Conflicts of Interest (COI) have the potential to adversely affect patient care. The obvious and major COI are the result of associations between doctors and industry (pharmaceutical and medical device companies). They need to be minimised and always disclosed in publications, but there are other COI that are often forgotten. I am concerned that doctors fail to disclose their COI and that regulators, institutions and journals turn a blind eye to lack of transparency. In some cases these bodies are complicit in concealment. I describe some examples from my own experiences.
Commission on Human Medicines
During 2014 and 2015 I was a member of an Expert Working Group set up by the Commission on Human Medicine (CHM) to review the UK licence for Alteplase for treatment of acute thrombotic stroke. The CHM is a division of the Medicines and Healthcare products Regulatory Agency (MHRA), which is the UK equivalent of the US Food and Drug Administration. The Working Group’s members included senior doctors (consultant and professors), some of who are stroke physicians. The secretariat of the CHM invited me to join the Working Group in September 2014. They wrote that disqualifying COI for membership were financial association with the corporations that manufacture and market Alteplase or involvement in the clinical trials of Alteplase in stroke that would be discussed at the meetings. I disclosed no COI because I had none.
Before each of the three meeting, members were sent an agenda, which was in black print except for a section highlighted in red print reminding us of the disqualifying COI and asking us to confirm again that we had none. At the start of each meeting, the chair, Prof Sir Ian Weller read out the COI exclusions. No member spoke up at any point to say that they had disqualifying COI. After the final meeting of the Working Group in June 2015, I discovered that four members had COI that should have disqualified them from membership. Two were professors and two were NHS consultants. I wrote to CHM in August 2015 pointing out my concerns that the COI policy had been breached. I had to write repeatedly before I got a reply ten months later in June 2016 from Dr June Raine (Director, Vigilance and Risk Management of Medicines at MHRA). She said that MHRA knew that some of the four had disqualifying COI and had agreed that those members could be members despite their COI. Other members of the Working Group were misled because we thought that the stated COI policy for membership had been enforced. The public were misled because the documents used by the Working Group and its minutes were placed on the MHRA website. The documents stated the Working Group’s COI requirements and implied that the policy had been enforced. One of those in breach of the policy was involved in the press conference at which MHRA claimed that an independent Working Group had made the decision on the licence. It seems to me that the MHRA knowingly allowed some members of the Working Group to have COI contrary to the stated policy and was complicit in misleading the public, the popular press and medical journals that had sent reporters to the press conference.
Migraine Intervention with STARFlex© Technology Trial
There were a number of issues related to the COI in the MIST (Migraine Intervention with STARFlex© Technology) Trial, sponsored by NMT Medical in the UK. It was a multicentre randomised double-blind trial comparing the effects on migraine of transcatheter closure of persistent foramen ovale versus a sham intervention. Dr Dowson (Director of headache services at King’s College Hospital, London) and I were the co-principal investigators in the trial. Dowson was the principal headache specialist and I was the principal cardiologist. The steering committee also included three other doctors.
I did not know until nearly three years later that three months before NMT appointed Dr Dowson to his role in the MIST Trial, Dowson had admitted entering false data on clinical record forms in an unrelated multicentre clinical trial (sponsored by Allergan) in which he was the principal investigator.1 The ethics committee immediately expelled Dowson from the Allergan sponsored trial and told him that he was not fit to be a trial principal investigator. 1 When NMT appointed Dowson to be my co-principal investigator in the MIST Trial three months later, they knew about his behaviour in the Allergan sponsored trial. Later when the regulator that licences doctors to practise in the UK, the General Medical Council (GMC), found Dowson guilty of misconduct in the Allergan sponsored trial, Dowson informed NMT about the GMC sanctions (conditions imposed on his practice), but Dowson and NMT did not inform me about Dowson’s record of research misconduct. I discovered it later from someone unconnected with the MIST Trial. (NMT’s decision to use as co-principal investigator in the MIST trial a doctor with a record of research misconduct may seem strange, but I later discovered that in another trial NMT had used as co-principal investigator a doctor who had been sent a FDA letter warning him that he might face criminal charges for misconduct in research.)
