Peer Review Anonymity on Trial

Keith Robison robison at lipid.harvard.edu
Mon Jan 15 20:01:41 EST 1996


U27111 at uicvm.uic.edu wrote:


: On 12 Jan 1996 18:34:53 GMT gregoryh at bcm.tmc.edu (Gregory R.
: Harriman) wrote:

: But...one would think that after 15 years, we would definitively
: know whether AZT actually works or not???

: Don't you think?

: Instead, we have dueling studies.

A 15-year retrospective on AZT use against AIDS would be a very
interesting study to do -- especially since it would require
_fabricating_ several years of data.

: And AZT is not alone in this... there are many drugs in which
: contradicting studies come out.

: And any half-way intelligent person should be able to recognize
: that such dueling studies aren't a factor of a poor understanding
: of complex issues.... but more due to intense competition at it's
: *highest* level.

Or perhaps a reasonable person would understand that differing
studies _may_ actually represent differences in the studies --
differences in protocols, study groups, etc.  Without actually
looking at the studies you can't tell either way.

An interesting sort of example appeared in _Cell_ last year.
Two different groups reported (in back-to-back papers) the finding
of the gene for a particular inherited disease.  Only problem is,
they reported two different genes (right next to each other).  
It's a legitimate scientific difference of opinion, and from
the data presented it is impossible to distinguish.  Sometimes
a cigar is just a cigar.

: >Nonetheless, the presence of a single lawsuit hardly
: >constitutes evidence of "a deep moral crisis of the present
: >research enterprise".

: Well... here we have the two schools of thought nicely displayed.

: On one hand... we have the 'tip of the iceberg' mentality.

: And on the other hand we have the 'rare event' mentality.

: And one thing I have learned when studying such debates... the true
: answer usually lies somewhere between the too extremes.

Agreed.

: >In any social institution or activity there will always be
: >people who try to take unfair advantage or cheat.  While this is
: >unavoidable, we should certainly do everything possible to
: >discourage and punish those that step over the line.

: How?

: Self-policing has been proven to be a myth.

: The ORI is generally ineffective, unless of course the person cited
: for misconduct 'agrees' to abide by their ruling.

Yup.  And don't expect things to get any better, no matter
what system you come up with.  Because misconduct cases involve
people's careers, they will go the legal route -- meaning legal
standards -- which means intent to deceive must be proven
for it to be fraud.  And proving that intent is a difficult
process.  

: Gallo is a perfect example... $24 million and growing... and
: building.

: We are actually awarding such behavior.

No "We" here.  If memory serves me, it's a bunch of Maryland 
politicians who gave Gallo et al this prize.  This is another
harsh reality anyone attempting to reshape the current reality
must face -- that there are multiple funding sources, and that
nobody controls them all.  You can't blame Gallo's new home
on APR -- it falls clearly in the domain of political earmarking.


: With all due respect sir, the whole system
: (funding/grants/APR/patents) *is* a breeding ground for
: corruption... and the honor system just isn't good enough anymore.

Again, what system would _you_ replace the current one with
that would actually improve matters?  How would _your_ system
deal with scientific misconduct artists, within the bounds of
the U.S. legal system?  What undesired side-effects have you
anticipated, and are you willing to accept them?

Keith Robison
Harvard University
Department of Cellular and Developmental Biology
Department of Genetics / HHMI

robison at mito.harvard.edu 








More information about the Bioforum mailing list