Peer Review violation

Published by the National Science Foundation, Office of Inspector General on 1997-03-31.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

                                    CLOSEOLTT FOR M96110042

On 21 November 1996, a Deputy Division ~irector'informed OIG of a allegation she had
received from a ~cientist.~  The scientist alleged that an ad hoc reviewer3 had received two
proposals4for review. The PIS for the two proposals were direct competitors. Allegedly, the
reviewer, in violation of the confidentiality of merit review, contacted PI #14 and discussed
the contents of the two proposals. Allegedly, the reviewer told PI #1 that PI #2's4 proposal
contained certain data and suggested that PI #1 submit comparable preliminary data to
strengthen her proposal. Allegedly, PI #1 told a student5 about the conversation with the
reviewer and the student, in turn, told PI #2. The Division noted that PI #1 had submitted an
update to her proposal in time for the panel review and that the update contained preliminary

The Division informed OIG that the reviewer had a pending proposal6 at NSF and the
curriculum vitae included in it identified PI #1 and PI #2 as his research collaborators. The
two PI'S proposals each included a citation to a paper7 co-authored by PI #I, PI #2, and the
reviewer. The reviewer's proposal was submitted before he received the two proposals for
merit review. However, he had not contacted NSF to discuss any possible conflict of interest
he might have with the two PIS when he subsequently received their proposals for review.
NSF's Proposal Evaluation Form (NSF Form 1) instructs reviewers to disclose any affiliation
that might be considered a conflict of interest, and in the absence of such disclosure NSF
assumes that the reviewer has no conflicting affiliations. According to NSF's Conflict-of-
Interest Rules and Standards of Conduct (Manual 15) in force at the time, collaborative
relationships within 48 months preceding the requested review are considered to be
potentially biasing. Program officers told OIG that they take this restriction seriously and
have disqualified reviewers because of existing or past collaborative relationships.

In response to OIG's inquiry, PI #1 said that she had not been contacted by any reviewer about

her proposal. She said she had.learned from a previous proposal submission that updating the


the Directorate forb

  Proposal     a
  The scientist is Dr. !1
                          is the Deputy Division Director for the Division o

                                             He is the Associate Chairman for b-a!t

    The Division deduced that the reviewer was Dr. a

(PI #I). PI #1 is a faculty member in the Department) of

                                                                            facultv member in the De~artmentof

                                                                             was submitted by ~r.4-b
                                                                                                                   ( in

                                                                                      a t ~ n i v z s i 6- ~. r o ~ o s a l

                entitled, '     -                                                                                 was
submitted by Dr.                  (PI #2). PI #2 is a faculty member in the Department4 of  - J                     at
-University.              PI #2 is the spouse of the ad hoe reviewer.
  The student, D       r       , was a graduate student in PI #2's laboratory.
  Proposal s                 entitled                          --       --         #I

   he paper is,                                                                a       n      d      . 1992.
           -                  - --
                                 -                               -                         R"         -- -
mP             -

                                               Page 1 of 2                                          M96-42
                              CLOSEOUT FOR M96110042

submission with preliminary data was a good idea. She had submitted the update this time
because of that experience, not in response to a reviewer's prompting. OIG also contacted the           -
student who said that he had spoken with PI #1 and that the PI did know that her proposal
was in competition with PI #2's. According to the student, PI #1 did not say how she knew
this nor did she say that she had been contacted by a reviewer. The student said the two PIS
had attended a scientific meeting and may have exchanged information at the meeting.

In response to OIG's inquiry about violation of confidential merit review and possible
undisclosed conflicts of interest the reviewer said that he was an experienced reviewer, he
understood and respected the rules, and he had not contacted PI #1 in violation of those rules.
He said that he knew both PIS but he had no current collaborative relationship with them. He
characterized his prior collaboration with them as "limited" and said he had disclosed it in his
proposal because, even though the research for the paper was completed in 1991, the paper
was finally published in 1992 (technically within the 48 month restriction period). He said he
had provided PI #1 with a test compound used in the paper and experiments he had conducted
in 1988 were repeated and included in the paper. PI #lts work on the project with PI #2 was
independent of him. He said he had not disclosed his past collaborative relationship with the
PIS to the NSF program officer when he received the two proposals for review because he did
not feel his past affiliation created a conflict of interest and he felt he could be objective in his

OIG concluded that there was no evidence that the reviewer had contacted PI #1 or that he
had violated the confidentiality of merit review. Although the reviewer had failed to disclose
to NSF affiliations that created the appearance of a conflict of interest when he received the
proposals for review, it is doubtful that the relationship he described would be considered
disqualifiing or limiting. However, for the merit review process to work as fairly and
objectively as possible it is NSF, not the reviewers, that must determine whether a reviewer's
collaborative relationships disqualifl or limit his or her review activities. OIG told the
reviewer that he should have disclosed this relationship to NSF before he submitted his
reviews or, at the latest, along with them and instructed him to disclose relevant collaborative
relationships in the future.

This inquiry is closed and no further action will be taken in this case.

cc: Staff Scientist, AIG-Oversight, IG

                                        Page 1 of 2