The results of the MIST trial were published in the journal Circulation.2 Another member of the steering committee (Dr Simon Nightingale) and I refused to be authors because the steering committee had been denied access to the full trial data, but even without sight of the full data it was clear that the paper contained false statements about the methods in the trial and false results. After the paper was published I spent considerable effort getting Circulation to publish a correction and new version of the paper, which confirm the falsity of the original version.3 Simultaneously NMT Medical sued me for libel for comments that I made about the trial, but the legal case collapsed after four years when NMT went into liquidation.4,5 Before that happened, NMT admitted in legal pleadings that the paper contained a false statement that the authors had full access to the data. NMT admitted in legal papers that they had not allowed any author to have full access to the data. NMT also admitted that a vice president of the corporation had written part of the paper although that COI was not disclosed in the paper. In addition, the Financial Director of NMT confirmed that 10 of the 15 authors (NHS doctors) were paid as consultant by NMT, but only two disclosed financial COI in the paper.6
I asked the GMC to consider the misconduct of Dr Dowson, who was the first author of the paper in Circulation. A Fitness to Practise Tribunal found him guilty of several counts of misconduct including two counts of dishonesty in the MIST Trial and he was suspended from the Medical Register in 2015.7,8 However the decisions of both the GMC and the tribunal with respect to Dr Dowson’s COI in the MIST Trial seems perverse. The GMC put charges relating to Dr Dowson’s failure to disclose in the paper that he owned shares in NMT during the trial.9 The tribunal decided that share ownership valued at less than US$50,000 was too small to influence anyone and because Dr Dowson’s share ownership was worth less than US$50,000, his failure to disclose his share ownership was not misconduct.9 I am not sure that many would consider that sum insufficient to influence them. The tribunal also heard that Dr Dowson was paid about £2,000 per week during the trial as Chief Investigator with payments starting in July 2004 and continuing until March 2008.9 (I refused the offer of payments from NMT to avoid a COI.) In addition Dr Dowson was paid for each patient visit.9 Dr Dowson’s contract with NMT also included the possibility of a bonus depending on certain performance outcomes,9 which most would consider a COI. He was also paid for his role in another NMT Trial (MIST 3).9 I calculate that in total, Dr Dowson received in excess of £400,000 from NMT, which was very much greater than the value of his share ownership and much greater than $50,000. However the GMC put no charges over Dr Dowson’s failure to disclose in the paper his massive payments from NMT and as a result the tribunal could not make a finding with respect to his non-disclosure of COI other than his undisclosed share ownership.
The tribunal found that a vice president of NMT had made significant contributions to the published version of the paper, but that Dr Dowson, the corresponding author, had done no wrong by his failure to disclose the vice president’s contribution to the paper, because Dowson told them that the vice president did not wish to be an author.9 So it seems that in this case, the views of the GMC and the tribunal on what constitutes a COI and what constitutes proper disclosure of a COI is out of line with the opinion of many members of the public.
Dr Mullen was another member of the MIST Trial steering committee and second author on the published paper.2 He was a consultant cardiologist at the Royal Brompton Hospital. He was one of two doctors in the MIST Trial to disclose COI. His were that he owned shares in NMT and was paid by the corporation. The Royal Brompton Hospital where Mullen worked placed false information about the results of the MIST trial on its website.10 The website claimed that the treatment had a high rate of success.10 The trial actually found that the intervention did not improve migraine and that there was a high incidence of life threatening complications from it. At the bottom of that Royal Brompton Hospital webpage was a link to a webpage on NMT’s website. On that NMT webpage one could enter your postcode (the UK equivalent to a ZIP code) to find the nearest cardiologist who would implant NMT STARFlex device for migraine relief as a private patient even though the MIST Trial showed that the treatment was ineffective. The Royal Brompton Hospital stated that the information on the webpage had been put up at the request of Dr Mullen. The hospital removed the webpage when the dishonesty it contained was reported in Dr Ben Goldacre’s “Bad Science” column in the Guardian Newspaper.10 (I downloaded copies of the webpages before they were taken down.) The GMC took no action against Dr Mullen over his role in using a NHS website to make false claims about a medical device when he owned shares in the corporation marketing the device and he was paid by the corporation.
Circulation and the MIST Trial
I have said that Circulation published a correction to the MIST Trial paper.3 It took a lot of effort on my part to get the correction published. Circulation repeatedly refused to consider correcting the paper, but eventually after 18 months Circulation published a correction that is 700 words long and is accompanied by a new version of the paper and a data supplement. The corrected version still contains errors. I am concerned that the delay in publishing it was in part because of COI involving the editor of Circulation, Dr Joseph Loscalzo, and his colleagues. Loscalzo is the Hersey Professor of the Theory and Practice of Medicine at Harvard Medical School and Chairman of the Department of Medicine at the Brigham and Women’s Hospital. NMT was set up in Boston to manufacture and market devices for closing holes in the heart under licence from the Harvard affiliated Boston Children’s Hospital. The Brigham and Women’s Hospital and the Children’s Hospital face each other across Shattuck Street in Boston and are connected by walkways across the narrow street. Boston paediatric cardiologist, Professor James Lock invented the prototype device. The Children’s Hospital, Lock and other Harvard cardiologists received royalties from sale of the devices made by NMT. At the time I was attempting to persuade Loscalzo to correct the original version of the MIST Trial paper, cardiologists in his department were engaged in NMT’s CLOSURE I Trial of closure of persistent foramen ovale to prevent stroke, one of his colleagues was on the CLOSURE I steering committee and Loscalzo’s hospital is listed as receiving payments from NMT for participation in CLOSURE I.11
Other conflicts of interest at journals
Journals have other conflicts of interest. Major journals, such as the New England Journal and the Lancet, make large sums of money from reprint sales to industry and from advertisements to accompany research articles if they publish an article suggesting that a pharmaceutical or medical device corporation’s product is a therapeutic advantage. Dr Richard Smith, a previous editor of the BMJ has pointed out that major journals can receive more than $0.5 million dollars from commercial companies for reprint sales of a single article if the article shows their commercial product in a favourable light.12 So journals are far more likely to publish “positive” research about a product than they are to publish negative studies or studies of non-commercial treatments. The resulting biased content influences patient care.
