Making medicines safer for all of us

Adverse drug events are now the fourth leading cause of death in hospitals.

It’s a reasonable bet they are an even greater cause of death in non-hospital settings where there is no one to monitor things going wrong and no one to intervene to save a life. In mental health, for instance, drug-induced problems are the leading cause of death — and these deaths happen in community rather than hospital settings.

There is also another drug crisis — we are failing to discover new drugs. [Read more...]

Archive for Study 329 – Page 2

Restoring Study 329: Letter to BMJ Jan 2016

Fiona Godlee

Editorial Note: This was our final letter in the correspondence that began here and ran through here. There has been no response from Dr Godlee.

28 January 2016

Dr Fiona Godlee
Editor, BMJ

Re: BMJ handling of conflict of interest

Dear Dr Godlee

Thank you for your email response of January 11 2016 to our letter of July 8 2015 about Dr Elizabeth Loder’s conflict of interest in relation to editing our Restoring Study 329 paper. However, we are disappointed with the outcome.

When Dr Loder and reviewers of our submitted paper suggested that we had conflicts of interest, we argued in our responses that conflict of interest is less important than access to data. With access to the data, others can decide whether or not perceptions of conflicting interests might have translated into actual bias. However BMJ made several more comments during the review process about our conflicts, and on July 6 2015 required that the most important part of our work, the reporting of adverse events, be independently checked because of perceived bias.

It was against this background that we wrote our letter of concern about Dr Loder’s apparent COI, intending to point out the double standards that appeared to us to be operating. We had hoped that you and Dr Loder would concede that there was indeed a perceptual issue with her organisation’s present and past relationship with GSK and that of her husband’s law firm, but would reassure us that there was no actual conflict because Dr Loder had worked hard for the publication of the paper. Instead BMJ dismissed the issues we raised as ‘so attenuated’ that requiring their declaration would be ‘absurd’, a dismissal that seemed inconsistent with BMJ’s attitude towards our perceived bias.

Study 329 will be an object of study for years to come, and it is important that the full story be available. The BMJ reviews and correspondence are the data about the handling of our paper. Reading them will make it possible for others to make a judgement as to whether any bias may have been involved. Yet these documents are not included under the ‘Peer review’ tab for our published paper, contrary to BMJ policy We request that they be posted there.

Meanwhile, the correspondence is available on the Restoring Study 329 website. We would be happy to include on that site a BMJ view as to how readers should read the various reviews and correspondence, or any other commentary that you might want to add.

Yours sincerely

Jon Jureidini
On behalf of 329 RIAT team

Restoring Study 329: Letter to BMJ

Editorial Note: The posts have gotten slightly out of sequence. This post should have preceded the responses from BMJ. It is the letter to which they were responding. This and other letters are on

8 July 2015

Dr Fiona Godlee

Dear Dr Godlee

Re: “Restoring Study 329: A randomised, controlled trial of the efficacy and harms of paroxetine and imipramine in the treatment of adolescent major depression”

I note your earlier comment about BMJ’s risks being ‘more editorial than legal’. Coincidentally when we were looking up Dr Elizabeth Loder’s profile in relation to her concerns about our handling of headache in our adverse events analysis, we became aware of a potential conflict of interest.

BMJ staff have understandably been very careful about any perceived conflict of interest on the part of our team, given that some of us have previously criticized GSK’s Study 329; we now have concerns about Dr Loder’s indirect but significant links with GSK.

  • Her hospital (Brigham and Women’s) received over $12 million in research funding from GSK in 2014, up from just over $100,000 in 2003


  • Dr Loder has made public statements favourable to GSK products, including: ‘For all of these reasons my mantra is that “You haven’t failed sumatriptan until you have failed to respond to a full dose of injectable sumatriptan given early in an attack!” There is also evidence that combining a triptan with an anti-inflammatory drug might improve the likelihood it will be effective’.

( ). This article was published just before GSK’s Treximet (combination triptan and anti-inflammatory) came on to the market.

  • Dr Loder’s husband, John M. Loder, is a partner in Ropes & Gray, a law firm retained by GSK in the US Department of Justice’s action against them, in which Study 329 was a central element

( ).

More recently, the law firm has supported GSK with its difficulties in China


Although, as Dr Loder’s COI declaration at BMJ points out, John Loder’s work is not in the healthcare field, as a partner in Ropes & Gray, he presumably profits directly from such work.

We believe that Dr Loder’s interests have been incompletely declared and that it might have been appropriate for her to recuse herself from involvement in the assessment of our paper to avoid any perception that GSK’s interests were being considered in BMJ’s deliberations.

While the timing for bringing these concerns to your attention is not ideal, we wanted to inform you as soon as possible after we became aware of these potential conflicts.

I look forward to hearing from you soon.

Yours sincerely

Jon Jureidini
On behalf of the RIAT 329 group

BMJ Award

Restoring Study 329: Correspondence with BMJ Jul 15 to Jan 16


Editorial Note: The letter from the Study 329 team to BMJ re Dr Loder featured in 50 Shades of Gray and in Conflicts of Interest. The actual letter and related correspondence is all on  The emails below and in the next post tell you what happened next.

From: Jureidini, Jon (Health)
Sent: Wednesday, 8 July 2015 10:32 PM
Subject: 329 and COI

Dear Dr Godlee

Please find attached a letter that raises some COI issues.


Jon Jureidini


On Wednesday, 30 September 2015, 23:52, “Jureidini, Jon (Health)” <> wrote:

Dear Fiona

Thank you for publishing and promoting our paper. We are very pleased that we and you saw this difficult process through.

We have not received a reply to the attached letter we sent you on 8 July, about our concerns about potential conflict of interest on the part of Dr Loder. We still think that the issues we raised are important, so we are seeking a response from you.




From: Fiona Godlee []
Sent: Thursday, 1 October 2015 6:46 PM
To: Jureidini, Jon (Health)
Subject: Re: Confidential

Dear Jon.

Thank you for your message. My apologies for not replying to your earlier message. I’m not sure how I missed it.

Let me discuss your concerns with Dr Loder. I will get back to you as soon as I have had a chance to do so.

All best wishes. Fiona

Dr Fiona Godlee FRCP

Editor in chief, The BMJ

London WC1H 9JR

44 (0)207 383 6002


—–Original Message—–

From: Jureidini, Jon (Health)
Sent: Wednesday, 4 November 2015 9:37 AM
Subject: FW: Confidential


Have you had a chance to evaluate our concerns?




On 14 Dec 2015, at 2:07 am, Jureidini, Jon (Health) <> wrote:


Just tidying things up before Christmas, and wanting to check if I will hear back from about this potential COI issue.

Happy new year



From: “” <>
Date: Wednesday, 16 December 2015 4:46 am
To: “Jureidini, Jon (Health)” <>
Subject: Re: Confidential

Dear Jon.

So sorry not to have got back to you about this. I referred it to our ethics committee which meets three times a year, but the last meeting was cancelled. They meet tomorrow and I should be able to give you their conclusion shortly afterwards. All best wishes. Fiona Dr Fiona Godlee FRCP

Editor in chief, The BMJ

London WC1H 9JR

44 (0)207 383 6002


From: Fiona Godlee []
Sent: Sunday, 3 January 2016 11:09 PM
To: Jureidini, Jon (Health)
Subject: Re: Confidential

Dear Jon,

Just to give you an update.

The ethics committee’s agenda was rather full for its December 16 meeting – having missed a meeting earlier in the year, and I am afraid we didn’t get to this item. However, I have asked our chair, Professor Marion McMurdo, to convene an additional meeting, which I hope will take place in the next few weeks, in order that we can get you a response to your concerns. My apologies for this further delay. I will be in touch again as soon as possible.

All best wishes,


Dr Fiona Godlee FRCP Editor in Chief, The BMJ BMJ, BMA House, Tavistock Square, London, WC1H 9JR T: 020 7383 6002 E: W:

Personal assistant:

Julia Burrell

T: 020 7383 6102



From: Fiona Godlee []
Sent: Monday, 11 January 2016 3:42 AM
To: Jureidini, Jon (Health)
Subject: Fwd: response to the Jureidini allegations

Dear Jon,

I am now able to respond in full to the concerns you raised last year about Elizabeth Loder’s possible conflicts of interest in relation to your reanalysis of the data from Study 329.

The ethics committee convened on Friday 8 January, 2015. It was chaired by the committee chair Professor Marion McMurdo, attended by committee members Julian Sheather, Elizabeth Wager, and Dr Adrian Sutton. Committee members who were unable to attend – Dr Rubin Minhas, John Coggon, and Dr Richard Hain – sent their comments by email. Also attending were myself, the BMJ’s Executive Editor Dr Theo Bloom, and the committee’s secretary Laura Templar who took the minutes.

The committee were in receipt of your letter and a response from Dr Loder, which is appended with attachments below.

The committee reviewed the issues you raised and, by a large majority, concluded that the possible conflicts you have identified are so attenuated by distance from Dr Loder – by virtue of the size and scope of her and her husband’s organisations, as to be highly unlikely to have impacted on her decision making. The committee felt that to require an individual to consider, manage or declare such attenuated links would take declaration of conflicts of interest to an absurd level, since such potential conflicts would be near impossible for an individual to keep track of or to be held responsible for. The committee was satisfied with Dr Loder’s response.

