How to avoid a hostile takeover of first authorship during medical leave?

A colleague has approached me with a tough situation. I'm unsure how to advise her.

She is a postdoc, and will soon take a month-long planned medical leave. Her adviser, a senior research scientist, has a history of reshuffling authorships of papers of his subordinates during leave, sometimes to make himself first author. She had previously hoped to have the paper in question in submission, but “serendipitously” some last minute data is unavailable. Adding to her suspicion, the adviser has asked her to deliver all drafts and all data analysis code before leaving.

Her question, which stumped me: how can she safeguard her present first-author position? I will meet with her at the end of the week (17th) and share this thread with her.

Some details:

• 
  

  This advisor has reshuffled authorship on several previous occasions of which I am aware. He has suffered no disciplinary action from the department in these situations, and has a reputation for working to damage the careers of those that work against him. The unit has a history of turning a blind eye in these situations.

  
  


• 
  

  There's very little accountability to be had within this unit. Pieces of proof like emails and widely witnessed public statements are unlikely to be actual safeguards.

  
  
  


• 
  

  The advisor really does control all the inputs and outputs. This is true even after her return: he may switch authorship at that point over any real or perceived failure.

  
  


• 
  

  My feeling is the successful path here will involve some clever way to move the power in the situation from his side to her side.

  
  


• 
  

  Certainly, she is looking for a exit from the lab. A solution will substantially increase the chances she can take this first authorship with her.

  
  



• 
  

  My sense from our discussion is that early drafts and data analysis already shared, but the most recent ones are not yet. That said, the entire situation seems to suggest a lack of trust on all sides.

  
  


• 
  

  The distribution of work put into the paper is roughly: 95 % by my colleague, 4 % by another co-author, and 1 % by the supervisor.

  
  

EDIT: This has concluded. She used several of the strategies here, documenting extensively. Her adviser remover her from the paper and sent it for publication within 48 hours of her leaving. Upon her recovery, which took somewhat longer than expected, she appealed to the editor and the department. Her adviser ended her postdoc position, citing insufficient funding, and responded to the editor (how I'm unsure). She appealed to the university, which did not intervene. She has had no additional communication from the journal or her old lab. She has left, in some disgust, for an industry job.

Answer

Well, there's always the Gordian knot solution: Don't take the medical leave and continue working, to the detriment of her health. If she's not on medical leave, the senior researcher can't use her lack of presence to screw her over. Whether that's something she's willing to consider depends on what exactly she's taking the medical leave for, and whether she's willing to take the damage to her health to avoid the damage to her career.