how to get your paper published

5
S3 article with strong epidemiologic interest may be less well received if it is sent to a journal in a country where that particular problem is of little concern. There are many journals, and the review process varies greatly from one to another. (The process for The Journal of Thoracic and Cardiovascular Surgery is described in greater detail later). You may want to have colleagues objectively assess your work and provide advice and criticism of its relative importance in rela- tion to articles generally published in the same field. If your priority is early publication, it is important to choose a journal with rapid review and publication cycles. On the other hand, the work may be of great and possibly enduring importance and involve complex investigations that require rigorous review to ensure universal acceptance of the conclusions made by the authors. Instructions to authors can pose limitations that influ- ence the choice of journal. A complex study with abun- dant data and figures simply may not be condensed enough to meet space limitations. Color photomicro- graphs may be critical to understanding the experimen- tal data but may not be allowed by a particular journal. The work may be such that it is enhanced by a journal that allows an electronic link to a website that allows review of a video. Prolific authors generate many articles of varying quality, importance, style, and emphasis over the course of their careers. The strength of having many excellent peer-reviewed journals is the opportunity for publica- tion of a broad range of articles, each ideally matched to the particular style and nuance of that journal. Many authors have “favorite” journals to which they direct their “best” work but use many journals to communi- cate their experimental methods and findings. Deciding which journal to use is critical in the strategy for having a manuscript published. Writing and assembling the manuscript Once you have decided where to submit your work, you need to consider the myriad details that go into writing, assembling, and sending your manuscript. An extremely helpful resource for anyone writing a scientific paper is the “Uniform Requirements for Manuscripts Submitted to P ublication of the results of an original scientific study entails producing a manuscript with content that reflects a well-designed, well-executed research question. It is a well-written description of the process, the results, and the wide implications. Publication also depends on one’s ability to choose the most appropriate journal—the one with the best match in emphasis between the other articles in the journal and the one being submitted. Selecting the appropriate journal Most investigators are eager to have their work pub- lished, and choosing the most appropriate journal for submission is a critical step in the process. Although there is no “standard approach,” the choice of journal in many instances should precede writing the manuscript. Each journal has a set of instructions for authors con- cerning the topics suitable for that journal and the types of papers that may be submitted, such as original arti- cles, reviews, and brief communications. Although it seems as if the choice of journal should be last, we believe it should be first. The process begins with careful consideration of the subject matter. There must be a match between the emphasis of other articles in the journal and the one being submitted. How much does the article emphasize basic science or clinical science? Is the approach taken usually organ specific, tissue specific, system oriented, technique driven, observational, or experimental? Who is likely to read the article? Are they whom your article is likely to interest? Choose the appropriate spe- cialty and approach to that specialty. Readers of a car- diology journal are more likely to be interested in results from cohorts of patients than in surgical tech- nique. Some journals are more focused on a local or regional audience than on an international audience. An Getting published depends heavily on starting with a well-designed, well- executed research question that is accurately described, and submitting it to a jour- nal with an appropriate audience. (J Thorac Cardiovasc Surg 2001;121:S3-7) Pamela W. Fried, MBA Andrew S. Wechsler, MD HOW TO GET YOUR PAPER PUBLISHED From MCP Hahnemann University, Philadelphia, Pa. Read on April 29, 2000, at the Eightieth Annual Meeting of The American Association for Thoracic Surgery, Toronto, Ontario, Canada. Copyright © 2001 by The American Association for Thoracic Surgery 0022-5223/2001 $35.00 + 0 12/0/114493 doi:10.1067/mtc.2001.114493

Upload: pamela-w-fried

Post on 21-Sep-2016

221 views

Category:

Documents


1 download

TRANSCRIPT

Page 1: How to get your paper published

S3

article with strong epidemiologic interest may be lesswell received if it is sent to a journal in a country wherethat particular problem is of little concern.

There are many journals, and the review processvaries greatly from one to another. (The process for TheJournal of Thoracic and Cardiovascular Surgery isdescribed in greater detail later). You may want to havecolleagues objectively assess your work and provideadvice and criticism of its relative importance in rela-tion to articles generally published in the same field. Ifyour priority is early publication, it is important tochoose a journal with rapid review and publicationcycles. On the other hand, the work may be of great andpossibly enduring importance and involve complexinvestigations that require rigorous review to ensureuniversal acceptance of the conclusions made by theauthors.

