Download Lectures on Biostatistics (1971). Corrected and searchable version of Google books edition
Download review of Lectures on Biostatistics (THES, 1973).
This may seem very odd to people outside the biosciences area, but in recent years a business has grown up that will write a paper for you, on the basis of data supplied by a pharmaceutical company. The person who actually did the writing will usually not appear as an author at all This practice is known as ghost writing.
It is obvious from the last two posts ( here and here ) that ghost writing of papers has became a major menace to science in the clinical world. The authors may have not even seen all of the data, and some “authors” may even not have seen the paper at all. Awareness of the size of the problem has increased, especially since the paper by Healy & Catell (2003) [ download the pdf ] (see also here ).
They say
“a process of changing scientific authorship that could conceivably culminate in a situation in which the dominant figures in therapeutics actually have comparatively little first-hand research experience and few raw data that they can share with others.”
See also “Declaration of Medical Writing Assistance in International Peer-Reviewed Publications”, Wooley et al ., 2006, JAMA, 296, 932 – 3 [get pdf ], and a Professional Ethics Report published by the American Association for the Advancement of Science.
The conversation
I recently had a conversation with a “medical writer” (don’t bother asking me who, I’m not saying). I’m very grateful that he spoke to me, because this is the sort of information that is usually concealed carefully from public view. He was clearly a person with integrity, but nonetheless what he had to say was quite disturbing. Here are some of the things that I learned.
- So called ‘Medical education agencies’ are also marketing agencies. They give expert advice to industry on how to select information from large clinical studies and how to cut and slice it to generate numerous papers. They advise on ‘publication planning’ and on which journals are best to reinforce particular messages that the companies want to convey.
- Medical writers who are acting as ‘ghost writers’ often interact with clients in the industry and don’t necessarily interact with the proposed academic authors. Opportunities to talk to the authors directly can be very limited and discussions are often mediated by a third party.
- Medical writers often do not see evidence that all authors have approved the submitted manuscript. It is perceived that only the primary author has actually approved the manuscript.
- Even if medical writers are aware of their responsibilities concerning ‘good publication practice’ and publication ethics it can be difficult to persuade clients to adopt all of these practices.
- Often medical writers do not see all of the raw data for a study or publication, they can get pre-selected data or just tables and figures for inclusion in the paper. Medical writers who are relatively inexperienced do not realise that this is not ‘best practice and will prepare a paper without requesting further background data.
- Medical writers can be pressurised by clients to include ambiguous data or marketing terms in manuscripts. When medical writers have problems with clients, support from their employers can be limited as clients are paying the bills. In this context, the ‘client’ is the company who produce the drug, not the people who eventually appear as authors.
- Ghost written papers have a number of authors. Often the most important is the primary author; this can be a selected ‘key opinion leader’ whose reputation is relied on by the company for promotion of the product.
In defence he did say that 99% of those working in the pharmaceutical industry are a pretty decent bunch. They understand that the consequences of manipulating data or employing unethical publication practices, if discovered, can be disastrous for the company and their product.
He also said that unethical publication practices occur regularly in purely academic environments. And in that he is, unfortunately, absolutely right. After all, the business of “medical writing” would not exist at all if senior academics were not willing to attach their names to papers that they have not written, and may well not even have seen all of the data.
Still worse, having been caught out doing that, they may be defended by their vice chancellors rather than condemned. Indeed anyone who dares to mention that it is going on is taking the risk of being fired. There is a shocking example here.
15 Responses to A conversation with a ghost writer
Leave a Reply
This site uses Akismet to reduce spam. Learn how your comment data is processed.
I wish I had had a ghostwriter for my thesis…..
Anyway, presumably this problem could be overcome if journals insisted that ghost written papers were declared as such and that signatures were required from all the authors stating that they had read it prior to submission.
It is encouraging to see this topic being aired by people with reliable, detailed knowledge, as opposed to those with evil-big-pharma axes to grind. Plos Medicine ran a relevant article date 25 Sept 07: http://medicine.plosjournals.org/perlserv/?request=get-document&doi=10.1371/journal.pmed.0040286 , including an analysis of how MECCs operate. This is an extract:
“In a primer on publication planning, the director of one MECC defines the activity as: “gaining product adoption and usage through the systematic, planned dissemination of key messages and data to appropriate target audiences at the optimum time using the most effective communication channels” [32]. These channels are such things as: “publications, journal reviews, symposia, workshops, advisory boards, abstracts, educational materials/PR.” Influencing scientific opinion in the service of marketing is the clearly stated goal here. The author of this article therefore makes scientific and commercial goals equal stakeholders in communication: in a chart he juxtaposes “Where shall we publish this study?” with “Who are our customers?” and “What can we claim from the results?” with “What are our customer needs?”…”
People need to be able to blow the whistle on unethical practices, pressure from the pharmaceutical industry or whatever, without fear of adverse consequences for their livelihoods. Maybe that’s a tall order in our money-driven world but failure to do so means scientific medicine is viewed by some with increasing suspicion. Something which doesn’t go unnoticed by promoters of quackery and pseudoscience.
Poor innocent soul that I am, I was genuinely shocked to learn in these blog pages that ghost writers actually exist in the biomedical literature. I was under the naive assumption that first and/or last authors actually wrote the papers they put their names to.
Is this sort of thing widespread, or am I right in thinking that clinicians are the principal “beneficiaries” of ghost writer’s services?
It just seems to fit with the kind of attitude that I have found in some clinical colleagues. They have to publish to advance their careers, although their real interests are in the clinic, and getting someone else to do the experiments, analyse the data and write the work up is just the most efficient way to get the job done.