On the basis of this advice I therefore intend to take no further action on this matter.

Thank you again for raising it.

All best wishes, Fiona


Dr Fiona Godlee FRCP Editor in Chief, The BMJ BMJ, BMA House, Tavistock Square, London, WC1H 9JR T: 020 7383 6002 E: W:

Personal assistant:

Julia Burrell

T: 020 7383 6102


———- Forwarded message ———-

From: Elizabeth Loder <> Date: 16 December 2015 at 11:11 Subject: response to the Jureidini allegations To: Fiona Godlee <>

Dear Fiona,

You asked me to respond to Dr. Jureidini’s accusations of conflict of interest regarding connections to GSK. These were 1) that my hospital, Brigham and Women’s, receives research money from GSK; 2) that my husband is a partner in the law firm Ropes & Gray, which has done work for GSK and thus he profits directly from this connection; and 3) that I have made public statements favorable to GSK products.

With regard to the first matter, I don’t keep track of nor do I have any practical way of knowing about the hospital’s many sources of research support. BWH is one of the largest academic medical centers in the US and receives millions of dollars of research support from a vast array of companies and government agencies.

With regard to the second matter, Ropes & Gray is a huge international law firm with thousands of clients. The firm has approximately 1200 lawyers working from 11 different offices around the world and well over $1 billion in annual revenues. I didn’t have any knowledge of their work for GSK. My husband does not work in their healthcare division. He is a securities lawyer. He works exclusively with firms that manage money. He advises the management of these firms or their independent boards of directors and represents them in government enforcement matters.  Ropes & Gray’s work for GSK and the firm’s relationship with GSK accordingly had no influence on the editorial process. It seems entirely unreasonable to expect that any journal would make disclosures of all of their editors’ spouses’ employers’ business relationships. Those relationships are much too attenuated to be likely to have any bearing on the objectivity of the editorial process. Furthermore, there is no practical mechanism for gathering information about such remote relationships.

With regard to the third matter, the examples provided by Dr. Jureidini are taken out of context. They do not illustrate that I have spoken favorably about GSK products. You will note that in the live Q&A with the Washington Post, from which he pulled a quote, I use the generic name sumatriptan, and am answering a question posed by a reader who used the brand name.

The question was specifically about sumatriptan so it made sense to speak of that drug. It is a matter of fact not opinion that the injectable version of the drug, and early administration, improve outcome. Furthermore, sumatriptan is the only one of the seven marketed triptans that is available in an injectable formulation. The injectable formulation is produced by a number of generic manufacturers in addition to GSK. GSK makes a branded version of sumatriptan which is called Imitrex or Imigran but I did not use the brand name.  Additionally, you will see in the attached review article I wrote for NEJM that I specifically recommend the use of a generic version of sumatriptan because it is the least expensive treatment.

With regard to the statement about combinations of triptans and NSAIDs being more effective than either drug alone, again this is a matter of fact and not opinion. There are many trials comparing combination treatment to the single drugs — see attached editorial I wrote about the combination of frovatriptan and an NSAID. It is hard to understand how a statement about the effectiveness of combination therapy constitutes an endorsement for a particular fixed dose combination. In practice I do not use the GSK product treximet (sumatriptan + naproxen) because I think aspirin is a better choice than naproxen for use in combination therapy.

I hope these answers are satisfactory. Let me know if you have any other questions.



BMJ advances healthcare worldwide by sharing knowledge and expertise to improve experiences, outcomes and value. This email and any attachments are confidential. If you have received this email in error, please delete it and kindly notify us. If the email contains personal views then BMJ accepts no responsibility for these statements. The recipient should check this email and attachments for viruses because the BMJ accepts no liability for any damage caused by viruses. Emails sent or received by BMJ may be monitored for size, traffic, distribution and content. BMJ Publishing Group Limited trading as BMJ. A private limited company, registered in England and Wales under registration number 03102371. Registered office: BMA House, Tavistock Square, London WC1H 9JR, UK.


On Thursday, 28 January 2016, 6:08, “Jureidini, Jon (Health)” <> wrote:


Please find attached a response to this email.



Study 329: Response to Keller & Colleagues


Editorial Note: This post offers our response to the Keller letter. It was published in BMJ in early February. Again all material is available on The response does not make much of the fact that not all of the Als in Keller et al have lined up to sign their letter. One of the intriguing points in all this is tracking the line taken by GSK and by Keller and colleagues.

Jon N Jureidini, David Healy, Mickey Nardo, Melissa Raven, Elia Abi Jaoude, Catalin Tufanaru, Joanna Le Noury

Re: Restoring Study 329: Response to Keller

The response by Keller and selected colleagues [1] to our Restoring Study 329 article alleges three overarching faults[2]: bias and lack of blind ratings in relation to harms; lack of detailed methodology; and failure to consider the available methodological knowledge regarding paediatric depression from twenty-four years ago. Regarding the second issue, there is in fact a detailed explanation of all the methods in our paper and its RIAT Audit Record (appendix 1). We tackle the first and third issues below.


While there was uncertainty twenty-four years ago about the appropriate rating scale to use in pediatric depression trials, there were serious methodological problems in the conduct and reporting of Study 329 that have nothing to do with that uncertainty. Instead, in their reporting of efficacy in Study 329[3], and their defence of it, Keller and colleagues have asked that the field suspend many widely held tenets about clinical trial analysis, by asking us to do the following:

  • accept that the a priori protocol is not binding, and that changes can be made to the outcome variables while the study is ongoing, without amending the protocol with the IRB or documenting the rationale for the change
  • ignore the requirement to correct the threshold of significance for the analysis of multiple variables
  • ignore the requirement that when there are more than two groups, preliminary omnibus statistical analysis needs to be done prior to making any pairwise comparisons between groups – an integral part of the ANOVA analysis declared in the Study 329 protocol
  • allow the parametric analysis of rank-order, ordinal rating scales [CGI, HAM-D and K-SADS-L Depressed Mood Items] rather than the expected non-parametric methods specifically derived for this kind of data
  • allow 19 outcome measures to be added to the original eight at various times up to and after the breaking of the blind, purportedly according to an analytical plan ‘developed prior to opening of the blind’ (In spite of multiple requests, neither GSK nor Keller and colleagues have ever produced this analytic plan, suggesting that either it does not exist, or that it contains information unsympathetic to their claims.)
  • accept the dismissal of protocol-specified secondary outcomes and the introduction of rogue variables on the grounds that ‘the Hamilton Depression Rating Scale (our primary outcome measure) had significant limitations in assessing mood disturbance in younger patients’, when none of the protocol-specified secondary outcome measures that they discarded were based on the HAM-D, and two of the rogue measures that they introduced were HAM-D measures
  • accept the clinically dubious improvements in four of these rogue variables as evidence of efficacy. (Although these measures achieved statistical significance in the pre-defined eighth (final) week of the acute phase of the study, they did not do so in the weekly assessments over the previous seven weeks, a pattern unseen in any known antidepressant; we are working on another manuscript analysing Keller et al’s rogue variables.)

There was no ambiguity about the appropriateness of these methodological manoeuvres when Study 329 was conducted and reported. However, although some of these problems were obvious when the paper was first published, others were not apparent until we had access to the raw clinical data. This lack of transparency erodes confidence that RCTs will be conducted, analysed and reported free from covert manipulation.

Furthermore, Keller and colleagues also failed to report on the continuation phase of Study 329, even though that was a protocol-specified outcome. A report of this phase is almost ready for submission by us.


With regard to harms, Keller and colleagues are simply incorrect in many of their claims about our purported bias and lack of blind ratings.

First, our paper makes it clear that both coders in the re-analysis were blind to randomisation status.

Second, there was no ‘re-scoring’. This odd choice of words raises doubts that Keller et al have much expertise in analysing harms. We used a dictionary that adhered much more closely to the verbatim terms used by the face-to-face interviewers. The fact that Keller and colleagues say that we have labelled emotional lability as suicidality makes us wonder if they have seen the individual patient level data; it was the SKBs coders who came up with the term ‘emotional lability’, not the face-to-face interviewers, whose verbatim terms were of suicidal thoughts and behaviour. Simply using the verbatim terms that the named authors or their colleagues had used when faced with these adolescents reveals a striking rate of suicidal events. To argue that our return to these verbatim terms was arbitrary is bizarre.

Third, we made it clear there is unavoidable uncertainty in coding, and we invited others to download the data we have made available and juggle it to see if they can improve on our categorisation of the data. In our correspondence with BMJ, we made it clear that there are items that GSK could argue are more appropriately coded differently. We would be receptive to a rationale for alternate coding of certain items that is cogently argued rather than simply asserted, but our hunch is that a disinterested observer reviewing the coding as presented by GSK across all 1500 adverse effects in this study (or 2000+ if we include the continuation phase) would conclude that our efforts are a better representation of the data.

Fourth, reading our paper makes it clear why we reviewed the clinical records of 93 subjects; these were the subjects who dropped out or became suicidal. Our claims about underreporting of adverse events stand independently of that non-random sub-sample.

