CTSU has received grant funding from the pharmaceutical industry for independent research; our approach to maintaining the independence of that research is described at https://www.ctsu.ox.ac.uk/research/ctsuindependent-research_27june14.pdf and to maintaining our own independence by not taking honoraria or consultancy payments from industry is described at https://www.ctsu.ox.ac.uk/about/ctsu_ honoraria_25june14-1.pdf.
Jane Armitage, Colin Baigent, *Rory Collins [email protected] Clinical Trial Service Unit & Epidemiological Studies Unit (CTSU), Nuffield Department of Population Health, University of Oxford, Oxford OX3 7LF, UK
1
2 3
4
5
6
Husten L. The Lancet versus BMJ: a dispatch from the statin wars. New York: MedPage Today. Sept 15, 2016. http://www. medpagetoday.com/Cardiology/ CardioBrief/60251 (accessed Oct 11, 2016). Horton R. Offline: Lessons from the controversy over statins. Lancet 2016; 388: 1040. Godlee F. Rapid response re COPE complaint. London: BMJ. Sept 14, 2016. http://www.bmj. com/content/351/bmj.h3908/rr-8 (accessed Oct 3, 2016). Cholesterol Treatment Trialists’ Collaboration. Renewed call to retract BMJ articles that mislead on statin effects. https://www. cttcollaboration.org/news/renewed-call-toretract-bmj-articles-that-mislead-on-statineffects (accessed Oct 3, 2016). Collins R, Reith C, Emberson J, et al. Interpretation of the evidence for the efficacy and safety of statin therapy. Lancet 2016; published online Sept 8. http://dx.doi. org/10.1016/S0140-6736(16)31357-5. Godlee F. Statins: we need an independent review. London: BMJ. Sept 15, 2016. http://www.bmj.com/content/354/bmj.i4992 (accessed Oct 11, 2016).
Department of Error Brinkman SA, Johnson SE, Codde JP, et al. Efficacy of infant simulator programmes to prevent teenage pregnancy: a school-based cluster randomised controlled trial in Western Australia. Lancet 2016; 388: 2164–75—In this Article, the name of the funding organisation Healthway should have been written as “Western Australian Health Promotion Foundation (Healthway)” in the Summary Funding section. In figure 1, both percentages written as 45% in the control group should be 50%. In table 1, in the first row, the numbers in brackets are the mean number of participants per school, not percentages, and the table footnote has been modified accordingly. In the second paragraph of the Discussion, the participation rate at the individual level in the control schools should be 50%. These corrections have been made to the online version as of Nov 3, 2016, and the printed Article is correct. Shaw D, Guise J-M, Shah N, et al. Drivers of maternity care in high-income countries: can health systems support woman-centred care? Lancet 2016; 388: 2282–95—In figure 1A of this Review, the label for the y-axis should have read “MMR (per 100 000 livebirths)”. This correction has been made to the online version as of Nov 3, 2016, and the printed Article is correct.
www.thelancet.com Vol 388 November 5, 2016