Dr. Blackstone’s comments re: writing a paper

[While these comments were in reference to a specific manuscript draft, they have general applicability]

You'll also forgive me for approaching the paper more from the vantage point of a long-time journal editor.

MESSAGE

The single most important thing to establish before writing a paper is its message. It can be stated generally in 2 sentences--we call this the ultra-mini-abstract for JTCVS. Being clear about this guides your statement of objectives and organization of the paper, such that it drives in a focused way toward supporting the message (conclusions/inferences/essence). [Note: It requires on average 3 hours to write a 50-word ultra-mini-abstract.]

Ifind it difficult at times to discern where you are heading with this paper, because what has been done seems not strictly tied to the objectives, nor do I see the tie-in between your conclusions and the data presented.

INTRODUCTION

I tend to take the advice of the NIH. You capture the attention of readers within 2 sentences or the application / paper will not be read! Yours is not a review paper; it is a focused paper (or should be). It takes you until paragraph 3 to know where you are heading with this paper and to state the importance of the study!

I have carefully read and reread the statement of purpose and find it difficult to connect the purpose with anything you present in a logical step-by-step path--there are dots, but I fail to see how the dots are connected. You state you are performing primarily a validation study--but where is that?

METHODS

You say you want to validate a specific measurement of unbalance, but then give your criterion for unbalanced. It would seem that you would be using a data-driven approach to define unbalanced on the basis of the measurement.

I then come upon a competing risk analysis, and it comes out of the blue. How does this relate to your purpose(s)? How does it help with validating the measurement? What is it addressing? It may be the right thing to do, but I don't know why it is needed and appropriate.

RESULTS

Patient characteristics belong under Patients and Methods. Everything prior to time zero (where the study begins) should be under Patients and Methods.

Then mortality...is this part of the strategy to validate the measurement?

Then a frequency plot--strange for a heading; headings should not be about methods or presentation, but about what the topic is. Ideally, every section under Results (and Discussion) should relate directly to one of your objectives (and if you have an overall objective, then a sub-aim).

Under the title of competing risks (again a method), I don't see the tie in directly to a purpose that is keyed to defining unbalanced.

DISCUSSION

Again, although I like subsections of a discussion, each should match an aim, which in turn matches a section of Results, etc, etc.