BMJ  2005;330:305-307 (5 February), doi:10.1136/bmj.330.7486.305

Education and debate

Submission to multiple journals: a method of reducing time to publication?

¹ York Trials Unit, Department of Health Sciences, University of York, York YO10 5DD

Getting a manuscript accepted by a journal can be a long, drawn out process and delays dissemination of clinical research. Allowing authors to submit to several journals simultaneously could speed up publication

Introduction

Most medical journals do not allow simultaneous submission of articles to more than one journal. The need for sequential submission is an important factor in delaying the publication of research. We propose that journals should allow authors to submit to two or more journals at the same time. This would lead to greater competition among journals and shorten publication delay, which would benefit both patients and authors.

Publishing delays

Timely publication of research findings is crucial because delays will have a harmful effect on patients' health. In a review of AIDS trials, Ioannidis found a delay of between 1.7 and 3 years between study completion and publication, with negative trials taking significantly longer to be published.¹ Furthermore, a study looking at economic evaluations found that on average the economic results were published two years after the clinical results.²

Morally, as well as ethically, all those involved in the research process have a duty to report their findings as quickly as possible. An important barrier to early publication and dissemination is often the researcher. Many researchers take too long to write up their findings. However, another big factor is sequential submission, whereby authors are allowed to submit to only one journal at a time.

Process of publication

Authors intending to submit a manuscript that they consider to be of high quality and general appeal may consider a general medical journal (BMJ, JAMA, Lancet, New England Journal of Medicine, etc). These journals have a fairly rapid turnaround. Even so, unless the journal considers the paper for its "rapid" publication section, a decision usually takes six to eight weeks. If the decision is positive (usually subject to amendment), the study is then published within a few months of the final manuscript being received.

However, many papers are not accepted by the first journal and resubmission to a second or third journal is required. Ironically, in our experience, the most interesting and methodologically sound papers are often delayed the most as these are usually more likely to be sent to highly cited and competitive journals. On the other hand, many authors overestimate the value of their work and aim too high and therefore contribute to the delay in publication. Nevertheless, it is not uncommon for a paper to be rejected two or three times before it is finally accepted. Indeed, major general journals reject most of the papers they receive.

Because our anecdotal experience was that such delays were widespread we undertook a small survey of corresponding authors of randomised controlled trials. We searched Web of Science with the phrase "randomised controlled trial" for a single month (January 2004). We emailed 95 corresponding authors asking how many times they had submitted their manuscript before it was accepted. Of the 40 who replied, about half (18/40) had submitted the paper to two or more journals, and for a quarter of those the time to publication was 20 months or more (compared with about 12 months for those who published in their first journal). This delay will not be entirely the fault of journals: some of the authors will have been inefficient in resubmitting their manuscripts. Nevertheless, a large proportion will be due to the requirement of journals that papers are submitted to one journal at a time.

Sequential submission

Journals require sequential submission for several reasons, one of which may be to reduce competitive forces between journals. Altman described various methods journals have used to maintain an advantage in order to increase circulation and profits.^{3 4} Sequential submission can be seen in this context. During the time an article is under submission to a journal it cannot be sent elsewhere for possible publication; in effect, the journal is holding a temporary monopoly on the paper.

If the paper is rejected, the journal will suffer some loss in terms of the time and costs of peer and editorial review. Researchers and research consumers also lose from rejection. Delays in publication can adversely affect researchers' careers (publish or perish) and their institutions' financial status, as distribution of public research funding in the United Kingdom is partly decided by a research assessment exercise. Research consumers (patients, doctors, and policy makers) lose out because the results of effective or ineffective treatments remain unknown. One method suggested some years ago that could address this problem is multiple submission.⁴

Multiple submission

Multiple submission is different from duplicate publication in that if the same manuscript were sent to two different journals and was accepted by one, the submission would then be withdrawn from the other. Multiple submission has a potential benefit of reducing the delay incurred by sequential submission. For example, an author might submit a paper to a general journal and a specialist publication. If the general journal accepted the paper for publication, it could be withdrawn from the specialist journal. To avoid encouraging duplicate publication, journals adopting a multiple submission policy could insist that authors inform them of the other journals that the article has been sent to. Acceptance or rejection letters could be sent not only to the authors but also to the other journals. This would prevent duplicate publication and also stop authors from waiting until they received a better offer from another journal.

