My role as Editor-in-Chief of the Journal of the American Society of Echocardiography (JASE) has given me the opportunity to review not only a lot of manuscripts, but also an even greater number of reviews provided by our large pool of peer reviewers. While I am not an unbiased observer, I would observe that the large majority of the reviews we have received have been done carefully and constructively. I learned a few years back, from a book about medical editing, that an effective editor must be able to provide to authors written feedback that is timely, courteous, clear, and informative. Of course, such feedback depends heavily on comments and suggestions from the reviewers as well as those that stem from discussions with the editors. While it is for others to judge, I would like to think that the critiques that we provide to our prospective authors are useful and do contribute to the quality of the articles that we publish, and even to those that eventually wind up being published elsewhere.
In my opinion, the peer review process is really about quality improvement; a decision to accept or reject a given manuscript is more the byproduct of the review process than a specific goal. While I believe that JASE’s peer review process generally does work well, I have seen an occasional and notable “gaffe”. In hopes of suggesting some areas for quality improvement in the peer review process, I would like to discuss briefly ten mistakes that good reviewers would no doubt avoid. Just to be crystal clear, the following is a list of things not to do when reviewing a manuscript!
10. Use highly critical or pejorative language. If the purpose of peer review is to provide constructive feedback to prospective authors, and not just to decide if their manuscript ought to “live or die”, then it is not very effective to couch comments in such a way that they sound mean and nasty. I don’t believe that much is accomplished by comments such as “the authors clearly did not recognize that…” or “the authors have ignored work by …”, or “the authors obviously do not understand that…” There are ways to point out areas for improvement in a kind yet firm manner without being insulting. An adage that seems to apply here is the one about “attracting more bees with honey than with vinegar”. The good reviewer sticks to factual observations, avoids excessive criticism, and phrases suggestions in a constructive manner. This does not mean that critical comments should be sugarcoated, only that they are more effective if phrased in a balanced manner.
9. Make comments that suggest that you did not read the paper carefully. In my career, I have received plenty of rejection decisions from journal editors— every author has. I believe that nothing makes an author more upset than the idea that his or her paper was turned down because of critical comments made by reviewers who did not take the time and make the effort to read the manuscript carefully. As an editor, I have on occasion (not often) run across a comment that criticizes a manuscript for failing to provide certain data or to discuss a particular issue, when it was clear from my reading of the paper that the authors did in fact include those data or discuss those issues. I confess that I would find it difficult to send such comments to authors, since they imply that we have not reviewed their paper with sufficient care. Reviewers are inevitably also authors, and good reviewers always remember the Golden Rule.
8. Make a habit of citing your own publications. Most often, the peer review process is conducted anonymously; in fact, some journals have tried sending articles for review with the names of the authors removed. I believe that most reviewers are more comfortable writing a critique when they believe that the author of the paper they are commenting upon will not know for sure who wrote the critique— after all, the tables might be turned in the future! Every so often, I see a reviewer make liberal reference in his comments to the authors to publications that were overlooked and should have been considered. When the same author wrote all of these publications, it takes little imagination for the authors of the article being critiqued to figure out who might be so familiar with all of those articles. Clearly, a reviewer should point out relevant literature that ought to have been discussed, since sometimes an author may have missed these publications. However, it is best when the reviewer makes balanced suggestions to consider key publications, and not just his own.
7. Use English language badly while being critical of the authors’ use of English language use. This is one of my favorites, and I have kept track of a few examples. One reviewer commented “The addition of extraneous words and sentences throughout the manuscript appear distracting from the essential message” when he could have said “Please write simply and clearly”. Another reviewer wrote, “Multiple spelling and grammer [sic] mistakes noted”, failing to note that he had failed to include a verb in his sentence, and had spelled “grammer” incorrectly! I have great sympathy for readers, and do not like to see them forced to read papers that are badly written and therefore hard to understand. In my view, it is great when reviewers point out ways in which writing can be improved and points made more effectively. However, if a reviewer is going to criticize the authors’ use of English language spelling, grammar, and phrasing, then he ought to do it using correct language conventions.