With regard to suicidal ideation and attempts, Keller et al. refer to a reanalysis by Bridge and colleagues, which found that there was no significant difference in suicidality between paroxetine and placebo. But Bridge et al. relied on Keller et al.’s misleading 2001 report.

With regard to bias, our point was that the best protection against bias is rigorous adherence to predetermined protocols and making data freely available. We, like everyone, are subject to the unwitting influence of our bias. The question is whether the Keller et al publication of 2001 manifests unconscious bias or deliberate misrepresentation.

The original and restored studies, the study data, reviews and responses are all available at, offering a broad range of options when it comes to consideration of authorship, research misconduct and the newly described species, ‘research parasite’[4].


1 Keller MB, Birmaher B, Carlson GA, Clarke GN, Emslie GJ, Koplewicz H, Kutcher S, Ryan N, Sack WH, Strober M. Restoring Study 329: efficacy and harms of paroxetine and imipramine in treatment of major depression in adolescence. Response from the authors of the original Study 329. BMJ 2015;351:h4320

2 Le Noury J, Nardo JM, Healy D, Jureidini J, Raven M, Tufanaru C, Abi-Jaoude E. Restoring Study 329: efficacy and harms of paroxetine and imipramine in treatment of major depression in adolescence. BMJ. 2015 Sep 16;351:h4320.

3 Keller MB, Ryan ND, Strober M, et al. Efficacy of paroxetine in the treatment of adolescent major depression: a randomized, controlled trial. J Am Acad Child Adolesc Psychiatry 2001;40:762-72.

4 Longo DL, Drazen JM. Data sharing. N Engl J Med. 2016;374:276-7.

Competing interests: As noted in our RIAT paper

Restoring Study 329

Study 329: Nobody Pinned Anything on Us

Editorial Note: Following the publication of Study 329 Restored, Martin Keller and his co-authors gave an initial response in which they said Nobody Pinned Anything on Us and also that they would respond in more detail later. It took four months for this response to appear. Our Response will follow in the next post and after that we will update our background correspondence with BMJ.

This and earlier correspondence and all the data from the Study 329, BMJ reviews of the study, a timeline of the history of SSRIs and all controversies linked to these drugs is all available on

There are two further Study 329 articles making Study 329 the most intensively studied Clinical Trial ever. The Study 329 website will make all this material available for anyone who wants to see how clinical trials operate – this study is not an anomaly, it is standard industry practice.

The picture above was dubbed Three Amigos by its creator. It features Charlie Nemeroff, Marty Keller and Alan Schatzberg. A picture that is at least as appropriate is below linked to the conflict of interest declaration.

BMJ 2015; 351 doi: (Published 16 September 2015) Cite this as: BMJ 2015;351:h4320

18 January 2016

Martin B Keller, Boris Birmaher, M.D., Gabrielle A. Carlson, MD, Gregory N. Clarke, Ph.D., Graham J. Emslie, M.D., Harold Koplewicz, M.D., Stan Kutcher, M.D., Neal Ryan, M.D., William H. Sack, M.D., Michael Strober, Ph.D.

attn: Martin B Keller, MD, 700 Butler Drive, Blumer 120, Providence, RI 02906, USA

Re: Restoring Study 329: efficacy and harms of paroxetine and imipramine in treatment of major depression in adolescence. Response from the authors of the original Study 329

The BMJ article entitled “Restoring Study 329: efficacy and harms of paroxetine and imipramine in treatment of major depression in adolescence”[1] reanalyzed data from the original Paroxetine 329 study[2], a double-blind placebo controlled comparison of paroxetine to imipramine. Paroxetine 329 was designed between 1991 and 1992. Subject enrollment began in 1994, and was completed in 1997. Academic psychiatrists designed the study, with very little change by GSK, which funded the study in an academic / industry partnership. The goal of the study was to advance the treatment of depression in youth, rather than primarily as a drug registration trial.

Overarching issues with the “Restoring Study 329” include:

  1. Authors of “Restoring Study 329” evidenced both bias and a lack of blind ratings. In a recent article about “Restoring Study 329” in the Chronicle of Higher Education, Dr. Jureidini is quoted as saying: “We don’t think we’ve done the definitive analysis. It’s not something that can be done absolutely objectively, particularly the interpretation of harms. We can’t protect ourselves completely from our own biases.”[3]     Biases are a serious consideration for Restoring Study 329 because Dr. Jureidini, as he declares in a Footnote on the subject of “Competing interests”, served as an expert witness for plaintiff’s lawyers in legal suites against GSK related to Study 329. In that work Dr. Jureidini would have studied all available data looking at both efficacy and suicidal side effects, using many different approaches to best capture any potential harms.
  2. The “restoring invisible and abandoned trials” (RIAT) approach to reanalyzing published studies may provide general guidelines but we could not find publications or available working RIAT documents on detailed protocols. Lack of detailed methodology is a serious concern because there is general consensus in the field that there is not, nor never will be a single correct approach to reanalysis. Small differences in analysis frequently make big differences in statistical results and conclusions.
  3. “Restoring Study 329” did not consider available knowledge 24 years ago, when Paroxetine 329 was developed and performed. Clinical research methodology has evolved considerably in the past two decades. These aspects are addressed in comments by established investigators not involved in Paroxetine 329. For example, Referring to “Restoring Study 329” as reported in Psychiatric News Alert [4] Mark Olfson said, “However, the new reanalysis does not alter the totality of clinical trial evidence that continues to support the safety and efficacy of SSRIs for adolescent depression.” And Daniel Pine said “We have known for some time that antidepressant medications have both significant benefits for some children as well as significant risks for other children. This new analysis really does nothing to change this knowledge, and provides no new insights into what we have known about these medications for the past few years.”


Antidepressants considered as a group are superior to placebo for the treatment of anxiety disorders and for depression in adolescents, with similar overall response rates in anxiety and depression. [5]

The two primary outcome measures in Paroxetine 329, did not reach statistical significance. The abstract of the published paper noted: (1), “The two primary outcome measures were endpoint response (Hamilton Rating Scale for Depression [Ham-D] score ≤ 8 or ≥50% reduction in baseline HAM-D) and change from baseline HAM-D score.” In Table 2, the p value for the first primary endpoint (Ham-D score ≤ 8 or ≥50% reduction in baseline HAM-D) was reported at p < 0.11 for paroxetine versus placebo. In the same table, the p value for change in HAM-D total score is reported at p < 0.13. While both outcomes were in the direction of a better response for paroxetine over placebo; neither reached our critical alpha level of 0.05. This is clear in the abstract and text of the publication.

In the interval from when we planned the study to when we approached the data analysis phase, but prior to the blind being broken, the academic authors, not the sponsor, added several additional measures of depression as secondary outcomes. We did so because the field of pediatric-age depression had reached a consensus that the Hamilton Depression Rating Scale (our primary outcome measure) had significant limitations in assessing mood disturbance in younger patients. Taking this into consideration, and in advance of breaking the blind, we added secondary outcome measures agreed upon by all authors of the paper. We found statistically significant indications of efficacy in these measures. These secondary outcomes were clearly reported as separate from the negative primary outcomes.

Thus, the authors of “BMJ-Restoring Study 329” were incorrect in stating that “Both before and after breaking the blind, however, the sponsors made changes to the secondary outcomes as previously detailed. We could not find any document that provided any scientific rationale for these post hoc changes and the outcomes are therefore not reported in this paper.” Rather, secondary outcomes were decided by the authors prior to the blind being broken. Secondary outcome measures are frequently, and appropriately, included in study reports even when the primary measures do not reach statistical significance. The authors of “Restoring Study 329” state “there were no discrepancies between any of our analyses and those contained in the CSR [clinical study report]”. The disagreement on treatment outcomes rests on this arbitrary and non-blind dismissal of our secondary outcome measures.

In the abstract we stated “Conclusions: Paroxetine is generally well tolerated and effective for major depression in adolescents.” In this sample and with the state of knowledge at the time, it was justified and appropriate.

Our goal was to learn as much as possible about the use of this compound in youth, so that we could understand what role it (and by extension other SSRIs) could play in the treatment of adolescents with MDD. For us the question was given (1) the data distribution and statistical results for efficacy and side effects that we saw in the study, and (2) that there was well replicated research evidence of efficacy and relative safety of paroxetine in adults, what conclusions should be drawn from our data? The clinical outcomes were substantially in the right direction, with a number of them reaching the 0.05 level of statistical significance. The clinical results comparing paroxetine placebo to and imipramine to placebo paralleled those reported in adults. Side effects were similar to what was known in adults. Thus we reached, the conclusions reported.


The “Restoring Study 329” reanalysis uses the FDA MedDRA approach to side effect data, which was not available when our study was done. That one can do better reanalyzing adverse event data using refinements in approach that have accrued in the 15 years since a study’s publication is unsurprising and not a valid critique of Paroxetine 329 study as performed and presented.

We emphatically disagree with the “Restoring Study 329” position that statistics are not useful in understanding adverse side effects and that each individual reader should decide for herself when a difference in rates of adverse side effects is meaningful. Statistics offer several approaches to the question of when is there a meaningful difference in the side effect rates between different treatments.

