By the time this editorial is printed, we will have been coeditors for almost a year, and we thought it time to offer some observations based on our experience to date. While it would be fun to share our stories of awards the journal has received for the excellence of our authors' work, agonizing decisions involved in the process of accepting a paper, or even the re-markable improvements in some manuscripts based on reviewers' comments, we will save those for another time. Instead, we offer an insider perspective on the process by which (some) manuscripts become published HSR papers. This process has been a major focus for us in the last year because of the transitions we have been making in the “back office” functions here at HSR. Having submitted many articles ourselves to HSR and other journals, however, we recognize how often the author is in the dark about what is happening with his or her paper and how a little information might be helpful to everyone. What follows is a description of how things should operate, recognizing that there are occasional exceptions and that some of the details will change as we make yet another transition to a new web-based submission and management system. (More on that at the end of this editorial.)
Getting Started
The process ideally begins well before the paper is completed. The author has hopefully examined a number of recent HSR issues to determine the types of papers that we usually publish, the methods typically used, and the style of the journal. Generally speaking, one finds that in the past we have not published extensive literature reviews, highly theoretical expositions, or clinical case reports. That is not to say that we may not be open to such papers, but that an author must provide a compelling case for why such a manuscript will be important to our readers. The wise author will have checked our web site for “instructions to the author” (see http://www.HSR.org in the section called “info for authors”). This will provide important guidance with respect to length, style, and other features as well as requests for specific information, such as disclosures of conflicts of interest, and recommendations for reviewers. Since we ask that manuscripts be no more than 25 pages, it is far better to write a 25-page paper the first time, knowing that reviewers will almost certainly ask for elaboration of some points, than to start out with a 40-page paper. (The latter is also more likely to result in a grumpy reviewer—not a good strategy.)
Health Services Research publishes articles in both hard copy and electronic form. The editors have a “page budget” for each volume, and going over the budget is expensive in terms of printing costs. Electrons are (almost) free, so we can put extensive additional information in appendices in the electronic version, which can then be referred to in print. Knowing this, the wise author provides as much detail as is needed in appendices, leaving fewer unanswered questions for reviewers, and a cleaner, easier-to-follow text for the majority of readers. Designing the paper with such opportunities in mind is better than cutting and pasting after the fact.
The author, well before submitting, has also presumably considered carefully our policy on “prior publication” (also located on our web page). Although this is not as restrictive as those of many biomedical journals, this policy statement reflects guidelines developed in common with many sister journals publishing health services articles. This policy points out that there is little reason for us to publish material that is already widely available to our audience, irrespective of copyright problems. On the other hand, some types of prior dissemination are not a problem for us. Since the posting of our policy, we have had a number of authors ask in advance whether certain actions they are considering would jeopardize publication of a manuscript in HSR, and we have offered advice on a case-by-case basis. Authors sometimes have multiple papers on closely related topics. To allay concerns about duplicate publication, it is best to send copies of such papers and include in the cover letter a discussion of how they differ from the one submitted to HSR.
Anticipating the review process is a wise idea. Authors are encouraged to anticipate what our reviewers will be asked to judge about their manuscript and what types of expertise are needed to review the manuscript. Reviewers are in turn asked to provide constructive comments to help the authors improve the manuscript as well as to advise the editors about whether to reject, revise, or accept the manuscript for HSR. For these reasons, we believe it is important for authors to understand the reviewers' functions and to actively participate as reviewers for HSR too.
