Print

Print


Dear Martin,
Good question. The pharmaceutical companies do 
invest *much* more in research than government 
and not-for-profit groups (in the UK this is 
about 5:1). However, often the research is me-to 
or marketing research that doesn't answer our 
clinical questions. So the research relevant to 
clinicians may be better balanced. I can't tell 
you how many of the 20,000 new randomised trials 
published each year are non-pharmaceutical 
interventions, but in a review of things we 
picked for the EBM journal we found about a 50:50 
split over 3 years. The split we found is similar 
to the split you can find in Figure 3 of a recent 
UK analysis of non-commercial trials:
Chalmers I, Rounding C, Lock K. Descriptive 
survey of non-commercial randomised controlled 
trials in the United Kingdom, 1980-2002. BMJ. 2003 Nov 1;327(7422):1017.
Cheers
Paul Glasziou

At 09/03/2006, you wrote:
>I have been an observer of this listserv for a 
>couple of months as I get acquainted with the 
>main papers dealing with EBM.  I am just 
>starting a PhD which involves general 
>practitioners use of online evidence resources, 
>so the ‘Best EBM methods papers’ is extremely useful for me  - thanks
>
>
>
>I don’t know whether this listserv is used for 
>much ‘discussion’, but I am interested to 
>followup on what Steve was highlighting and ask for comments…
>
>
>
>Question: if research studies are biased to 
>pharmaceutical interventions, and EBM relies on 
>exploring the available evidence, then EBM will 
>also be biased to pharmaceutical interventions.
>
>I know this is a bit simplistic, but is this 
>true?  Is there any solutions to counter this problem?
>
>
>
>Regards
>
>
>
>Martin
>
>
>
>
>
>
>
>Martin Halperin MBBS MPH
>
>Academic GP Registrar and Lecturer
>
>Department of General Practice
>
>Monash University
>
>867 Centre Rd
>
>EAST BENTLEIGH Vic 3165
>
>Phone:  (03) 8575 2289
>
>Fax:    (03) 8575 2233
>
>[log in to unmask]
>
>
>
>
>
>
>
>
>
>"Simon, Steve, PhD" <[log in to unmask]> wrote:
>
> > I looked through my files and found ten papers published in 2005 that
>
> > dealt directly or indirectly with EBM issues. I put these on my web
>
> > site. Look for the March 6, 2006 entry on my Research Methodology weblog
>
> > at:
>
> >
>
> > http://www.childrens-mercy.org/stats/weblog.asp
>
> >
>
> > Three papers merit special attention, because they call into question
>
> > the very research database that we rely on for EBM.
>
> >
>
> > 1. Medical journals are an extension of the marketing arm of
>
> > pharmaceutical companies. R. Smith. PLoS Med 2005: 2(5); e138
>
> > 2. Contradicted and initially stronger effects in highly cited clinical
>
> > research. J. P. Ioannidis. Jama 2005: 294(2); 218-28.
>
> > 3. Randomized controlled trials of aprotinin in cardiac surgery: could
>
> > clinical equipoise have stopped the bleeding? D. Fergusson, K. C. Glass,
>
> > B. Hutton, S. Shapiro. Clin Trials 2005: 2(3); 218-29; discussion
>
> > 229-32.
>
> >
>
> > The first article in the list, by Richard Smith, is one among a large
>
> > series of articles that identify how commercial interests have made the
>
> > entire research endeavor open to question. The second article, by John
>
> > Ioannidis also calls the entire research endeavor into question and
>
> > points out the many factors that cause incorrect conclusions to be drawn
>
> > in most published research findings. The third article by Dean Fergusson
>
> > and others, argues that we do too much duplicative research that just
>
> > confirms what is already well established. The authors cite a particular
>
> > example involving placebo controlled trials which raises serious ethical
>
> > concerns, because 64 trials were run, but clear and convincing evidence
>
> > emerged by the time the 12th study was published.
>
> >
>
> > There are several other articles published in 2005 that corroborate the
>
> > problems cited by these three articles, but I could not easily locate
>
> > them. Sorry!
>
> >
>
> > You could argue that these are not truly EBM articles, but I believe
>
> > that they raise important issues that anyone who uses EBM should be
>
> > aware of.
>
> >
>
> > Steve Simon, [log in to unmask], Standard Disclaimer.
>
> > Look for my book "Statistical Evidence in Medical Trials"
>
> > which has just been published. For more details, see
>
> > http://www.childrens-mercy.org/stats/evidence.asp
>
> >
>
> >

Paul Glasziou
Department of Primary Health Care &
Director, Centre for Evidence-Based Practice, Oxford
ph: 44-1865-227055