Are abstracts confidential during the review process?

Question

In an answer to a question about the confidentiality of reviews, I basically said you cannot reveal information about the review until the paper is publicly available. This question about revealing information after publication is making me rethink my answer. The question I have is, is the information (e.g., title, authors and abstract) that you are given to decide if you want to review confidential?

It seems to me that the process of agreeing to do a review is NOT

1. Please review our reviewer guidelines (including confidentiality policies),
2. If you accept these guidelines please look at this abstract and let us know if you want to review.

Rather, it seems it is

1. Please look at this abstract
2. If it interests you, please consider our review guidelines.

This suggests to me that the initial information (title, authors and abstract in a non double blind review) are not confidential.

Answer 1

To complement other answers and comments: indeed, as far as I can tell, from a quasi-legalistic viewpoint "consent" cannot be pushed on me by sending me something in email, e.g., an abstract and asking whether I'd review the article. Nevertheless, it is my firm impression that, there is a strong expectation that any such information is kept confidential in perpetuity, whether or not one agrees to referee/review.

Yes, I agree, there is something a touch unfair or burdensome about this, since one can imagine that a malicious editor could wreak havoc with one's work by sending a steady stream of one's competitors' as-yet-unpublished work... thus seemingly obliging one to disrupt one's own work... ?

And, yes, something like this does sometimes happen when one served as NSF reviewer (in the older system), especially, where work-in-progress is sometimes portrayed.

Despite the potential for abuse in having others put obligations upon us, it seems that the potential for abuse, in the line of "conflict of interest", is substantially greater if confidentiality is not essentially promised implicitly, and in perpetuity.

Answer 2

I checked the peer review policy of Nature and Science, I guess that should is the best we can get. The policy can be found here, here (both Nature) and here (Science).

A bunch of relevant quotes from the Nature policies:

As a condition of agreeing to assess the manuscript, all reviewers undertake to keep submitted manuscripts and associated data confidential...

Nature journals keep confidential all details about a submitted manuscript and do not comment to any outside organization ...

Referees of manuscripts submitted to Nature journals undertake in advance to maintain confidentiality of manuscripts and any associated supplementary data.

And some from the Science review policy:

Reviewers are contacted before being sent a paper and are asked to return comments within 1 to 2 weeks for most papers.

The submitted manuscript is a privileged communication and must be treated as a confidential document.

I would read it conservatively. That is, consider an abstract to be a part of the manuscript, hence the same rules governing general manuscript should apply. Apart from that, both journals state that in advance to being sent the manuscript, the (I guess still potential) reviewers agree to keep the matter confident.

I think an answer may also lie in the exact process handling review. First, a potential reviewer is invited and sees a title and possibly paper's authors. Then he/she has to log-in into a review system, but I guess, somewhere in the process accepts general terms and conditions which established the contract. And only then the reviewer sees the manuscript. I am not sure if all journals do it this way, but those I reviewed for did so. Well, besides some special issues handled outside the journal submission/review system, where the case would be unclear.

---

I just checked the reviewer invitations I received from journals in my field and I found that the only disclosed thing in the invitation e-mails was the submission's title. I must accept/decline to review and only afterwards I saw the manuscript, or its abstract.

Answer 3

Well, as you said, there is no rule that explicitly says that the information mentioned in a “invitation to review” is confidential. Yet, as with the related questions, there is a big grey area around the process of peer-review. Many people would expect this information to be kept confidential… and it does make sense: after all, the reason you gained access to what is (at that stage) privileged information is for the purpose that you may review it. It is part of the review process.

Let's see it another way: this situation is not so different from the case where you would accept a review, then drop it (write to the editor to pull out) upon learning that you won't have time to do it. Morally, those two situations are close to one another.