Instructions to authors can pose limitations that influ-ence the choice of journal. A complex study with abun-dant data and figures simply may not be condensedenough to meet space limitations. Color photomicro-graphs may be critical to understanding the experimen-tal data but may not be allowed by a particular journal.The work may be such that it is enhanced by a journalthat allows an electronic link to a website that allowsreview of a video.

Prolific authors generate many articles of varyingquality, importance, style, and emphasis over the courseof their careers. The strength of having many excellentpeer-reviewed journals is the opportunity for publica-tion of a broad range of articles, each ideally matchedto the particular style and nuance of that journal. Manyauthors have “favorite” journals to which they directtheir “best” work but use many journals to communi-cate their experimental methods and findings. Decidingwhich journal to use is critical in the strategy for havinga manuscript published.

Writing and assembling the manuscriptOnce you have decided where to submit your work, you

need to consider the myriad details that go into writing,assembling, and sending your manuscript. An extremelyhelpful resource for anyone writing a scientific paper isthe “Uniform Requirements for Manuscripts Submitted to

P ublication of the results of an original scientificstudy entails producing a manuscript with content

that reflects a well-designed, well-executed researchquestion. It is a well-written description of the process,the results, and the wide implications. Publication alsodepends on one’s ability to choose the most appropriatejournal—the one with the best match in emphasisbetween the other articles in the journal and the onebeing submitted.

Selecting the appropriate journalMost investigators are eager to have their work pub-

lished, and choosing the most appropriate journal forsubmission is a critical step in the process. Althoughthere is no “standard approach,” the choice of journal inmany instances should precede writing the manuscript.Each journal has a set of instructions for authors con-cerning the topics suitable for that journal and the typesof papers that may be submitted, such as original arti-cles, reviews, and brief communications. Although itseems as if the choice of journal should be last, webelieve it should be first.

The process begins with careful consideration of thesubject matter. There must be a match between theemphasis of other articles in the journal and the onebeing submitted. How much does the article emphasizebasic science or clinical science? Is the approach takenusually organ specific, tissue specific, system oriented,technique driven, observational, or experimental?

Who is likely to read the article? Are they whom yourarticle is likely to interest? Choose the appropriate spe-cialty and approach to that specialty. Readers of a car-diology journal are more likely to be interested inresults from cohorts of patients than in surgical tech-nique. Some journals are more focused on a local orregional audience than on an international audience. An

Getting published depends heavily on starting with a well-designed, well-executed research question that is accurately described, and submitting it to a jour-nal with an appropriate audience. (J Thorac Cardiovasc Surg 2001;121:S3-7)

Pamela W. Fried, MBAAndrew S. Wechsler, MD

HOW TO GET YOUR PAPER PUBLISHED

From MCP Hahnemann University, Philadelphia, Pa.Read on April 29, 2000, at the Eightieth Annual Meeting of The

American Association for Thoracic Surgery, Toronto, Ontario,Canada.

Copyright © 2001 by The American Association for ThoracicSurgery

0022-5223/2001 $35.00 + 0 12/0/114493doi:10.1067/mtc.2001.114493

Page 2: How to get your paper published

Biomedical Journals” (UR)(available online at http://www.acponline.org/journals/annals/01jan97/unifreqr.htm)or in reprint form (American College of PhysiciansCustomer Service Department, phone 800-523-1546, ext2600; fax 215-351-2799). Written initially in 1979 by asmall group of medical journal editors, UR providesinstructions to authors on how to prepare manuscripts forsubmission to participating medical journals, which nownumber in the thousands. Editors of participating journalsdo not refuse because of style manuscripts preparedaccording to UR criteria.