The Initial Decisions: The Managing Editor and Coeditors
The managing editor is the first person to receive a manuscript, logging it in and assigning a code number. The managing editor in general helps to “manage” the administrative details and the flow of the manuscript through all the respective steps, monitoring for logjams or other communication needs to push it along the review process. We try to have blinded reviews, so the authors' instructions indicate that author names should only appear in certain places. Many word processing programs, however, place the name on your computer in the file automatically, so this needs to be stripped out. Currently done manually by the managing editor, this will be handled automatically by our new “Manuscript Central” web site, which converts the manuscript to a PDF file format and simultaneously strips it of identifying properties. After completing these initial administrative details to prepare the manuscript for review, the managing editor assigns the manuscript to one of the two coeditors. This is usually done in a manner to balance workflow, but authors can request one of us based on perceived expertise, and we routinely juggle assignments to avoid conflicts, such as when authors are at one of our institutions or are otherwise close colleagues.
The coeditor examines the manuscript, making several decisions. The first is whether the manuscript clearly does not belong in HSR. Typically about 20 percent of all new submissions fall into this category and are rejected at this step, but the reasons vary. Authors will always be disappointed by a rejection, but if we can determine at the outset that a paper is almost certain to be rejected, we believe that it is best to let the authors know quickly so they can move on. Occasionally the quality of the paper is clearly not up to our standards, and this is apparent after even a cursory review. Some papers are apparently well done, but address issues not of primary interest to the HSR readership. Sometimes the topics are appropriate, but the findings add little that is new for the general reader, even though they may be of interest to a narrower audience. If possible, we offer a few suggestions to the author along with the rejection.
Manuscripts that clear this first hurdle are then assigned to one of the senior associate editors (SAEs), who are given the coeditor's suggestions for reviewers. The SAEs serve three major roles: they (1) provide another layer to determine whether the manuscript is worthy of external review; (2) make the final selection of reviewers, and (3) initially receive the reviewers' comments and advise the coeditor about whether to accept or reject or to suggest how to revise the manuscript. The choice of SAE is usually based on expertise relevant to the manuscript, but we also give some consideration to balancing workload. Authors may suggest an SAE, and we will usually honor that request, although we are also careful to try to avoid potential conflicts of interest. The coeditor will occasionally add some notes for the reviewers to focus on when reading the manuscript, or to the SAE offering a rationale why he or she thinks the paper is worthy of publication and technically solid. For example, the findings may not be new, but the methods used may be innovative and worthy of spotlighting.
The Reviewers
The process of identifying two people who will commit to reviewing this manuscript in a timely manner begins after the SAE's concurrence that the manuscript is worthy of external review. The coeditors and the SAEs are continuously developing a list of potential reviewers based on various sources, and they select a set of potential reviewers for a specific manuscript based on their own knowledge of the field and the kinds of expertise particularly needed to evaluate the manuscript.
Reviewers are asked to read a manuscript very carefully, to provide constructive feedback to the authors, and to provide a rationale to the editors to back up their advice about publication. Note that we explicitly ask reviewers to separate their advice to the authors about how to improve the manuscript from their advice to the editors about whether to publish it and their summary evaluations.
Reviewing manuscripts is one of those time-consuming, yet underrewarded, tasks necessary for the field. The principal reward to reviewers for several hours' work is primarily knowledge that they are helping the field improve, although we try to provide some feedback and thanks from the journal. Good reviewers who provide these timely and helpful critiques are hard to find, and thus are often “rewarded” by editors by overloading them with manuscripts.
Especially when manuscripts address rather narrowly focused topics or use uncommon methods, the pool of potential reviewers may be small, and may be unknown to the editors. Hence, suggesting new reviewers, especially with appropriate contact information, is always valuable. In particular, we encourage authors to suggest potential reviewers for their manuscript (as well as to let us know if there are people who the author feels might not be unbiased reviewers). Aside from following all our submission rules, this is the most important place where the author can help address the biggest logjam in the whole review process: finding two good reviewers who agree to look at the manuscript within a short timeframe. We also ask authors to select keywords that describe the subject area and methods used in their manuscript. These too can be used to search our database of reviewers for potential matches.
