In an earlier editorial, we gave recognition to 227 persons who reviewed manuscripts for Tobacco Control from late 1993 to the end of 1995.1 Appended to this editorial is another list of 227 individuals—those who reviewed papers for the journal as outside referees or co-editors in 1996 and 1997. This list includes those who reviewed papers for two special supplements to the journal which were published in 1997.2-3 As usual, we express our appreciation to them all, as the quality of the journal depends on the vital service they perform.

In a 1993 editorial I explained how we performed peer review at that time.4 We have used a rigorous peer-review process since the inception of the journal, but that process has been refined and strengthened in recent years. So I’ll take this opportunity to explain how we now conduct peer review at Tobacco Control.

The most significant change is the use of an editorial committee which makes consensus decisions on papers submitted to the journal. The committee, modelled after a similar committee used by theBMJ, is often referred to internally as the “hanging committee”. That name did not originate from the macabre meaning of the word “hang”; rather, it is borrowed from the Royal Academy of Arts in London, which has used a “hanging committee” to decide which paintings to hang on its walls (personal communication from Dr Richard Smith, editor of BMJ).

Before 1997, I myself made interim and final decisions on papers—whether to accept or reject them, or to send them back to authors for revisions. Those decisions were aided by comments from peer reviewers and recommendations made by senior and associate editors. Beginning with papers submitted in January 1997, editorial decisions have been made by our three-member hanging committee, consisting of myself and our two senior editors—Drs Ross Brownson and K Michael Cummings. Dr Brownson is professor and chair of the Department of Community Medicine at Saint Louis University’s School of Public Health. Dr Cummings is chair of the Department of Cancer Control and Epidemiology at Roswell Park Cancer Institute in Buffalo, New York. The hanging committee meets weekly by conference call.

When papers arrive at our editorial office, they are assigned to Dr Brownson or Dr Cummings on an alternating basis, so that one of them takes the lead on each paper. At the next conference call, the lead senior editor makes a recommendation to the committee as to whether a new paper should be rejected outright or sent for external peer review. If the paper is designated for peer review, the committee identifies referees to whom the paper will be sent. Generally we send papers to three peer reviewers, so that if one reviewer is unable or unwilling to review it, we are left with two others who can. If a paper has substantive data analysis, it is also sent to one of our three consulting editors for methods and statistics: Dr Seth Emont, a senior programme officer at the Robert Wood Johnson Foundation; Dr Todd Rogers, who until recently was a senior research scientist (for more than a decade) at Stanford University’s Center for Research in Disease Prevention; and Dr Michael Siegel, an assistant professor at Boston University’s School of Public Health.

Tobacco Control has 12 associate editors, each of whom covers a distinct area in tobacco control for the journal. The associate editors, and the areas they handle, are shown on the inside front cover of the journal. If the subject matter addressed by a paper falls within the purview of an associate editor, we will often request assistance from that individual. In that circumstance, the associate editor will typically nominate peer reviewers for the paper, and will make recommendations to the hanging committee concerning the disposition of the paper (accept, reject, revise) as it goes through the various stages of editorial and peer review. In some cases the associate editor carries on correspondence with the author.

When peer review is completed on a paper, or when a revised paper is received from an author, it is placed on the agenda of the next conference call. Relevant materials for each manuscript on the agenda are distributed in advance to each of the three members of the committee. During the call, the senior editor who has the lead on a particular paper summarises the content of the manuscript; the feedback from those who reviewed the paper (outside referees, the consulting editor for methods and statistics, and the associate editor); and in the case of a revised paper, the changes made to the earlier version. After discussion, the committee reaches a consensus decision, representing interim or final action on the paper. With very few exceptions, decisions of the hanging committee are unanimous.

When we request that authors revise their paper, we provide them with “blinded” copies of peer review comments to guide them in their revisions. When we reject manuscripts, we also send authors the comments from peer reviewers, to help the authors improve their papers should they wish to resubmit them to another journal.

In addition, we usually share all the peer review comments (with identifiers removed) and our letters to authors with each of the reviewers. This gives them feedback on editorial decisions regarding the paper and allows them to compare their own judgments on the manuscript with those of their peers.

The hanging committee approach yields several benefits. First, three editors make joint decisions based on careful review and discussion; thus, decisions are more likely to be well-informed, fair to the authors, and free from any bias that one particular editor might bring to the table. Second, the approach distributes the workload among three individuals, which allows papers to be handled more efficiently. Third, a weekly meeting agenda ensures that the process moves forward continuously. In a future editorial, we will report trend data on how quickly decisions are made on manuscripts, how long it takes for accepted papers to be published, the number of papers submitted, and the percentage that are accepted for publication.

As we have noted previously,1 the peer review process is far from perfect, but it is the best form of quality control we have in “journalology”. We will continue to strive to improve the process, and we welcome comments on how to do so from our readers and contributors.