Authorship

1
Authorship

• Bernard Lo, M.D.
• August 27, 2009

2
Questions

• Looked self up in Pub Med?
• Omitted as author?
• Co-author didnt deserve it?
• Asked to add author who didnt deserve it?

3
Survey of UCSF fellows

• Omitted as author 20
• Co-author didnt deserve it 38
• Asked to make someone author who didnt deserve
  it 37

4
Case 1 Prior agreements fail

• You are second author
• First author not analyzing data or writing paper
• You want to take lead, get paper out
• What would you do?

5
Case 1 Prior agreements fail

• Participants enrolled, data collected and entered
  into statistical program
• First author not analyzing data or writing paper
• You want to take lead, get paper out
• What would you do?

6
Case 1 Prior agreements fail

• What would you do?
• Send an ultimatum to your colleague
• Get your mentor to pressure him
• Forget about the project and move on
• Not sure

7
Case 2 Added author

• Division chief asks to be author
• Comments in seminars and on abstract
• Not participate in design or analysis
• What would you do?

8
Case 2 Added author

• What would you do?
• Hold your nose and do it
• Refuse and stand up for your principles
• Ask your department chair to intervene
• Forget about the project and move on
• Not sure

9
Question

• How did you feel about this authorship experience?

10
Outline of session

• Criteria for authorship
• Problems with authorship
• Practical dilemmas

11
Why have authorship?

• Recognition
• Job, grants, promotions
• Accountability
• Prevent fabrication, fraud, plagiarism

12
Criteria for authorship

• Conception and design or data analysis and
  interpretation, AND
• Drafting or substantially revising article, AND
• Approving final manuscript

13
Criteria for authorship

• Not merely
• Funding or equipment
• Collection of data
• Supervision of research group

14
(No Transcript)
15
Questions?
16
Problems with authorship

• Publish articles that shouldnt be published
• False, fabricated data
• Duplicate publications

17
Problems with authorship

• Fail to publish articles that should be
• Negative results
• Too many authors honorary authors
• People listed who shouldnt be
• Too few authors ghost authors
• People omitted who should be authors

18
Advantage study (2003)

• Randomized trial of 5557 patients
• Refecoxib vs. naproxen
• Discontinue Rx for GI reasons 5.9 vs. 8.1

19
Advantage study

• MIs 5 on rofecoxib vs. 1 on naproxen
• 3 additional rofecoxib deaths not reported

20
Statement by lead author

• Merck designed the trial, paid for the trial,
  ran the trial. . . Merck came to me after the
  study was completed and said, We want your help
  to work on the paper. The initial paper was
  written at Merck, and then was sent to me for
  editing

21
Ghostwriting

• Asked by medical education company to write a
  review paper on interactions between warfarin and
  dietary supplements sponsored by drug company
• JGIM 2005 20 546

22
Ghostwriting

• Received draft article, with name on title page
• Company developing oral anticoagulant
• No mention of product
• Biased against warfarin

23
Ghostwriting

• Later asked to review same paper
• No mention of ghost author
• No mention of drug company sponsorship

24
Are these isolated cases?
25
Problematic authorship

• Honorary authors 21
• Ghost authors 13
• Ghosts acknowledged 0
• JAMA 1998 280222

26
Problematic authorship

• No substantial contributions 26
• Provided subjects, materials,
• lab, technical assistance 58
• Collected data 25
• JAMA 1994 271 438

27
Preventing ghost authorship

• All persons who had input into writing must be
  author or acknowledged
• All persons named as authors or acknowledged must
  complete financial disclosure

28
Journal requirements for industry-sponsored
research

• Full responsibility for trial
• Access to data
• Data analysis
• Control over publication
• Including data detrimental to product
• Disclose financial relationships
• Including payment for writing

29
Questions?
30
Duplicate publication

• Articles in systematic reviews 1234
• Duplicate 103 (8)
• No cross reference 63
• Translations 12
• JAMA 2004 291 974

31
Types of duplication

• Identical sample and outcomes
• Combine several articles
• Report different outcomes on sample
• New data added to preliminary article
• Part of larger trial, same outcomes

32
What is wrong with multiple publications?

• Inefficient transfer of information
• Bias in evidence base

33
Problems with authorship

• Authors in wrong positions

34
Whos on second?

• Less prestige than first
• Middle authors contribute even less
• Last author often senior
• Not cited after 6th

35
Who understands order?

• Not journal editors
• Not deans

36
Survey of department chairs

• Fictitious article and authors
• Infer authors contributions
• Epidemiology 2004 15 125

37
Contributions of authors

• Little idea of roles of any author
• If corresponding author, more credit

38
Documentation of authorship

• Describe specific contributions
• In manuscript
• In promotion packet

39
Concussions in NFL players

• Retrospective review of data from team physicians
• Return to play not associated with increased risk
  of second concussion

40
Conclusion

• Might be safe for college/high school football
  players to be cleared to return to play on the
  same day as the injury
• Keep an open mind to possibility that present
  analysis of professional football players may
  have relevance to college and high school
  players.

41
Dispute among 5 authors

• Two disagreed
• One said passage added without her knowledge

42
Lead author

• Proofs were sent to each author
• No need to point out new passage
• If people who are not scientists or physicians
  are misinterpreting it, that is not the
  responsibility of those who wrote it.

43
Reactions?
44
Case 1 Prior agreements fail

• What would you do?
• Send an ultimatum to your colleague
• Get your mentor to pressure him
• Forget about the project and move on
• Not sure

45
Excuses

• Its in the pipeline
• Its next in the pipeline
• BMJ 1994 309 1739

46
Excuses

• Im reanalyzing the data
• The data are on a Windows computer
• I cant find the right statistical test to prove
  it worked

47
Pragmatic concerns

• Power differences
• Future repercussions
• Is it worth the hassle?
• Can you live with yourself?

48
Just do it, diplomatically

• I know youre very busy. Im willing to take
  over as 1st author and write a draft.
• If I havent heard in 3 weeks, Ill assume
  youre too busy to be first author.

49
Questions?
50
Case 2 Added author

• What would you do?
• Hold your nose and do it
• Refuse and stand up for your principles
• Ask your department chair to intervene

51
Just say no, diplomatically

• The journal insists that all authors sign that
  they have met a list of requirements. I would
  feel very awkward signing this. Id like your
  permission to give you a big thank you in the
  acknowledgments.

52
Questions?
53
Emotional impact of authorship disputes

• After disputes commonly feel
• Angry
• Hurt
• Taken advantage of

54
Take home points

• Be explicit about authorship positions and
  responsibilities
• Spell out arrangements in advance