Sometimes logjams or other complications accompany the process of obtaining two external reviews. Most reviewers are very responsible about providing a useful and a timely review, but crises occasionally occur and obligations are occasionally forgotten. Periodic reminders from the managing editor when reviews are late are often, but not always, effective. We strongly prefer to have two reviews (plus the SAE's assessment) for each manuscript, but occasionally we will rely on a more detailed review by the SAE in place of a second external review so as to not hold up a decision too much.
Finally, having a paper rejected when the comments to the author seem to raise only minor points may be a source of confusion for authors. As we describe below, our decision is based on more than the comments that are provided to authors. For this reason, we usually try to summarize the substance of the confidential comments in the letter from the SAE and coeditor assigned to the manuscript.
Making Decisions
Reviewers are valued consultants to the editors, but their recommendations are just that—recommendations, not votes or “blackballs.” The SAEs will sometimes decide that even though both reviewers liked the paper, it is not worthy of publication. Sometimes the opposite occurs—both reviewers “pan” the paper, but the SAE sees a “diamond in the rough” that, with substantial work, can become a publishable and perhaps even an important paper. The SAE has four primary categories to choose from: Accept, Minor Revision, Major Revision, and Reject. After making a preliminary decision and adding comments to the author, this is passed on to the coeditor (typically the one who began the process) for further review and occasional discussion before finalizing it to send to the author.
Having been on the receiving end of many “revise and resubmit” letters that resulted in rejections even after responding effectively to each point raised by the reviewers, we are quite sensitive to how our letters to authors will be interpreted. Manuscripts undergoing initial review are hardly ever accepted outright. If the SAE thinks the authors can respond easily to the reviewers' comments and that the necessary changes are unlikely to alter the major findings of the paper, this is considered a “minor revision,” even though it may require a good deal of work by the authors, and there is still no guarantee of acceptance. The revised manuscript may be sent back to the reviewers, especially if they had requested to see the revised draft. Assuming their concerns are addressed and no new flaws are uncovered (as occasionally happens on a second reading) the paper is then likely to be accepted, or accepted conditional on relatively minor changes, such as a shortened or reworded title.
Reviewers occasionally identify “fatal flaws” or feel the author has used “inappropriate methods” and thus recommend rejection in fairly strong terms. (Usually, these conclusions are made in the confidential comments.) The SAE or coeditor, however, may feel that the authors address an important issue or have a valuable dataset, and that something might yet be done that would be a contribution. This is a difficult situation, because the potential analyses may not be feasible, and even if successful, may end up with findings that are not very interesting. For example, the authors' initial attempt may have produced new and exciting results, but when redone with appropriate methods, the findings are rather humdrum. An encouraging “revise and resubmit” letter that requires substantial effort and yet results in an ultimate rejection is a double disappointment. Therefore, in such situations the letter we send is clear about the uncertainty involved and attempts to clearly alert the author to the risks involved. Not all of these “major revision” letters result in a resubmission, which means our message of risk is being understood.
“Reject” letters are fairly clear, although we have had authors who misinterpreted them and subsequently sent in an uninvited revised manuscript. Other authors may, after careful reflection on the reviews, feel we and our reviewers misinterpreted the manuscript and that what we thought was a fatal flaw is not. Sometimes a request for reconsideration, especially if there is an admission of a major problem in the presentation that somehow led us all astray, is successful, but rarely so.
The Revision Cycle
With a “revise and resubmit” letter of any variety, the ball is back in the author's court. Occasionally an author feels the required changes are too substantial to undertake. In that case, we very much appreciate a note so that we can close the file on the initial submission. In other cases, the revised manuscript is resubmitted, with a memo that provides a detailed response to each of the points raised by the reviewers and editors. These responses are sometimes pointers to how the paper has been changed. In other instances the author offers an explanation of why he or she does not think changes are warranted. It is certainly possible that a reviewer may be wrong in his or her critique, but sometimes that error is due to the fact that the author's approach, while correct, was not well explained; such cases require at least a minor rewording of the text.