The most recent version of UR covers issues to con-sider before submitting a manuscript, such as redun-dant or duplicate publication, acceptable secondarypublication, and protection of patients’ rights to priva-cy. It provides detailed guidelines for preparation andsubmission of manuscripts and statements on the fol-lowing issues of relevance to authors and publishers:

• Definition of a peer-reviewed journal• Editorial freedom and integrity• Conflict of interest• Project-specific industry support for research• Corrections, retractions, and expressions of con-

cern about research findings• Confidentiality• Medical journals and the popular media• Advertising• The role of the correspondence column• Competing manuscripts based on the same study

Authors writing an article are responsible for beingclear, consistent, and correct. Busy readers balk at hav-ing to wind their way through a maze of complicatedprose, inconsistent thoughts, and incorrect analysis.Although many pairs of eyes see each manuscript, fromthe journal editor to the copy editor, no one knows thesubject matter more intimately than does the author.The correctness of the information is the author’sresponsibility.

An author preparing a comprehensive manuscript isresponsible for the elements in the following summary.The topics are discussed in depth in the resources list-ed in Selected Readings.

1. Title and abstract. Considerable thought shouldgo into the title and abstract because they providereviewers and readers with their first impression of apaper. They are the first elements read by reviewers;they are often the only parts of a manuscript accessibleelectronically (via MEDLINE), and they are the first,and possibly only, elements read by readers. The titleshould be concise but informative; it can be either adeclarative or an interrogative statement. The initial

words should represent the most important aspect ofthe article, followed by other terms in descending orderof importance. The abstract may be structured (notmore than 250 words) or unstructured (not more than150 words), depending on the type of article and spe-cific journal requirements. The abstract should explainthe objective of the study and what was done, found,and concluded.

2. Text. Medical journal articles have the followingpredictable pattern:

• Introduction: a description of the general problemfollowed by a statement of the specific problemand the motivation for the study

• Methods: what was done• Results: what was learned• Discussion: how the results fit into the larger pic-

ture, the limitations of the study, and comments on“what next?”

3. Figures. Clear, succinct graphs, charts, pho-tographs, and drawings can quickly convey the prima-ry findings of research. This is greatly appreciated bybusy physicians and scientists. Check with the journalto which you are submitting your article to determinethe specific technical requirements, especially for on-line submission. The following are guidelines:

• Submit photographic prints, laser-quality prints, orelectronic files. When submitting electronic files,use graphics software (eg, Photoshop, Illustrator,Freehand). Do not use presentation software (eg,PowerPoint, CorelDraw, Harvard Graphics) orword-processing software (Microsoft Word;WordPerfect) for illustrations.

• Avoid direct juxtaposition of patterned bars in bargraphs; this can cause a moiré, or shimmering,effect, which is jarring to the reader.

• Avoid the use of gray filler in bar graphs becausegray does not reproduce well.

• Minimize the use of black.• Identify figures on the back with a soft pencil (ball-

point pens cause indentations on the surface of thephotograph that are difficult to hide during repro-duction. Include author’s name, the figure number,and the word top.

• Avoid the use of paperclips with photographs.Paperclips cause surface indentations.

4. Tables. Tables should be self-explanatory and self-contained. Textual description should summarize andhighlight table content. Tables can be used to displayprecise numeric values (figures are better for conveyingtrends or proportions), to present a large number of

S4 Fried and Wechsler The Journal of Thoracic andCardiovascular Surgery

April 2001

Page 3: How to get your paper published

numeric values in compact form, to summarize infor-mation, to explain variables, and to present the wordingof survey questions, to name a few options.

5. Permission. To avoid breaking copyright laws, theauthor needs to obtain permission to reproduce materi-al from a copyrighted source. For example, to repro-duce a figure, table, or a large amount of text, the writermust obtain permission from the original copyrightholder, cite the original source, and include therequired credit line. An author citing another author’sdata or original idea should name the original source. Ifa copyright holder charges a fee to reproduce the mate-rial, the author usually is responsible for paying the fee.

6. References. The author is responsible for the cor-rectness and completeness of all citations. According toUR, references are numbered in order of first mention.The textual reference should be Arabic numbers inparentheses. It is always best to limit references to themost pertinent. An excessive number of references fre-quently leads to a request from the journal editor to cutentries from and renumber the list, which can betedious. Personal communications should be cited inthe text only, not in the list of references, and shouldinclude the name of the person and the date of the com-munication. The author should obtain written permis-sion and confirmation of accuracy from the source ofthe personal communication. The sources cited inSelected Readings contain extensive sections on how tocite references.