As well as speeding up publication of important findings, such a system would also lead to competition between journals for the best articles. Awareness that competing journals were also looking at the paper would provide a strong incentive to rapidly peer review the manuscript and make a final decision. This effect could increase the speed of peer review and publication. Indeed, early entrants to such a competition would, in our view, benefit from receiving higher quality submissions. For example, if the Lancet and BMJ were the only general journals to allow multiple submission, many authors would be tempted to prioritise those two journals before other general journals.

Summary points Most journals do not allow authors to submit a manuscript to another journal simultaneously Most manuscripts are sent to two or three journals before acceptance Simultaneous submission could reduce the delay in publication and introduce competition between journals Safeguards could be put in place to avoid duplicate publication

Although journals could compete in such a system, they might also collaborate. For example, the BMJ might collaborate with the British Journal of Obstetrics and Gynaecology. Authors with a manuscript on obstetrics that could appeal to a general medical readership would submit the paper to both journals simultaneously. If the BMJ thought that the paper was not of sufficient interest to a general audience then no time would be lost in the paper being considered by a prestigious specialist journal.

Discussion

Multiple or simultaneous submission could introduce valuable competitive forces among journals for the best manuscripts. Multiple submission is allowed in some specialties. Piron compared his experience of writing and submitting papers to economics, finance, maths, and psychology journals, which do not allow multiple submissions, with law review journals, which do.⁵ He noted that law review journals in the United States had the "fastest turnaround times of any set of journals on the planet."⁵

Journals may have other reasons than preventing competition for not allowing multiple submission. Multiple submission would increase the workload of journal staff through the increased flow of manuscripts. For some journals, the extra administrative burden would not be worth while as it may slow down their decision making processes and allow a competitor to "scoop" the article. This would leave them with the sunk costs of mailing the paper to reviewers etc, without having had the chance of publishing the paper.

Workload would also increase for researchers as they will be asked to review more papers. This is a burden some of us would bear to increase turnaround. Indeed, as a condition of allowing multiple submission, journals could set the condition that one or more of the authors of the submitted paper would agree to review one of the journal's other recent submissions.

Workload for both journals and reviewers would fall, however, if a collaborative model of multiple submission was adopted. In this model, a journal would allow multiple submission on the condition that the paper went to a partner journal. Both journals could then share the reviewer's reports and one journal's staff could handle the administration.

Some journals use an author pays system—for example, BioMed Central journals. It is feasible that journals could have a single (free) submission policy or a multiple (pay) approach. This would allow the journal to offset some of its increased costs from losing an article to a rival but it would also depress the demand for the service. However, this approach might be less than ideal given that some organisations can better afford to pay than others. Additionally, authors may be more likely to pay if they have positive findings than negative results. Allowing papers with negative findings to be submitted at no charge might offset this problem.

Several models of multiple submission exist. Journals could adopt any of these, and they might even experiment with different models using a randomised trial.

---

Contributors and sources: DT and JA originally discussed the idea for the paper. SC, JD, and EP made suggestions for the paper. SC, JD, and EP identified a sample of papers and developed a short email questionnaire to corresponding authors. All authors emailed to a subset of corresponding authors. JD collated and analysed results for survey. DT will act as guarantor.

Competing interests: All the authors have a career interest in having their manuscripts published as soon as possible.

References

1. Ioannidis JPA. Effect of the statistical significance of results on the time to completion and publication of randomized efficacy trials. JAMA 1998;279: 281-6.[Abstract/Free Full Text]
2. Greenberg D, Rosen AB, Olchanski NV, Stone PW, Nadai J, Neumann PJ. Delays in publication of cost utility analyses conducted alongside clinical trials: registry analysis. BMJ 2004;328: 1536-7.[Free Full Text]
3. Altman LK. The Ingelfinger rule, embargoes, and journal peer review: part 1. Lancet 1996;347: 1382-6.[CrossRef][ISI][Medline]
4. Altman LK. The Ingelfinger rule, embargoes, and journal peer review: part 2. Lancet 1996;347: 1459-63.[CrossRef][ISI][Medline]
5. Piron R. They have the world on a queue. Challenge 2001;Sep. www.findarticles.com/cf_0/m1093/5_44/79151031/p1/article.jhtml?term= (accessed 13 Sep 2004).

(Accepted 10 August 2004)

CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    What's this?

