Thank you, Steve. I've been watching the unfolding discussion with
interest and deciding when to write JUST such a response. There is a
difference between scepticism and cynicism. The latter is a lot easier
and usually unhelpful because it doesn't add to the discussion.
Some other observations on the discussion to date.
Too much bad research:
Yes, I believe there is. This is largely because of the nature of the
'machine' in tertiary education - many have commented on this already.
As a personal summary, I am not a professor. I will not become a
professor until I've published a lot more than I have. I will not
publish more until I have grant income to sustain a team to undertake a
meaningful research programme that will generate useful findings. I
won't generate grant income until I research/publish more. I won't
research/publish more until I teach less. I won't teach less until I
have grant income to buy out my teaching time - and even then there is
actually no guarantee based on past experience
As has been said by Rakesh, research is about learning. Not every
research project should be published, some provide internal learning
telling us how to get to the answer rather than the answer itself.
Whilst papers describing and critiquing methods are helpful my pet hate
is finding the perfect abstract and pulling the paper to find it is
describing how the study will be done not whether the method works or
what the results are. PLEASE can journals stop publishing this CV
packing bull.....
In direct contrast to this view of too many papers, by-and-large I
believe research involving patients or volunteers (possibly include
animal studies too here depending on your view of ethical approaches to
using animals in research) should be published. If it isn't designed to
generate an answer you shouldn't be doing it -and BTW your Ethics
Committee or IRB should have withheld permission until it was. Once
completed you have a duty to publish BUT you also have a duty to publish
what WAS found not what you WISHED you'd found. We can all be guilty of
wanting to 'sex up' the results - we need to resist it and behave like
scientists.
Publication:
There are too many journals publishing too much 'quick and dirty'
research. Compounding the problem there are so many journals that
peer-review is compromised; there simply are not enough good reviewers
to go around. I don't know whether I'm a good reviewer or not but I do
know it takes me around half a day, sometimes more, to conduct a
standard manuscript review. I try to be critical of the research, not
of the researchers. My experience as an author is very varied. My
heart sinks when I get a 20-point review back from a journal but at
least I know the reviewer has read my article, thought about it and has
raised some points for consideration. When I get one back with two
lines I feel initially pleased but there is always a nagging doubt
later.
On a note re: publication. When are we all, by which I mean
researchers, editors and reviewers, going to start publishing
manuscripts that show no difference or negative results? They are still
a rarity and it isn't always the drug companies suppressing them, as
seems the prevailing view amongst the delightfully naive.
Conflict of interest:
I am an academic. I've received public research grants. I have also
received contribution from "Big Pharma" and from device manufacturers
through no-cost supply for trials. Am I evil? I don't think so. I
don't feel in the slightest bit compromised because in the first two
situations the companies had no influence on the design of the study and
- in fairness to them - sought no influence on or even forewarning of
the results prior to publications. They got a copy of the paper on the
day the journal embargo expired.
I have likewise been paid an honorarium to speak at meetings sponsored
by Big Pharma. Am I evil? I don't think so. I was given some slides
to review which I politely declined to accept and wrote my own based on
my reading of the data. I said some things the company probably
approved of and some things they almost certainly didn't want to be
highlighted. They didn't complain they just didn't invite me back or
send me a Christmas card. I'm assuming they acknowledged that was the
risk they were taking by picking an independent speaker and found
someone else. Presumably that person provided a more palatable version.
I'm not annoyed, that is their game and the other person's
conscience....and, by-the-way, my view may have been wrong.
The answer:
Until a miracle occurs and the problem is solved by divine intervention
I will continue to behave in a way that I believe is appropriate. This
is, of course, influenced by morals and social norms so it is reasonable
to expect that most other moral researchers will be doing the same. I
will continue to teach my students to think about what they are doing
and why - both in research and, more importantly to me, in practice. To
me that is actually what being an academic is all about - thinking more.
Happy Guy Fawkes to you all.
Jeff
-----Original Message-----
From: Evidence based health (EBH)
[mailto:
[log in to unmask]] On Behalf Of Steve Simon,
P.Mean Consulting
Sent: Friday, 5 November 2010 11:35 a.m.
To:
[log in to unmask]
Subject: Too much skepticism is just as bad as too little skepticism
The list has been filled with comments like this one:
Since one's main interest is to generate income, the primary objective
of the research is to perpetuate itself, i.e. to ensure more grants.
Patient outcomes and/or better allocation of limited public resources
are irrelevant, let alone maximizing the health of populations.
I can't disagree more. In my book, I write about the value of things
like randomization, intention to treat analysis, blinding, etc. and then
I turn around and say that randomization is overrated, intention to
treat is overrated, blinding is overrated, etc. The reason for this
turnaround is that there is too much black and white mentality in the
EBM community--if a study is randomized, it is good, and if it is not,
the study is bad. Everything needs to be placed in context. A well
conducted observational study is far better than a shoddy randomized
study.
The same is true with conflicts of interest. Conflict of interest is
overrated by many in the EBM community. It has to factored in, but you
can't disqualify a study just because there is a conflict of interest.
That's just as bad as disqualifying a study because it was not blinded.
A well conducted study is persuasive even if the authors have a
financial conflict of interest. The key is transparency of the methods
and objectivity in the approach. A financial conflict is a fatal flaw in
a subjective overview of the research, but it is not a fatal flaw in a
systematic overview. The systematic overview, with an open protocol and
objective ways to extract and combine the information from multiple
studies, is certainly defensible, even with a financial conflict of
interest.
Furthermore, to claim that the ONLY goal of academic researchers is to
ensure more grants is just not true. I know these people. I work with
them on a daily basis. They are concerned about getting grants, but that
is NOT to the exclusion of caring about their patients. In my
experience, what motivates medical researchers most is that they see a
problem in how health care is delivered and they want to fix it. Getting
the grants is a means to an end, and not the end itself.
The other problem with all of this discussion is a lack of appreciation
for the varying ranges of severity associated with conflicts. Everyone
has pressures that can influence how they conduct their research, but
some influences are minor and easy to resist, and others are major and
far more difficult to resist. Accepting a free trip from a drug company
makes you more conflicted than accepting a free meal. And non-financial
conflicts, while they can't be ignored, are less of a concern than major
financial conflicts of interest.
I believe that a skeptical attitude is mostly helpful in EBM, but too
much skepticism can be very harmful.
--
Steve Simon, Standard Disclaimer
Sign up for The Monthly Mean, the newsletter that
dares to call itself "average" at
www.pmean.com/news 