Specific methodology problems in the reanalysis of the “harm” data are as follows: 1) The authors choose a non-random subsample of 85 subjects who were withdrawn from the study plus 8 subjects whom the authors labeled “suicidal” based on their inspection of the data; 2) a different instrument was utilized to re-score the harm effects and only one of the authors was trained in the scoring of the instrument; 3) some side effects were arbitrarily interpreted (e.g., upper respiratory symptoms were labeled as “dystonia” and emotional lability labeled as “suicidality”); 4) in the original paper, side effects were analyzed only during the acute phase, but in the reanalysis, the authors analyzed them during the acute phase, as well as the tapering and follow up phases of the study; and 5) in the original study patients were interviewed face-to-face whereas the reanalysis was based only on the interpretation of the data; and 6) importantly, the two authors were not blind to patients’ randomization status.

Suicidal ideation and attempts

Our field’s understanding of how to approach analysis of suicidal ideation, suicide attempts, and completed suicide has advanced enormously since publication of study 329. Two definitive reanalyses of the suicidality with antidepressants in adolescents include: 1). The 2003 FDA reanalysis of all RCT data of SSRI studies in youth for all indications.[6] In the FDA analysis the average risk ratio for SSRI versus placebo treated subjects was 1.96 (CI: 1.28-2.98). Considered separately Study 329 did not reach statistical significance for increased suicidality (CI: 0.42-33.21). 2). The methodologically superior reanalysis by Bridge and colleagues also found that in study 329 there was no significant risk difference between paroxetine and placebo. [7]

Paroxetine treatment in youth does not appear to significantly differ from other SSRIs in the risk of suicidal ideation or attempts and whether SSRIs increase or decrease completed suicide remains an open question. [8-12]


We strongly support efforts to make anonymized raw data from scientific studies available for reanalysis. The validity of “Restoring 329”, however, is doubtful because of author bias and substantial problems with RIATT methodology. To describe Paroxetine 329 as “misreported” is pejorative and wrong based on both state-of-the-art research methods 24 years ago, and retrospectively from the standpoint of current best practices.


Martin B. Keller, M.D. Boris Birmaher, M.D. Gabrielle A. Carlson, MD Gregory N. Clarke, Ph.D. Graham J. Emslie, M.D. Harold Koplewicz, M.D. Stan Kutcher, M.D. Neal Ryan, M.D. William H. Sack, M.D. Michael Strober, Ph.D.

  1. Le Noury J, Nardo JM, Healy D, et al. Restoring Study 329: efficacy and harms of paroxetine and imipramine in treatment of major depression in adolescence. BMJ 2015;351:h4320.
  2. Keller MB, Ryan ND, Strober M, et al. Efficacy of paroxetine in the treatment of adolescent major depression: a randomized, controlled trial. J Am Acad Child Adolesc Psychiatry 2001;40(7):762-72.
  3. Basken P. Landmark analysis of an infamous medical study points out the challenges of research oversight. Secondary Landmark analysis of an infamous medical study points out the challenges of research oversight 2015.
  4. Reanalysis of JAACAP Study on Paroxetine Sparks Controversy. Secondary Reanalysis of JAACAP Study on Paroxetine Sparks Controversy 2015.
  5. Bridge JA, Iyengar S, Salary CB, et al. Clinical response and risk for reported suicidal ideation and suicide attempts in pediatric antidepressant treatment: a meta-analysis of randomized controlled trials. Jama 2007;297(15):1683-96.
  6. Hammad TA, Laughren T, Racoosin J. Suicidality in pediatric patients treated with antidepressant drugs. Arch Gen Psychiatry 2006;63(3):332-9.
  7. Bridge JA, Birmaher B, Iyengar S, et al. Placebo response in randomized controlled trials of antidepressants for pediatric major depressive disorder. Am J Psychiatry 2009;166(1):42-9.
  8. Gibbons RD, Brown CH, Hur K, et al. Early evidence on the effects of regulators’ suicidality warnings on SSRI prescriptions and suicide in children and adolescents. Am J Psychiatry 2007;164(9):1356-63.
  9. Olfson M, Shaffer D, Marcus SC, et al. Relationship between antidepressant medication treatment and suicide in adolescents. Arch Gen Psychiatry 2003;60(10):978-82. 10. Gibbons RD, Hur K, Bhaumik DK, et al. The relationship between antidepressant prescription rates and rate of early adolescent suicide. Am J Psychiatry 2006;163(11):1898-904. 11. Valuck RJ, Libby AM, Sills MR, et al. Antidepressant treatment and risk of suicide attempt by adolescents with major depressive disorder: a propensity-adjusted retrospective cohort study. CNS Drugs 2004;18(15):1119-32. 12. Gibbons RD, Mann JJ. Strategies for quantifying the relationship between medications and suicidal behaviour: what has been learned? Drug Saf 2011;34(5):375-95.

We merely supplied the patients

This picture comes from Carl Eliot based on Johanna Ryan’s noting a curious phrase in the conflict of interest statement from Gabrielle Carlsson below.

Competing interests: Please see attachments to this response – the attachments are available on


Study 329: Big Risk

Light bulb hit by hammer

Editorial Note: This is the Fourth Crusoe Report.

“Death waits for these things like a cement floor waits for a dropping light bulb”

Big pharma

Study 329 seems to fit the classic picture. It has Big Pharma ghostwriting articles, hiding data, corrupting the scientific process and leaving a trail of death, disability and grieving relatives in its wake.

Pharma began in the middle years of the nineteenth century when advances in the chemical and biological sciences underpinned the development of analgesics and antipyretics and later antibiotics. Within medicine, these were the first drugs that reliably worked. Within business, they led to developments in patenting and trade-marking, big profits and the emergence of an industrial-scientific complex.

The Second World War put a premium on pharmaceuticals. The development of Atabrine for malaria and Penicillin for everything else helped win the War – a lesson not lost on Governments. Pharma had become a strategic industry.

Government investment led to a cornucopia of new treatments in the 1950s and 1960s that transformed medicine. Lives were saved that would have otherwise been lost, research flourished, and the specter of premature death began to lift

Previously drugs had been developed in the pharmaceutical divisions of chemical companies. These divisions were now hived off as independent companies (Little Pharma) and these new companies rapidly became the most profitable on the planet.

In the late 1960s and early 1970s, Little Pharma called in management consultants in a bid to keep the goose laying the golden eggs. These outfits advocated outsourcing clinical trials to Clinical Research Organizations (CROs) and medical writing to Ghost Writing Agencies. They also advocated having businessmen and marketers as CEOs of the company rather than chemists or scientists or medics. They insisted on five year development plans that put a premium on the selling and reselling of popular diseases where even less effective products could be made into blockbusters rather than developing medicines for conditions that had no treatments. If the company was in the business of making profits, this switch in focus was a no brainer.

This advice created the very model of a modern major pharmaceutical – out of which came Big Pharma and Study 329.

But is this the whole story?

Big risk

The nineteenth century also gave rise to another profitable industry – the linsurance industry which is now a broader risk management industry.

The collection of data by the first life insurance companies in the eighteenth century led in the nineteenth century to the creation of public health and the idea of preventive medicine. (This will come as an extraordinary claim to many, but the underpinnings of this can be seen in books like The Creation of Psychopharmacology).

The interests of the insurance industry to manage risks laid the basis for epidemiology and an interest in numbers in health. It led to calls to eradicate filth even before the germ theory had established what it might be about filth that caused problems. Today the preventive impulse in public health medicine leads to calls to eliminate poverty – which brings medicine into politics and politics into medicine.

The first public health physicians in the early nineteenth century were called Hygienists or Sanatarians. In addition to campaigning against filth, and the adulteration of food, and for temperance, the Hygienists in Germany and Britain advocated strongly for pensions as a public health measure which in turn furthered the growth of the insurance industry. And ultimately healthcare today worldwide is (or will be with the latest Trade Treaties) delivered through insurance schemes of one sort or the other.

In the second half of the nineteenth century, therefore, the growth of the economy and of the modern world got a huge boost from both the emerging biomedical and epidemiological sciences and their linked industries.

We celebrate the gains that medicine made in the 1950s that stemmed from the discovery and production of new drugs. We miss the transformation of medicine that data from yearly insurance check-ups produced in the 1960s. These data created the notion of risk factors such as hypertension, raised cholesterol levels and raised blood sugar. From a risk management point of view the data put a premium on treating risk factors – giving drugs to people the vast majority of whom had nothing wrong with them. This was not a Pharma plot – or not solely a Pharma plot. The story has been told in Jeremy Greene’s Prescribing by Numbers.

And just as the dynamics of modern corporations transformed pharmaceutical companies from companies at the forefront of an effort to discover drugs that treat the disorders that need treating for which we have no treatments into companies that focus on the production of drugs that make a profit, so also these dynamics changed the insurance industry. It changed from an industry that viewed the environment as risky and aimed to ensure our safety from these threats and to provide our families with a safety net in the event of our death, into an industry that located risks within us and wanted to protect itself from us. Big Risk will refuse to cover anyone who is in fact risky.

The marriage of pharma and risk

The pharmaceutical and insurance industries were initially not perfect bedfellows. The insurance industry was hostile to individual doctors doing whatever they liked such as using the latest drug. But most doctors believed that medicine cannot be practiced by numbers – that the duty of the doctor is to the patient in front of her rather than to the population.

