Should I become first author of a paper for which I did the statistical analysis?

In medicine, as in most biological fields that I know of, the first author is the person who did most of the hands-on work, while the last author is the one who had the idea and planned the project. From what you've said, then, it would be appropriate for you to be first author and the doctor to be last.

That said, if you don't feel comfortable being first author (and especially in light of your comment that I do not agree fully with all the conclusions) it's also appropriate for you to request to not be first author -- some middle position (second author, or whatever) is also perfectly suitable for your contributions, from the sound of it.

Answering this from the perspective of a biomedical researcher who is often in the position of "We had this question, and this data...can you make something of it?"

First, the conventional view of authorship positions":

• First author: The person who did most of the work
• Last author: The senior author on the paper. Usually a conceptually guiding force, and the type of person who could be summed up as "Without X, none of this would have been possible..." even if they wrote no code, swabbed no samples, etc.
• Others: Everyone else.

From what you're describing, it sounds like there are only two of you. As such, it seems like the logical ordering is you as first author and the doctor as second author. Both of these are relatively prestigious positions in terms of authorship, so putting her last isn't a negative drain on her career. It indicates exactly what's happening - she's running a lab and directing research.

So basically the reason she wants to put me as first author is to put herself as last author.

Her reasoning is correct.

I feel uncomfortable being first author and would prefer being second author. I am probably the one putting the most hours into the project by preparing all the data, doing the statistics and explaining the results to the doctor. The statistical analysis will be a big part of the paper. But the research question is not my idea, I have no medical background, the doctor is writing the paper. I am basically getting paid to do the analysis.

If there are only two authors, her reasoning is entirely correct, and first author is the logical place for you to be.

While I think it is interesting research, I do not identify with it and I do not agree fully with all the conclusions.

This is marginally more concerning to me than whether you are first or last.

So far I worked under the assumption to become co-author and the idea of making me first author came up recently. My opinion is that I should only be first author of my own research, which would be in statistics, not medicine.

I would be extremely confused if you were last author on this paper, and would make assumptions about what that meant regarding the project that were much more inaccurate than the ones I'd make if you were first author.

Is it harsh/unusual to insist on being co-author only?

This would only make sense if there are more people on the team, and even then...

Who would you suggest should be first author in this case?

You.