Many editors fail to retract articles that are shown in official investigations to be fraudulent because they do not wish to adversely affect their journal’s reputation by drawing attention to the fact that they published fraudulent research.13 The possibility of financial loss from litigation by an author if they publish a notice of retraction also represents a conflict of interest. As a result research that was proved to be false continues to be cited and continues to influence future research and patient care.
Protecting institutions and colleagues
Research misconduct harms patients directly. It also wastes health care resources on useless treatments and it wastes the research resources of those tempted to do follow up research. Senior managers and senior doctors cover up research misconduct and other types of misconduct by doctors in order to protect the reputation of their institutions, which is a clear COI. Examples include:
Dr A K Banerjee was erased from the UK Medical Register after I reported his misconduct to the GMC. King’s College London had covered up his research fraud for 10 years. King’s College Hospital covered up the concerns about his clinical skills for the same period.14 When the West Midland’s Deanery had concerns that “he was the worse registrar we ever had” they sacked him from the training programme, but did not bring those concerns to the attention of the Royal College of Surgeons or the GMC.
Dr Clive Handler was suspended from the Medical Register after I reported that he had embezzled charitable research funds.15 The hospital’s trust board and its medical director, Prof Peter Richards, covered up the misconduct for years and agreed to conceal it from the GMC, even though Prof Richards was, at the time, a member of the GMC and chair of its Professional Conduct Committee.15 The Professional Conduct Committee was the GMC’s disciplinary committee (but is now renamed) and at the time Richards was effectively its chief “judge”.
In summary, I believe that undisclosed COI that adversely affect patient care are common in medicine.
- General Medical Council: Fitness to Practise Panel – Misconduct. Andrew John Dowson. Transcript March 24, 2006.
- Dowson A, Mullen MJ, Peatfield R, et al. Migraine intervention with STARFlex Technology (MIST) Trial: a prospective, multicenter, double-blind, sham- controlled trial to evaluate the effectiveness of patent foramen ovale closure with STARFlex septal repair implant to resolve refractory migraine headache. Circulation 2008;117:1397-1404.
- Dowson A, Mullen M, Peatfield R, et al. “Correction”, Circulation 2009;120: e71-e72, available from http://circ.ahajournals.org/cgi/content/full/circulationaha;120/9/e71
- Dyer C. Libel case against Wilmshurst collapses as firm goes into liquidation. BMJ 2011;342:d2646.
- Wilmshurst P. The effects of the English libel laws on medicine and research – a personal view. Prometheus 2011;29:67-71.
- Wood S. Former co-PI of MIST trial alleges data mismanagement, misinformation. Heartwire October 26, 2007.
- Dyer C. Migraine doctor is suspended for serious breach of standards. BMJ 2015:350:h982.
- Mr Justice Edis. Judgement in Andrew John Dowson versus the General Medical Council. 23 November 2015. Neutral Citation Number:  EWHC 3379 (Admin).
- Medical Practitioners Tribunal Service. “Transcripts: case of Dr Andrew Dowson”. (2014-15)
- Goldacre B. NMT are suing Dr Wilmshurst. So how trustworthy are this company? Let’s look at their website…. Guardian December 11th Available at: http://bit.ly/elGc8Q
- Furlan AJ, Reisman M, Massaro J, et al. Closure or medical treatment for cryptogenic stroke with patent foramen ovale. N Engl J Med 2012;366:991-9.
- Smith R. The trouble with medical journals. London, Royal Society of Medicine Press, 2006.
- Sox H, Rennie D. Research misconduct, retraction, and cleansing the medical literature: lessons from the Poehlman case. Annals Intern Med 2006;144:609-13.
- Wilmshurst PT. Poor governance in the award of honours and degrees: an extreme example of a systemic problem. BMJ 2016;352:h6952.
- Wilmshurst, P. Dishonesty in medical research. Medico-Legal Journal 2007;75:3-12.