But still the early interplay between science and business within the health domain and between preventive medicine and biomedicine worked to the advantage of all. New drugs liberated us from the specters of disease. Insurance highlighted things we could do to safeguard ourselves, our families and communities.

But the situation became more ambiguous as the twentieth century went on. With the virtual elimination of mortality linked to bacterial infections some of the greatest hazards to health came from pollution linked to other new industries such as the lead and tobacco industries. Tackling the health problems that stem from industries that are important to the economy and jobs cannot expect to mobilize the same degree of community or political support, as fighting Tuberculosis or Ebola can.

In addition, the links industry developed with science in the nineteenth century left it well placed, and financially more able than academia, to mount epidemiological studies in the twentieth century. This awareness of the benefits of research along with greater resources to sponsor studies was deployed to great effect for instance in the defense of tobacco smoking and lead where industry demonstrated it had learnt to exploit the radical doubt that drives science.

There is also the tricky balance of working out where politics ends and medicine begins. There were vicious disputes in the nineteenth century between biomedicine and public health over filth. Mainstream medicine didn’t see it as its job to clear up filth. Public health insisted it was. Mainstream medicine discovered germs and embraced the elimination of germs as a legitimate medical contribution. Many in public health held out against the germ theory.

Is eliminating poverty (the modern equivalent of filth) a medical task? Or should medicine make its contribution by recognizing that many poor (a.k.a. non-white) children live in slums that still have lead in their paint and that lead poisoning knocks several points off a child’s IQ and is associated with criminality and that the medical contribution is to flag this up and find ways to eliminate lead poisoning in the face of determined efforts by a powerful industry to block them – leaving poverty to politicians?


The shipwreck of the singular

Whatever balance you opt for in the above disputes, today, as has ever been the case, when you take a problem to a doctor for help, both you and she expect to be able to draw on the best evidence to solve your problems.

In 1990 at the start of the Big Pharma era, you and your doctor lived in a world where medical issues were found in journals, textbooks and a small number of popular books. Today there is likely to be a health story on the front page of the newspaper, with an entire section inside devoted to health. The amount of health related material on the Web is second only to pornography and even pornography is grist to the medical grind.

The political has become personal in an extraordinary fashion.

Unlike any time in medicine hitherto, when you go to a doctor today you will have to take your place in a queue of people, many of whom have been summoned to a consultation by a clinic screening for a wide range of things none of which bother the people who have been summoned. They will come to the clinic unaware of any problem but will leave with diagnoses and on medication. The doctors call them in not out of concern for them but because the doctors have targets to meet in order to get reimbursed – targets set by Big Risk.

Big Pharma play on this pitch but it’s Big Risk that draws the lines and sets up the goalposts.

When you do get in to see the doctor, you’ll find someone who adheres to Guidelines. She will do so in good faith, figuring this the way to bring the best evidence to bear on your case. She will not recognize she is being guided to see problems in certain ways and to deliver on patent treatments. She will not be treating you according to the Guideline for Treating You. If there were such a Guideline, the first point would be pay little if any heed to Guidelines for treating diabetes, or hypertension or depression or the menopause. (See The Macbeth Test).

If the problem is a mental health one, both you and your doctor are likely to be aware of conversations denigrating biological reductionism claiming that it risks dehumanizing clinical encounters. In practice however biology contributes almost nothing to clinical encounters about nervous problems.

These encounters are being dehumanized but the problem lies with an informational reductionism linked to the use of rating scales and operational criteria.

Within the mental health domain, a great deal of public discourse claims the medical model is inappropriate, diagnosis unhelpful, and the word “patient” to be abjured along with an increasingly long string of politically correct replacements. But in practice patients seek diagnoses, and the appeal of the language of chemical imbalances lay in the fact it was destigmatizing. The allure of biomedicine lies in its promise of treatments that work.

But for the first time in a century, today’s first line treatments are likely to be less effective than yesterdays.

In all of medicine, one of the greatest sources of morbidity and mortality – perhaps the greatest – now stems from the treatments patients have been put on, the multiplication of hazards by polypharmacy and the denial of the possibility of risks by corporations whose own health depends on the continuing consumption of the greatest possible number of medications by the greatest possible number of patients from the earliest possible age.

In most of the medical and lay media, Big Pharma is the only whipping boy for these evils. But is it?

Epidemiological methods are used to deny these treatment related risks. RCTs come from Big Risk not Big Pharma. When they were introduced first they were a way to contain the pharmaceutical industry. Pharma lobbied vigorously against them. They began as a Risk Management Tool but have become the gold-standard way to hide risks – as Study 329 shows so dramatically.

Economically you might have thought it was in Big Risk’s interests to map out the epidemiology of treatment induced morbidity – the problems treatments cause. But it doesn’t do this. Big Risk’s traditional methods of prevention – Guidelines and RCTs – don’t work for treatment induced problems. And why solve a problem that generates more turnover?

Meanwhile so uncertain has Big Risk made access to care – so shredded has the safety net become – that any suggestions that consuming fewer drugs might be healthier are drowned out for most people by concerns about access to medicines. The ACLU for instance will not take up the issue of whether treatment induced violence might have led to inappropriate incarceration for fear it might complicate their efforts to ensure that prisoners have access to healthcare.

Just as a balance in drug development has tipped so that it no longer serves medical treatment, so also a balance within prevention has been perverted.

Big Risk should make it impossible for Big Pharma to take separate patents on drugs as similar as two drops of water by refusing to reimburse the second drop of water. It should make it impossible to ghost write over 90% of the literature for on-patent drugs and to sequester the data from clinical trials, in contravention of the fundamental norm of empirical science – but it doesn’t do any of these things.

Big Risk underpins a comprehensive failure to diagnose and treat in the face of morbidity and mortality on an epidemic scale. Before blaming Capitalism, the problem is the market isn’t working. It’s Big Risk that should make the market work and they aren’t. What we are looking at is the behavior of Corporations. This behavior is shaped by Rules and at the moment the Rules are not working for us.

Medicine is no longer what it was. Your doctor needs to relearn the skills of listening to, seeing and touching you. She will have to engage with a biology that recognizes the brain as a social organ rather than with the biobabble that stems from Pharma marketing. She will have to ignore an epidemiology that figures you can design authoritative RCTs without understanding the biology being investigated (most RCTs).

Both Big Pharma and Big Risk justify the status quo by saying they don’t want to impinge on the sanctity of the doctor patient relationship. So she will have to be able to take the dynamics of industrial power into account and Industry will have to figure she is made of the Right Stuff – unless we can find a way to rescue her from the pot in which she is now stewing.

Until such treatment becomes possible, we are all shipwrecked. We are all Crusoe.

Death waits


“This generation thinks that nothing faithful, vulnerable, fragile can be durable or have any true power. Death waits for these things as a cement floor waits for a dropping light bulb. The brittle shell of glass loses its tiny vacuum. This is how we teach metaphysics on each other”.

The quote is from Saul Bellow’s Herzog. In Bellow’s imagery, the vacuum in the dropping light bulb contains our hopes, our aspirations, our fears. Big Pharma and Big Risk were once our allies in keeping our hopes alive – in keeping our children alive and well. They are now a threat. And of the two – Big Risk is the bigger threat.

Study 329: BMJ Transparency


Two weeks ago The BMJ ran an editorial by Richard Smith (former editor) and Fiona Godlee (current editor) on the retraction of a 1989 article by R K Chandra under the heading of A Major Failure of Scientific Governance.

While making money from the publication of pharmaceutical company trials, and in the face of a complete failure by industry to adhere to basic scientific norms and make data available, BMJ and other journals, although BMJ in particular, have run a series of articles on supposed Academic Fraud. These articles feature instances of fraud sometimes as bizarre as researcher claiming he cannot show the data as it was eaten by termites. A common theme, as with Chandra, is the academic community are held back from tackling the issue by fear the dude will sue. The universal feature is that these are academic studies and academic fraud is an issue in academia. Talks on this topic are often given by representatives of pharma such as Frank Wells for the Association of the British Pharmaceutical Industry (ABPI).

A day after the editorial appeared, Leonie Fennell submitted a Rabid Response (RR) to the editorial that used several of the words of the editorial itself. This RR was posted. Then the response disappeared. A while later it was re-posted. This time it was there long enough for someone to like it and for the screenshot above to be taken. But establishing its reality by liking it was apparently not enough. It was removed again, and has not reappeared since.

This is a comment in limbo. Until recently many Irish believed that stillborn babies went to limbo and the vague promise their parents held on to was their children would be saved on the Last Day.

“To err is human, to cover up is unforgivable, and to fail to learn is inexcusable”. I agree with both BMJ editors (past and present) on the latter, yet I still have some concerns. Surely if the BMJ had actually learned from this, it would have been more proactive with Study 329, where scientific fraud has so obviously once again prevailed. Is the BMJ, as David Healy suggests, terrified when publishing anything that might make a pharmaceutical company uncomfortable? It’s interesting that an earlier article regarding Study 329 was reviewed and turned down by the BMJ (reviewers included the former editor in chief, Richard Smith).

