shade data. The wholesale shift of these responsibilities to universities an account, giving the accused a chance to reply. Other penalties
would be problematic. Academic statisticians have their own research may apply.” Does JAMA want to list offending individuals on the
programs; performing service activities or analyses of industry data journal’s web site, as FDA does in disbarring scientists from partici-
will be of lower value for promotion and tenure decisions. pating in IND investigations?
But the part that puzzles us most is why DeAngelis and The JAMA editorialists have been too eager to prescribe reme-
Fontanarosa believe money cannot corrupt academic statisticians, dies for the statistical profession—with little input or counsel from
but can corrupt those in for-profit companies. (We are not com- those affected. n
pletely clear how the editorialists would classify government work-
ers or those who work for private, not-for-profit corporations.)
When a sponsor pays an academic statistician to analyze a data set
for submission to the FDA or for publication, that statistician must
make a substantial time commitment (weeks or months of analy-
ses) and will generally charge substantial fees. Thus, the academic
statistician has a potential conflict of interest. Worse, separating
Bentley’s components of a statistician’s role into the set of activities
performed by the in-house statistician removes intellectual coher-
ence from a study.
If the editorialists’ real problem is distrust of the sponsor,
their solution does not prevent nefarious practice. A company
could select a different statistician for each prospective paper,
with the potential for inconsistent studies and analyses. Or, they
could send the same data set to several statisticians and pick the
most favorable analysis.
but Wait, There’s More
The editorial says, “All journals must require a statistical analy-
sis of clinical trial data conducted by a statistician who is not an
employee of a for-profit company.” Why? If this recommendation
becomes effective, it pays an academic handsomely for doing work
that could be done by a company statistician, a contract research
organization, a consultant, or an employee of a private research
company. The amount of time involved is no less (and, in many
cases, more) and the profit to the individual statistician is hidden.
Why have the editorialists singled out statisticians as the evil-
doers? Why not prohibit private physicians who do not have an
academic appointment from participating in clinical studies? We
prefer wide participation in medical research as long as the research-
ers have scientific integrity; these proposals put forward artificial
structural solutions; they do not address integrity. Referring to
the phases of statistical involvement in the studies, this proposal
unwisely excludes statisticians from the design and data manage-
ment operations. Does JAMA want to pay academic statisticians for
a detailed review of the submission with full access to the data? But
the fundamental issue is deeper. Having an academic statistician
analyze data does not prevent the sponsor from submitting only
Invitation to ASA’s Centenary Celebration
part of the data to the statistician or the journal. For example, if
(1939), ASA Record Book, Box 9, Folder 1,
a safety data set covers only three months after drug cessation (as
American Statistical Association Records,
specified in a protocol), late adverse events would be missed.
MS 349, Special Collection Department,
Iowa State University Library
Unworkable ‘Solution’
The URL for the archive is www.lib.iastate.edu/spcl/
The editorial then recommends penalties for authors who fail
to disclose conflicts, suggesting they be reported to the dean or
manuscripts/MS349.html.
department chair. This ‘solution’ is unworkable. A journal will
If you have questions, email spclref@iastate.edu or
rarely discover such violations, and a responsible dean or depart-
call (515) 294-6672.
ment chair would be reluctant to take any action without extensive
investigation and hearings. A simpler version might be, “If such a
conflict of interest is not disclosed and later discovered, the journal
will take appropriate action. The journal may include publishing
JULY 2008 AMSTAT NEWS 3
Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50  |  Page 51  |  Page 52  |  Page 53  |  Page 54  |  Page 55  |  Page 56  |  Page 57  |  Page 58  |  Page 59  |  Page 60  |  Page 61  |  Page 62  |  Page 63  |  Page 64  |  Page 65  |  Page 66  |  Page 67  |  Page 68  |  Page 69  |  Page 70  |  Page 71  |  Page 72  |  Page 73  |  Page 74  |  Page 75  |  Page 76  |  Page 77  |  Page 78  |  Page 79  |  Page 80