Relevant Articles

Submission to multiple journals to reduce publication times: Idea needs further evaluation

Prabhat Kumar Sinha
BMJ 2005 330: 602-603. [Extract] [Full Text]

Submission to multiple journals to reduce publication times: Access to information might become truly universal

Adrian S Blaj and Lucian Alexandrescu
BMJ 2005 330: 603. [Extract] [Full Text]

Submission to multiple journals to reduce publication times: What about the readers?

Harvey Marcovitch
BMJ 2005 330: 603. [Extract] [Full Text]

Submission to multiple journals to reduce publication times: Summary of responses

Birte Twisselmann
BMJ 2005 330: 603. [Extract] [Full Text]

Would multiple submission reduce the time to publication?
BMJ 2005 330: 0. [Full Text]

Questions and answers

Kamran Abbasi
BMJ 2005 330: 0. [Extract] [Full Text] [PDF]

This article has been cited by other articles:

• Yu, G., Li, Y.-J. (2006). A scientometric assessment of the information loss caused by the delays in publication of Chinese science journals. Journal of Information Science 32: 78-87 [Abstract]  
• Blaj, A. S, Alexandrescu, L. (2005). Submission to multiple journals to reduce publication times: Access to information might become truly universal. BMJ 330: 603-603 [Full text]  
• Sinha, P. K. (2005). Submission to multiple journals to reduce publication times: Idea needs further evaluation. BMJ 330: 602-603 [Full text]  
• Marcovitch, H. (2005). Submission to multiple journals to reduce publication times: What about the readers?. BMJ 330: 603-603 [Full text]  

Rapid Responses:

Read all Rapid Responses

Some other issues that delay the publication of research

Wen-Hung Kuo
bmj.com, 4 Feb 2005 [Full text]

article submitted to multiple journals- If it happens as below...who is responsible???????

akhtar sherin
bmj.com, 7 Feb 2005 [Full text]

Competition is a good antidote against sleep

Giovanni de Girolamo
bmj.com, 4 Feb 2005 [Full text]

Killing peer review

William M. Tierney
bmj.com, 4 Feb 2005 [Full text]

Multiple submissions may lead to unethical pressures

Scott Kinkade
bmj.com, 4 Feb 2005 [Full text]

Iconoclastic thoughts

Adrian S. Blaj
bmj.com, 4 Feb 2005 [Full text]

The Journal Article "Match"

Alexander C. Tsai
bmj.com, 6 Feb 2005 [Full text]

A welcome suggestion

Ramaswamy. Subramanian, et al.
bmj.com, 6 Feb 2005 [Full text]

Multi-submissions on the same reviewer's desk

Jutta Loeffler
bmj.com, 6 Feb 2005 [Full text]

Mutual responsibility

S Anuradha
bmj.com, 6 Feb 2005 [Full text]

Multiple submissions - Needs further evaluation

Prabhat Kumar Sinha, et al.
bmj.com, 6 Feb 2005 [Full text]

Addendum Re: Iconoclastic thoughts

Adrian S Blaj, et al.
bmj.com, 7 Feb 2005 [Full text]

Another drawback of multiple submission

E Suresh
bmj.com, 7 Feb 2005 [Full text]

Re: Multiple submissions - Needs further evaluation

Robert I. Rudolph, M.D., FACP
bmj.com, 7 Feb 2005 [Full text]

Submission to multiple journals: a method of reducing time to publication?

waqas khan
bmj.com, 7 Feb 2005 [Full text]

The crux of the problem

Jerry George Mathew
bmj.com, 7 Feb 2005 [Full text]

Speeding up acceptance time

Sara Black
bmj.com, 8 Feb 2005 [Full text]

What about the reader?

Harvey Marcovitch
bmj.com, 8 Feb 2005 [Full text]

How authors can help

Tim Albert
bmj.com, 8 Feb 2005 [Full text]

Facilitating Publication

Vance W Berger
bmj.com, 10 Feb 2005 [Full text]

Submission to multiple journals: a method of reducing time to publication?

Dr.Satheesha Nayak
bmj.com, 17 Feb 2005 [Full text]

Preliminary enquiry before submission could be useful

Bolajoko O. Olusanya
bmj.com, 17 Feb 2005 [Full text]

Multiple submission or group submission?

Peter D Macdonald
bmj.com, 11 Mar 2005 [Full text]

Suggestion for submission for multiple journals

DENISE RAZZOUK
bmj.com, 15 Mar 2005 [Full text]

Submission to multiple journals: what happens if there are multiple acceptances

Marge Berer
bmj.com, 13 May 2005 [Full text]