The findings of Study 329, that Paroxetine was ‘safe and effective for adolescents’ led to the widespread medicating of children with Selective Serotonin ReUptake Inhibitors, subsequently causing many deaths. Saying that universities, authorities and the world of science have a chance to learn from the Chandra case is all well and good, but what have the BMJ actually done to right this latest, very evident wrong? Brown University, GSK and Keller et al are digging their heels in. The BMJ need to act now while there is still has time to put its own house in order. ‘Good men doing nothing’ is just not good enough. The BMJ’s current reputation as one of the leading medical journals is at stake here”.

Study 329: 50 Shades of Gray


Editorial Note: None of these posts about Study 329 should be taken as representative of a RIAT view, especially this one. See Study 329: Conflicting Interests for a prequel to this post and to make sense of the last comment.

Fiona Godlee to RIAT July 6 2015
Re: Study 329

Dear Dr Jureidini,

“Many thanks for your letter. I quite understand you concerns. You are right to say that there are few or no precedents against which to compare this article. We ourselves are feeling our way, both with the RIAT process since this is the first full RIAT research paper we will publish, and with the specific challenges posed by this particular study and you as the paper’s authors. I want to stress that we are proceeding in good faith with the clear aim of publishing the article as soon as possible provided we can do so safely”. …

All best wishes

Fiona Godlee FRCP
Editor-in-Chief The BMJ

In utero

The editor dealing with Study 329 through its year-long gestation in the belly of the BMJ was Elizabeth Loder. The paper improved considerably during this time but there was almost nothing constructive from BMJ that led to any of these improvements – readers can examine for themselves the various letters, reviews and revisions, bearing in mind that from our side the correspondence is written under constraint . Indeed almost all inputs from BMJ produced outcomes that likely made the editorial staff ever more uncomfortable.

As the features of Study 329 took shape in the BMJ womb, the New England Journal of Medicine (NEJM) had a series of articles suggesting that concerns about Conflict of Interest had gone too far. Some journals, NEJM said, were in a state of moral panic; they were hunting for witches. They should have run their articles at Halloween.

The BMJ in an editorial by Elizabeth Loder and Fiona Godlee led the way in responding with outrage to the NEJM, suggesting that this was not the time to step back from seeking transparency on conflicting interests. []. Ideally we need to go further and find ways to bring non-financial conflicts of interest into the frame. The medical literature is clearly still a mess and this can only be because we have not been zealous enough about conflicting interests.

The alternate view, not popular with some “progressives” who don’t do science, is that access to data is more important than access to information about conflicts of interest. It is only when there is access to the data that we can see if interests are conflicting and take that into account. Science needs conflicting interests – financial and non-financial. Problems don’t get solved unless someone is motivated for some reason. We need the bias that pharmaceutical companies bring to bear in their defense of a product, along with the bias of those who might have been injured by a treatment. Both of these biases can distort the picture but it’s when people with differing points of view agree on what is right in front of their noses that we can begin to have some confidence about what we have.

Outsiders have conflicts; Insiders don’t

Sixteen years ago I sent a paper to BMJ based on Lilly documents that had come into the public domain because of a Court Case. BMJ’s initial response was that they should perhaps publish given that they had published a Lilly article, the Beasley et al 1991 article, that had done a good deal to create the antidepressant and suicide problem. But Richard Smith resiled from this position. In later correspondence, he stated BMJ would never publish anything from me on this topic.

The real issue was that BMJ were lily-livered and were prepared to cite my supposed conflicts as their escape route even though the data was fully in the public domain and publishing would have laid bare whether my judgments were unduly conflicted. Another journal took the paper promptly.

At the time of the July 6 email above, BMJ stated they were thinking about sending parts of our material out to a third party to review. In fact they had already sent the material out for review when they told us they were thinking about doing so. We objected and made it clear that this was unacceptable and when the reviews came back, we paid little heed to them.

On July 8,  we emailed BMJ raising issues covered below.

A headache

Some of the drama in Restoring Study 329 centered on headaches. GSK coded headaches and dizziness one way in the original Study Report and a different way in the Keller et al paper. Headaches and dizziness were so common that moving them from Body as a Whole to Nervous System laundered out the signal not just from suicide events (coded as emotional lability) but from pretty well all psychiatric events.

We were less interested in the question of where exactly headaches and dizziness should be placed and much more interested to make sure readers of the article were aware that it was possible to move them around and in so doing to “re-author” the findings. Our concern wasn’t to show that deception had happened but rather to show how people might be misled – perhaps inadvertently.

Elizabeth Loder was upset. She might have regarded our approach as cavalier or she might have missed the point because of a conflicting interest.

EL is a professor of neurology and a migraine expert based in Boston. She has several books on headaches – something we hadn’t thought to check before July 6.

She has been the President of and serves on the Board of the American Headache Society which runs an influential journal. She serves on the Executive Council of the International Headache Society, and the Board of the Headache Cooperative of New England. She has helped write all manner of guidelines for treatment of migraines, and has chaired the “Choosing Wisely Committee” of neurologists.

Knowing this might have affected how we worded things.

Dr Loder has been with BMJ since 2006 and declares that she has no links to pharmaceutical companies since then. She declares she has previously been a speaker, received grant support, or been a consultant for: OrthoMcNeil, Endo, AstraZeneca, GlaxoSmithKline, Pfizer, and Allergan – a list of companies that overlaps with one that formed part of my conflict of interest statement as of that point in time.

Before moving to BMJ, in a 2003 paper she co-wrote about how we don’t need to be so afraid of using triptans, it says both authors have gotten research grants and done speaking on behalf of various migraine medicine manufacturers including GSK.

Imitrex – sumatriptan – is GlaxoSmithKline’s triptan. With Imitrex off patent, GSK turned to Treximet, a combination of sumatriptan and an NSAID that was approved for migraine in children over 12 in May this year.

Sumatriptan comes with a suicide problem. Robert Gibbons has been enlisted to pooh-pooh this.

In a recent Washington Post article Dr Loder answers readers’ questions about migraine. She says:

“you haven’t failed sumatriptan till you’ve failed to respond to a full dose of injectable sumatriptan given early in the migraine. That’s my mantra!”

She squeezes in that “there’s evidence it’s more effective when taken with an NSAID” – just before Treximet came out.

While Dr Loder’s name is not listed in OpenPayments or Dollars for Docs, a regular partner on much of her work is Stephen Silberstein. They are on all the same Influential Committees. Dr. Silberstein comes close to being the KOL’s KOL in the headache field. Here is a recent Disclosure Statement.

Meanwhile, her hospital, Brigham & Women’s, got $15-20 million from GSK in 2014.

50 shades of gray

Meanwhile in addition to helping articulate BMJ’s position on conflicts of interest, Dr Loder was involved in a Lown Institute Twitter Chat in July on conflicts of interest, where she makes clear we need to find out more about non-financial COI’s.

Her BMJ COI statement refers to her husband’s position in a law firm but she states that he does not have a healthcare involvement.

His law firm is the Boston based Ropes and Gray.

John M Loder is not in the Healthcare division of Ropes and Gray. He is in Hedge Funds, Investment Advisors, SEC compliance. Attorneys from Mergers & Acquisitions, Government Enforcement, and White Collar Defense divisions likely work with GSK every day of the week.

Central to the work Ropes and Gray do is a First Amendment commitment that horrifies those who are usually most vocal about conflict of interest. This company is one of the main supporters of off-label marketing, on the basis that to prohibit it would be an interference with free speech. They have achieved some success in these areas. Their position makes a certain amount of sense to me.

The irony was that BMJ were using a close to monopoly position to abuse our First Amendment rights in respect of Restoring Study 329. If we had the money, the perfect people to have hired to fight our case against BMJ would have been Ropes and Gray.

But more to the point Ropes and Gray are currently GSK’s lead attorneys on company fraud matters in China. Their ties to GSK are close enough that journalists describe them as “GSK’s law firm in Hong Kong”.

And for several years they were also central to defending GSK in the Department of Justice and related investigations that led to a $3 billion fine for GSK in 2012. A key element to the charge against GSK lay in their handling of Study 329.

Colleen Conry and Brien O’Connor, co-leaders of the Ropes & Gray government enforcement practice, were successful in winning an acquittal for former GlaxoSmithKline associate general counsel Lauren Stevens in the U.S. District Court for the District of Maryland back in 2011.

As they say on their website:

Companies and individuals worldwide turn to Ropes & Gray’s leading government enforcement practice to guide them through all manner of investigations, enforcement actions, and compliance issues. Our attorneys understand business and regulation and help our clients pre-empt, resolve or mitigate the impact of government investigations and enforcement actions.

A radical suggestion

Around the time of Fiona Godlee’s July 6 email, BMJ suggested that the entire data from the study be re-analyzed by a third party who had no prior contact with the issues. This would likely have taken a further year at the very least.

This was plain bizarre.

Intriguingly, Elizabeth Loder and I had both attended a Selling Sickness conference in Washington in February 2013, and in her account of the meeting in BMJ a few weeks later she floated a very similar proposal.