6. Copy your Comments to the Authors and use them as Comments for the Editors. It is worth noting that “Comments to the Authors” and “Comments for the Editors” serve very different purposes. As noted earlier, the reviewer’s comments to the authors ought to focus on the material presented in the manuscript, and ought to point out areas needing additional attention and revision. Comments to the authors ought to help the authors to improve their article— for example, by suggesting that the authors include additional measures or revise the methods used to make those measures, add more patients, analyze data differently, or expand their discussion of specific issues. Comments for the editors do not need to reiterate those points— after all, the editors can read the Comments to the Authors themselves, so copying those comments and pasting them into the Comments for the Editors section accomplishes nothing useful. Comments for the Editors ought to include the reviewer’s recommendation— that is, whether the paper ought to be accepted without changes (this is a rare occurrence for an initial submission), revised and reconsidered, or rejected outright. Even more important are the reasons for the recommendation, the logic that supports the advice given. The most helpful reviewers provide a solid, evidence based rationale to support their recommendation— so that it is not just an expert opinion, but one based on solid facts. Such comments help the editors decide not only if a given paper is important enough to publish, but also if that paper is better than other candidates for publication since priorities often must be established. The reasons supporting the reviewer’s recommendation are at least as important as the recommendation itself.
5. Do not include any Comments for the Editors. This ought to be self-explanatory. If, as just noted, the reviewer’s comments for the editors are where he ought to explain the rationale for the recommendation he has made, then failure to provide any comments effectively means that the reviewer has missed the opportunity to give weight to his recommendation and to help the editors understand the real strengths or shortcomings of the manuscript.
4. Recommend acceptance or rejection in Comments to the Authors. The decision to accept or decline a paper is one made by the editors, and not by an individual reviewer. It is not uncommon for two experienced reviewers to disagree on the merits of a paper, and in this instance the editors consider the pros and cons very carefully and sometimes get additional reviewer input. On occasion, an article that one reviewer thinks is “terrific” has major flaws that a second reviewer points out in a clear and convincing manner. Conversely, we have also seen articles that had important shortcomings when submitted initially, but which (by hard work on the part of the authors) were markedly improved after revision and ultimately merited publication. It is confusing to authors when they receive a rejection letter that includes a reviewer’s comment that “This is an extremely important paper and should be published”. Reviewers should make comments about the suitability of a manuscript for publication— but not in the comments to be transmitted to the authors. Such “recommendations” (and that is what they are – recommendations, and not final decisions) should be reserved for the reviewer’s comments to the editors. Clinicians are well aware of the adage “the physician advises; the patient decides”. Applied to peer review, this could be rephrased as “the reviewer advises; the editors decide”.
3. Consider that you must always accept invitations to review. Editors understand that it is the rare reviewer who is always able to review when asked, who always turns in his review promptly, and who always does a spectacular job. We have a few such reviewers, and they are truly special. It is probably fair to observe that good reviewers are almost always busy people who have other important commitments, and that most good reviewers are asked to review by multiple journals. Sometimes a reviewer simply does not have the time to review in a timely manner. I would rather have him decline my invitation promptly, and to suggest a colleague or two who would be well qualified to review, rather than to accept the responsibility to review and then not be able to come through with a good review in a timely manner. A quick reply that states “I’m sorry, I can’t take on this work now, but Dr. XXX would be a good alternate choice, and he can be reached at …” allows the editor to keep the review process moving along smoothly.
2. Make a habit of turning in your review late. I realize that doing a good job as a peer reviewer takes time and sometimes seems to offer few rewards; peer reviewers are not paid for reviewing, and they would probably make their bosses happier by using the time to write their own papers instead of using it to review papers written by others. On the other hand, authors seem invariably anxious about the fate of their submitted manuscripts, and editors like to be able to brag about the efficiency of their peer review process. I understand that sometimes circumstances arise that interfere with prompt completion of a promised review, but this should be the exception and not the rule. It is difficult for authors and for editors when peer reviewers are inevitably late with their comments, and I can say with conviction that probably the least enjoyable responsibility of an editor is having to hound a reviewer repeatedly to turn in his late review.
1. Agree to review a manuscript, but then don’t turn in a review and ignore requests to complete the work you promised to do. As you might imagine, this is an incredibly annoying situation, but it does occasionally happen. An editor must often be patient, but everyone’s patience has a limit, and sometimes it is necessary to move on and to “cut loose” the reviewer who agreed to review but then failed to do so. A promised review that never materializes forces the editor either to invite additional reviewers and thereby to prolong the review process, or to make a decision without the input of an expert whose opinion was wanted.
As noted earlier, I believe that peer review is about quality improvement. While most of our reviewers do an excellent job of critiquing manuscripts in a careful and constructive manner, the opportunity to observe the occasional mistakes made by others gives reviewers a chance to do an even better job the next time they are invited to review a manuscript, either by JASE or by another journal. I hope this short list is helpful in that regard. As always, if you have comments or suggestions, I can be reached at firstname.lastname@example.org .