It’s only quite recently that I became aware of it too. I have never encountered it in my own world. As far as I know, it is almost entirely restricted to the clinical world, and mostly to papers that papers that involve drug companies. It is sad that people will sink to this sort of thing. It is even sadder to see it defended by vice-chancellors.
Well, I’m not going to argue that the pharma companies are without sin, but I have done ghost writing on and off for 20 years and have never been pressured to distort messages. I have always had access to whatever data I wanted. On one occasion I was doubtful about some tables and was sent all the raw data (great – it was faxed!). Over the years I think I have tended to write the rather more humdrum papers, because it can be the devil’s own job to get the investigators around to doing it. I also have to re-write manuscripts because some medics write so badly. I usually have direct contact with investigators and they are actively involved in producing the manuscript.
Ghost writing is done as a pragmatic way to get studies written up. Like lots of activities it’s open to abuse, but perhaps I’m lucky not to have encountered any in my career.
I think the real concern is about ‘ghost management’ of research findings rather than simply ghost writing. I have, on occasion, rewritten technical articles, due to the impenetrable prose style of the original. As the Sergio Sismondo says in the Plos Medicine article referred to earlier: “…Ghost writing and honorary authorship are not in and of themselves scientific problems, though they become so when they shape science to meet particular interests…”
I wonder, do the No Free Lunch campaigns (USA and UK) have a specific position on this? It’s not a tangible like money or gifts but probably equally, if not more, damaging to public trust in medicine.
OK there is nothing wrong with improving the standard of English of a paper that has been written by the people who did the work (and being acknowledged for doing so), But that isn’t really what we are talking about. This comment is not aimed primarily at ghost writers (not the many honest ones anyway) but at the academics who hire out their names.
Sergio Sismondo says in Plos Medicine
To me, that remark shows starkly just how far standards have sunk in the clinical world. Can he really be saying that it is not a scientific problem if the authors have not seen the data, and if authors appear who have contributed nothing and may not even have read the paper? This is not science at all, but public relations (i.e. lying).
Even if there is no distortion of data, it is simply dishonest to put your name on a paper that you can’t take responsibility for. Any one who does it should be in deep trouble. Nature has the right idea. in its proposed rules.
I say “clinical world” deliberately. As far as I can see the biggest problems are restricted to that part of the clinical world that is concerned with assessing a marketable product. In basic biomedical science, ghost-writing is (as far as I know) quite unknown.
The same, sadly, cannot be said of honorary authorships. That is a growing problem and it is a form of dishonesty that is actively encouraged by government policy. Once everyone is assessed by the number of citations they get, it will get much much worse.
In fairness to Sismondo, he does conclude his article thus,
I agree totally with that part of his article. Sadly, it doesn’t happen. In fact universities (i.e. vice chancellors) connive in the corruption if enough money is at stake, as in the case of Blumsohn vs Sheffield.
Where there’s brass, there’s muck, it appears, and few with the guts to challenge. Very sad, ethics seem to be contracted out as well as the research. Sobering how many of these MECCs and CROs are owned by big players in the advertising and publishing industry.
Perhaps slightly off the specific topic of ghostwriting, this report of research newly published in BMJ deals with industry influence on conclusions of meta-analyses: http://www.newswise.com/articles/view/535380/?sc=rsmn
links to free full text included
I think a key issue here is what constitutes authorship from the perspective of a journal editor.
The Uniform Requirements recommended by the International Committee of Medical Journal Editors (http://www.icmje.org) state that authors should meet all of the following conditions: 1) substantial contributions to conception and design, or acquisition of data, or analysis and interpretation of data; 2) drafting the article or revising it critically for important intellectual content; and 3) final approval of the version to be published.
So, by these criteria, none of the authors on a paper actually need to have drafted the manuscript in order to qualify for authorship. On the contrary, a ghost writer will never qualify for authorship, as they will never satisfy the final point.
I would say that the vast majority of authors of reports on clinical studies are original investigators from the industry-sponsored studies in question, and would in no way be classed as “honorary authors”. Review articles, on the other hand, are a bit more of a grey area.
I would also agree that medical writers usually have direct contact with the authors. In fact, it is often the case that the pharmaceutical company has no direct contact with them and the writer acts as the intermediary.
On another point raised, the use of medical writers is not just restricted to supporting clinical authors. There are plenty of preclinical studies carried out during a drug’s development (some of which are carried out in academia with drug company support)where third party writers get involved.
“I would also agree that medical writers usually have direct contact with the authors”
My correspondent said precisely the opposite: that there is normally little or no direct contact.
It seems to me to be both pathetic and very dangerous that people don’t write the papers for whuch they are supposed to be responsible.
I’d certainly be interested to hear more about the last sentence. The fact that we don’t know more is presumably because the (non)-authors are (rightly) ashamed that they couldn’t be bothered to write up their work themselves and try to conceal it.
Not relevant to the discussion, apologies, but this report of ‘grassroots investor’ harassment of oncologists who did not support a potentially lucrative prostate cancer treatment casts a chilling light on money driven medicine: http://www.healthbeatblog.org/2007/08/wall-street-can.html#more
[…] (most strikingly in the UK in the notorious case of the University of Sheffield). We have seen ghost-writing and spin tolerated, and even encouraged, not to mention sham consultations and attempts to impose […]
http://www.pharmalot.com/2008/04/the-merck-theme-song-ghostwriters-in-the-sky/
from The Man in Black
[…] Ghostwriting of papers is serious corruption. A company writes the paper and senior academics appear as the authors, though they may never have seen the original data. Even in cases where academics have admitted to lying about whether they have seen the data, they go unpunished by their universities. See for example, the case of Professor Eastell. […]