How medical journals can help stop disease mongering

“It would be hard to collect a more fascinating bunch of topics or people in a hotel conference room. The 2013 Selling Sickness conference recently held in Washington, DC was among the most thought provoking and just plain interesting conferences I’ve been to in a long while, and I go to a lot of conferences. This third conference in a string of international conferences was characterised by its organizers as part of “a global reform movement” against the “marketization of health, the corporatization of healthcare, and the hijacking of patient and consumer language to disguise market interests.” Opening speaker Shannon Brownlee identified “the different heads of the hydra” as “disease-mongering, conflict of interest, and overdiagnosis.” In her view, the challenge of this meeting was to identify these aspects of selling sickness and “weave them together in a systemic movement that promotes systemic changes.”

“I participated in a roundtable discussion charged with examining the news and communication perspective on disease-mongering, and was joined on the journal side by Jocalyn Clark of PLoS MedicineModerator Gary Schwitzer of HealthNews Review and panelist Nancy Shute [] considered the problem from the vantage of medical journalists. When it came time to examine how medical journals might address their role in selling sickness and disease mongering, I suggested two possible strategies.

“First, why not quarantine apparently ground breaking studies about new treatments or interventions in a special journal until the findings are replicated and long term consequences explored? Print copies of the journal would arrive in plain brown wrappers which undone would show the journal’s cover logo of a skull and crossbones. During quarantine, any news stories or summaries of research from this journal would travel with a sternly worded disclaimer, along the lines of those that accompany investment company advertisements. Something like the following would do nicely:

“Warning! Taking any action on the basis of this research could result in injury or death. The results described in this study have not been replicated and the long term effects of this treatment are unknown. Past performance is no guarantee of future results. When subjected to further investigation, most published research findings turn out to be false.

“To fill the void, medical journals deprived of these sensational research studies could instead devote themselves to the promotion and prioritization of the less glamorous medical research that really matters: replication studies, comparative effectiveness trials, and long term pharmacosurveillance and safety studies.

“My second suggestion was that several parts of a typical research paper are too important to be written by the researchers or anyone else with a vested interest in the outcome of the research. These include the portions where “spin” is mostly likely to enter into the paper, namely the title, abstract, results, and conclusion sections, and any summary or “what this study adds” statements that authors are now sometimes asked to supply. These portions of research papers should instead be written by disinterested parties with subject matter expertise.

“I have no illusion that these things will come to pass but I can dream, can’t I?

“During the question and answer session fellow panelist Nancy Shute turned to me and said “I’m impressed by your radicalism.” I’ll take that as a compliment!”


One can wonder if Dr Loder ever suggested anything like this for any of Dr Silberstein’s articles, or GSK linked articles involving treatments for migraine.

While a suggestion like this didn’t appear formally in the correspondence between RIAT and BMJ, it was raised in behind the scenes negotiations.

Access to the data

On July 8, RIAT sent an email to Fiona Godlee outlining some of the points made above. There was no response.

Have conflicting interests had an effect in this case? There are data that can be consulted.

When finally published Restoring Study 329 came with a number of tabs, one of which is labeled Reviews. As of this year it is BMJ’s policy to publish the reviews of articles along with the article.

One of the first Rapid Responses to the article was from Elizabeth Wager, a medical writer who has previously had close links with BMJ and who had reviewed previous articles about Study 329 submitted to BMJ that did not get published. She asked where the Reviews were. There was no response from BMJ. I submitted a response to her letter indicating that the reviews were available in their entirety on BMJ did not publish my response.

RIAT re-sent the July 8 letter to Fiona Godlee on September 30. We have had no response.

My impression is that most people reading the entire correspondence on between RIAT and BMJ (mostly Elizabeth Loder) will agree that it shows an extraordinary level of difficulty. At several points the RIAT team came close to reaching the end of the line. The correspondence makes clear that some of us seriously entertained the view that BMJ did not want to publish. By July 6 we were exploring publication options with other journals.

In my opinion very few people will think BMJ were not biased by something. You don’t have to trust me – you can access the data and make your own mind up. Against this background, Elizabeth Loder’s competing interests will raise eyebrows for some.

Not tonight. I have a headache

Did John Loder’s involvement with Ropes and Gray cause our problems? In my opinion, no.

Did Elizabeth Loder’s prior links to GSK cause our problems? In my opinion, no.

Did Elizabeth Loder’s headache expertise cause our problems? In my opinion, no. She was finicky but could have been a lot more finicky than she was.

The strongest evidence I have as to why there was such an unconscionable delay is that BMJ have been and still are scared close to shitless about publishing anything that might make a pharmaceutical company uncomfortable.

They have no problem publishing a Beasley et al paper that shows an increased risk of suicidal behavior on Prozac but claims there is no risk and that the data exonerates Prozac. They have no problem publishing Lu et al in 2014 that claims to show warnings on antidepressants cause suicide – a shoddy piece of work if ever there was one.

Ironically providing access to the data seems to have increased BMJ’s difficulties. A key message from Restoring Study 329 is that when data is made available all authorship (interpretation) becomes provisional. Despite apparent support for access to data, The BMJ want science to be authoritative – Biblical – rather than provisional. If an article offers the indisputable Word of God, they can’t get sued. If there is scope to read the matter in another way as David Linden’s Response to Restoring Study 329 indicates there is and always will be, the BMJ and their lawyers have a problem.

In my experience journals run by ex-pharmaceutical company employees have been much more courageous than BMJ on matters like this. However, while they appear to have lied at least twice in the process, BMJ are a long way from being the worst in terms of courage and integrity – they do, though, take some beating in terms of prissiness.

What was galling about the exercise was that BMJ (and everyone else who colludes to put BMJ and other journals in this position) see fit to turn their lack of guts around and blame our conflicting interests.

This is to be expected from people who have themselves been abused or are living in an abusive situation. Medication Time.

Study 329: Conflicts of Interest

BMJ September 2015

Email from Fiona Godlee (Editor of The BMJ) to RIAT July 6th 2015
Re: Study 329

Dear Dr Jureidini,

Many thanks for your letter. I quite understand you concerns. You are right to say that there are few or no precedents against which to compare this article. We ourselves are feeling our way, both with the RIAT process since this is the first full RIAT research paper we will publish, and with the specific challenges posed by this particular study and you as the paper’s authors. I want to stress that we are proceeding in good faith with the clear aim of publishing the article as soon as possible provided we can do so safely. …

All best wishes

Fiona Godlee FRCP
Editor-in-Chief The BMJ

The path to publication

BMJ state that it takes on average 8 weeks from submission to publication of an article. The review process for Restoring Study 329 took a year, with a three month review process involving six reviewers to begin with, and then a further four reviews in a four month process, leading to a provisional acceptance in March that was withdrawn.

Ultimately there were seven versions of the article. All versions and all reviews and responses are available on Restoring Study 329 (See – timeline to publication).

At the time of this email from Fiona Godlee it was far from clear that the BMJ were committed to publishing the paper.

Conflicting interests

The Lancet and BMJ were among the earliest medical journals. They began in Britain in the early nineteenth century with a public health brief.

Public Health began in an effort to contain contagious diseases. No one knew what lay behind epidemics of plague, smallpox, cholera and typhus. The problem might lie in abattoirs that sat in the middle of towns in those days or in the unhealthy constitutions of those who became ill. A lot of powerful interests were at stake.

In the absence of effective treatments, the public health movement stimulated a revolution by focusing on the environment and changes to it rather than on the individual. Removing the handle on the Broad Street pump offered a wonderful symbol of its mission to prevent disease rather than intervening to treat when it it was often too late.

While there might be nervous abattoir owners on one side, in the late nineteenth and early twentieth century public health science and politics and business were largely on the same side. The new science created opportunities for new research based businesses. Establishing that bacteria caused infections opened up the prospect of vaccines and antibiotics and helped the growth of the insurance industry. The wealth that came with these new industries lifted many out of poverty and in this way helped health just as much as they hygiene helped by preventing diseases.

The conflicts of interest at that point lay, as they had done since Galileo, between science and religious or political beliefs – between the evidence that vaccination helped and those who for religious or political reasons were opposed to vaccination or to people being forced to take vaccinations.

Twentieth century conflicts

With the development of bacteriology and the demonstration that bugs caused infections and the hunt for a Magic Bullet, the focus drifted back from preventing disorders to treating individuals. This environment is fine if you take your pills – antibiotics, antidepressants, statins whatever.

This development was linked to another. Because the development of modern science was so interlinked with the development of modern industry, industry were among the earliest to embrace “research”. Long before the Cochrane Collaboration began systematic reviews, the lead industry were systematically collecting all articles in the lay or scientific press about the effects of lead, in order to monitor the threat to their business and to be able to assess what lead blood levels might “wash” with the public and politicians. The industry also supported scientists in research projects to minimize problems. But minimization of problems did not mean doing studies to determine the lowest blood lead level at which biological effects became apparent. It meant studies that would minimize problems to industry.

Lead is not on the radar these days for people in the way it once was or in the way that it and other heavy metal poisonings probably still should be. The epidemiology of schizophrenia for instance shows an initial emergence of this disease in the nineteenth century and a more recent decline and this rise and fall parallels lead usage and lead levels. Dementia and a range of other diseases like cancer likely stem from environmental sources that have a basis in industrial pollution.

In part because the tobacco industry’s famous aphorism Doubt is our Product crystallized a modus operandi for all modern corporations, there is some awareness today that an increasing amount of “scientific” research is done to muddy the waters on questions of environmental induced diseases. This is a use of data against Science, a capturing of the Appearances of Science.