At times a reviewer will ask for a new bit of analysis that results in what is essentially a private letter from the author to the editors and reviewer, which entails little or no change in the manuscript. In other situations an extensive appendix is prepared that explains in some detail what has been done. While in the past authors have had to omit these appendices from the published version, and at best refer interested readers to request details from the corresponding author, we now have much more flexibility in the electronic version of the HSR and can use this venue to make available additional detail for our readers.
Most revised manuscripts, along with the author's response to reviewers, are sent back to the initial reviewers for a reassessment. This is particularly important when reviewers have requested major revisions. Since most suggestions require additions to the manuscript, and few manuscripts are submitted that are below our page limits, other material will need to be deleted. Including an additional copy of the manuscript using the “track changes” or “compare documents” features makes it far easier for the reviewers and editors to see if some of the things they liked in the first version have been sacrificed.
Some papers are accepted after the first revision. Many more receive an acceptance conditional on minor editorial changes. Some come back with further significant revisions needed, and with some, it becomes clear that the changes are moving us away from an acceptable paper. While the process may seem interminable, it is intended to improve the paper, and usually converges on an acceptable product. Occasionally, however, the clarifications elicited by reviewer requests uncover irreparable problems. In other instances, authors are just unwilling to make the changes necessary for acceptance at HSR.
While the SAEs have substantial discretion in the decisions about their manuscripts, the coeditors at HSR are involved in the decisions at each round. In some journals, multiple rounds of revisions with an associate editor result only in moving up the final decision-making session at the top editorial level, at which the surviving manuscripts are culled to fit into limited journal space. At HSR, we generally make the decision about whether a subject warrants journal space earlier in the process, so if the paper meets our quality standards, we will usually decide to publish it. There are no guarantees, however. For example, if while the manuscript is under review, another article is published that essentially has the same findings, this would be likely to reduce the contributions of the paper under review.
Once accepted, there are still some steps before authors actually see their manuscripts in print. One of the coeditors takes responsibility for organizing the accepted manuscripts into an “issue.” Usually, manuscripts are placed into the next open issue, but on rare occasions a paper may be held over for the next issue to place several manuscripts together in a section or to incorporate it into a special section. Copyediting, getting author responses to questions, typesetting, and so on are still time-consuming, so these production steps can take five to six months. (However, an author can legitimately claim their article as being “forthcoming in HSR” as soon as it is accepted.)
Speeding up the Process
By the time this editorial is in your hands, we should be well along in our efforts to implement a second transition in our “back office” process. The first occurred in September 2002, when we ceased accepting hard copy manuscripts. E-mail has lowered FedEx costs and sped the transmission of information among the SAEs, coeditors, and the editorial office, but it has its own problems. Blinding manuscripts for review is a chore, and simply moving things around electronically has not either improved workflow or used the powers of computers effectively.
Our second transition is to Manuscript Central, a web-based system developed by ScholarOne and supported by Blackwell Publishing Company, HSR's publisher. The promise of this system is substantial, and we will let you know in a subsequent issue about how well that promise has been met. Aside from making our lives easier, it should offer significant benefits to authors and reviewers, including easy access to information about the progress of the manuscript through these many steps. Potential reviewers are invited to sign in at http://hsr.manuscriptcentral.com with their contact information and to select any number of areas of expertise and subject interest that describe themselves from a predetermined classification list. Reviewers can also let us know when they are periodically not available for new reviews. Authors, when submitting via this new system, select from the same classification list all those key words best describing their manuscript. The editors thus have substantially more information to help make an appropriate match between reviewers and manuscript. This will, we hope, result in better and faster reviews for our authors and a more well-matched and evenly shared burden for our reviewers. We also hope to use the new system to reduce the burden on our editorial staff and otherwise speed the process, benefiting us all. Our goal is to reduce the unnecessary steps and systematize the routine, thus allowing us to focus on improving the science that we publish in HSR.