Dwight Eisenhower in a famous address in 1961 just before he left presidential office caught some features of this new world: “In holding scientific research & discovery in respect as we should, we must be alert to the equal and opposite danger that public policy could become the captive of a scientific and technological elite.”

Female genital mutilation

The situation we now have is something like as follows. You want to submit an article to our Journal on the benefits of Female Genital Mutilation? You must declare your “conflict” and we might decide not to publish no matter how good your data looks because this looks like data marshaled in support of a belief system rather than science proper.

This example is not all that far-fetched. Many books and articles routinely carry claims about the supposed results of studies demonstrating the health benefits of circumcision (MGM), about which all men have a conflict.

Whatever anyone’s beliefs about FGM and MGM, at the end of the day the power of science lies in an acceptance that we resolve issues with data. The idea that pharmaceutical companies can hang onto data by Force Majeure is the antithesis of science.

The new biology, pharmacology and trials-ology risks undoing Galileo and putting the individual back at the center of the universe. We can’t randomize environments but we can randomize individuals and if some drug produces a difference in a situation the simplest and most profitable conclusion will usually be that we have corrected a defect in the individual.

The new biology, pharmacology and trials-ology risks undoing the Bible even. The script is being rewritten to tell the story of a vulnerable Goliath, who while laboring away at making the medicines critical to all of us, and providing jobs, is at increasing risk from an unscrupulous and unethical David.

While there is the occasional headline such as a former editor of the BMJ resigning an honorary post from Nottingham University because it took money from the tobacco industry, the new rules of the game are as outlined by Fiona Godlee – anyone who has attempted on the basis of data to challenge a pharmaceutical company about the adverse effects their drugs cause must ipso facto be deemed to have a Conflict of Interest on a par with someone presenting data apparently showing health benefits from FGM.

The BMJ and NEJM reserve the right not to publish material like this, on the basis that while we are committed to resolving issues on the basis of data, anyone who has accessed data on this issue or in this fashion clearly has some strange belief system and must be deemed to be suspect – unlike companies like Pfizer whose actions follow the data in a disinterested fashion.

Medication time

What’s really going on is that journals are scared of industry. This is partly because the leading figures in medicine have become minions concerned more for the health of industry than the health of people and there is no-one who is likely to back up a journal willing to take a tough stand.

There is another reason. While data might really be new data – previously unavailable – the BMJ are scared that if they feature it they will be put out of business. Journals are a business that feeds on science rather than part of science.

These journals are caught in a 1960s style double-bind. They are dealing with a Nurse Ratchet, who writes friendly letters on first name terms to the journal’s Editor – Dear Fiona – but who has made it clear she will sue and put them out of business if they don’t play ball at medication time.

Close to the last thing a journal is willing to do today is feature any article about the biggest public health issue of our day – the biggest source of environmental toxicity – the morbidity and mortality caused by treatment.

In the face of a complex situation like this, the temptation as Daniel Kahneman told us 40 years ago is to resort to the Fundamental Attributional Error which is to look for who to blame. But rather than blame the powers that might house-arrest them (as happened to Galileo), the BMJ, along with society’s new religious authorities, locate the problem in Galileo – the person attempting to produce the data.

In the case of Study 329 there also appeared to be a desperate hunt for some box to tick that would make BMJ fireproof, some proof that the scurvy knaves behind the Restoration of Study 329 were adhering to some protocol that would make the results objective and definitive – exactly the opposite message to the message of the final article which is that the results are provisional and that objectivity as regards what these drugs do for good and for bad arises from having the data out there so they can be contested.

Science involves getting data into the public domain where it can challenge beliefs. But decades ago BMJ and other journals gave up on any attempt to access the clinical trial data behind the claims being made.

Now BMJ are signed up to an AllTrials coalition that includes GSK and is against Data Access.

Study 329 in Japan

jp paxil

Editorial Note: By 2002 GlaxoSmithKline had done 3 studies in children who were depressed and described all three to FDA as negative. As an old post on Bob Fiddaman’s blog reproduced here outlines, several years later they undertook another study in children in Japan.

Regular readers of this blog will know how I broke the news back in 2009 regarding GlaxoSmithKline’s attempts to push Paxil on kids in Japan. [See Here]

I was so outraged at this that I wrote to the Japanese Embassy and the Japanese Ministry of Health, more or less to give them a detailed view of how GSK had previously claimed Paxil was safe for kids to take…when in actual fact they knew that it wasn’t.

I never heard back from either one of them.

I also contacted GlaxoSmithKline in 2010, you can see the email I sent to them here.

In 2008, one year before I broke the news, Glaxo were recruiting kids for a clinical study. I say Glaxo, they, in actual fact were sponsoring the study.

The study was designed to compare the efficacy of oral paroxetine 10 to 40 mg/day (initial dose:10 mg/day) versus placebo administered once daily.

Oral paroxetine is a sickly orange syrup, I’ve been on it myself. It was the only safe way to taper from this highly addictive antidepressant.

And just who were being used as the guinea pigs in Japan?

One look at the inclusion study criteria would have showed you.

Ages Eligible for Study: 7 Years to 17 Years.

Yup, that’s right folks. Despite being dragged through numerous courts in the US where evidence was shown that Glaxo manipulated previous clinical trials in children, here they were again back in 2008 recruiting more kids.

A marketing campaign went out in the form of a poster… which I just happened to obtain from a source at Medwatcher Japan. Medwatcher were also furious at this particular clinical trial involving kids and Paxil.

Take a good look at the imagery used in the recruitment poster.

The results

Well, folks… **drum roll** – the trial has been terminated.

According to GSK’s clinical trial database the study was terminated in 2011. They give no reasons as to why this study was terminated.

What we do know is that 56 kids were enrolled. 29 were in the Paxil group whilst the remaining 27 were in the placebo group.

The study results claimed that there were 3 reports of suicidal ideation in the placebo group but none in the Paxil group.

The subjects enrolled had to have a diagnosis of a depressive disorder before being allowed into the study.

So, it would appear that 3 of the kids taking placebo had suicidal ideation. Not one report in the Paxil group. Glaxo must have loved this.

Unfortunately for the Glaxo sponsored trial, Paxil didn’t really show much efficacy.

Open the Outcome Measures on the Clinical Trials website and it tells us how Paxil failed.

Paxil didn’t reduce the depression scores of the children sufficiently to be considered effective and the primary purpose of this study was efficacy. In this study all participants had to have a depression score of 45 or greater to be included. A 50% reduction on the CDRS-R is required to consider children have responded to treatment.

So, not only did Paxil not reach the standard for efficacy but in comparison to, let’s say, fluoxetine, it would be seen to be less effective.

In the Fluoxetine studies [1] [2] the average decrease was 28.9%. In this Japanese study, the decrease was only 16.9% therefore the kids would not be considered to have responded to Paxil treatment.

No wonder the study was terminated, right?

It would appear that GlaxoSmithKline didn’t want to expose the fact that Paxil is less effective than that of their competitor.

Another interesting finding from the Japanese study was the participants only had to have been free from any antidepressant for 1 week prior to the trial commencement.

Anyone who’s anyone will know that one week off an antidepressant is hardly a time to get the champagne corks popping and decorate rooms with bunting and balloons. Any number of these participants could have been suffering withdrawal even before they were entered into the Japanese trial. Any of these patients suffering withdrawal, which remember can mimic depression, would have had immediate relief if they were selected for the Paxil arm of the trial. As the phases of the trial progressed they would have, obviously, reaped the benefits of Paxil but not for their apparent depression, their benefits from Paxil would have merely meant they would not be going through withdrawal anymore.

Take the three patients from the placebo arm of the study who, according to the results, suffered suicidal ideation, and we may just find that these three were also taking antidepressants a week or so before they entered the Japanese study.

Could their suicidal ideation have been caused by the withdrawal effects of the medication they were taking prior to the Japanese study?

Glaxo pretty much shot themselves in the foot with this study, a study that should never have taken place given the findings of the Paxil 329 study.

So, once the Japanese trial was over did the sponsors, GlaxoSmithKline, do any follow-up to see if these kids were okay? The doses used in the study were between 10mg and 40mg, the latter being enough to put a horse into a coma.

The withdrawal phase of the Japanese study lasted three weeks. Two weeks later the participants were contacted to see how they were.

Can you imagine a 7 year old child on 40mg of Paxil a day just having three weeks to taper? Even if the 7 year old was on a lower dose it’s still mind-boggling how one adult human could give someone so young a pill known to increase suicidal thoughts, known to increase completion of suicide.

What on earth were GlaxoSmithKline thinking by using kids in a study for Paxil?

The Japanese public, particularly the children and adolescents, just don’t know what a lucky escape they’ve had from this truly awful abomination of an antidepressant.

Bob Fiddaman

[1] Psychometric Properties of the Children’s Depression Rating Scale–Revised in Adolescents – J Child Adolesc Psychopharmacol. 2010 December; 20(6): 513–516.
[2] Early Prediction of Acute Antidepressant Treatment Response and Remission in Pediatric Major Depressive Disorder – J Am Acad Child Adolesc Psychiatry. 2009 January; 48(1): 71.