The review processOnce a manuscript arrives in the editorial office, it

begins its journey through peer review. The followingare the procedures of the Journal of Thoracic andCardiovascular Surgery. We believe our process is rig-orous, ensures quality, and provides authors with infor-mation that enhances current and future manuscripts.

When an article is received, an associate editorassigns at least three reviewers known to be experts inthe subject of the study. Each reviewer receives theentire article, including the text, legends, figures, andtables. The editorial office checks adherence to therequirements of the journal’s instructions to authors.Each reviewer is asked to read the manuscript in detailand to comment on it paying particular attention to thefollowing:

• Importance of the hypothesis• Reliability of the results• Appropriateness of the methods• Validity of the statistics used• Relevance of the discussion

The Journal of Thoracic andCardiovascular SurgeryVolume 121, Number 4

Fried and Wechsler S5

• Reasonableness of the conclusions• Clarity of the abstract• Length of the article• Relevance of the work in general• Priority the work should receive for publication

In some instances the reviewer may recommend pro-fessional statistical review, raise ethical concerns, orquestion duplicate publication by the authors. Thereviewers are particularly sensitive to failure of authorsto disclose relationships with industry or the use ofindustrial support for investigative work. The authormust make industrial relationships clear in the disclo-sure form provided.

The reviewer recommends acceptance withoutchange, need for revision with subsequent review, needfor revision without additional review, or rejection. Thereviewer writes comments for transmission to theauthors and comments for transmission to the editor.When all reviews are completed, the editor studies themanuscript and the reviews. Consultation with the sub-specialty-specific associate editor frequently is needed.A joint decision is reached, and the information is trans-mitted to the authors. This process ideally takes about 1month. Reality occasionally falls short of the ideal, how-ever, given the many conflicting demands on the review-er’s time. If revision is requested, the authors are askedto respond within 2 weeks. The process goes on fromthere, resulting in acceptance or rejection. We use mailand fax for this process. Conversion to a Web-basedelectronic method should be complete within the nextyear. This eventually will shorten the review period.

Revision is critical and is the greatest source of delayin publication. Providing revised manuscripts toreviewers and editors while the work is fresh in theauthor’s mind is of great benefit. Authors should re-spond in a clearly identifiable manner. The revisedmanuscript should be marked to show every additionand deletion so there is never a need for the reviewer torevert to the original manuscript. A cover letter shouldaddress every suggestion the reviewers have made andidentify each change made or defend why a change wasnot made. Above all, authors should recognize that thegoal of the reviewer is to enhance the quality of themanuscript and the journal. They should not takereviewers’ comments personally.

Authors have rights and responsibilities that can aug-ment the likelihood of publication of their work.Sometimes there is keen competition between inves-tigative groups, and reviews may reflect that competi-tion as much as they do critical assessment of the man-uscript. An author can influence this process by

Page 4: How to get your paper published

suggesting to the editor, at the time of manuscript sub-mission, that certain persons not be used as reviewers.Most editors respect such requests. Authors of workwith a narrow focus may help review by suggesting thenames of several possible reviewers. Editors usuallyhonor such a request for at least one of the reviewers, ifthe person is not related to the author.

PublicationOnce a manuscript is written, reviewed, and accepted

for publication, it is sent to the publisher. The publisher’sresponsibility is to copy edit, lay out, print, bind, andmail the journal. The copy editor reads the manuscriptline for line, editing for grammar, journal-specific style,clarity of language, and completeness. The copy editorensures that the manuscript adheres to the journal’sinstructions and that the author has not forgotten toinclude materials such as tables and figures. The pub-lisher sends proofs, and usually an order blank forreprints, to the author and asks that the materials bereturned in 24 to 48 hours. Some journals send manu-scripts to the publisher as soon as the article is accepted.In this case, the manuscript is typeset immediately andplaced in a queue for publication. Depending on thebacklog of articles, publication occurs months afterthe author sees the proofs. An alternative is to assign thearticle to a specific issue and typeset the manuscript forthat issue. The article appears in print soon after theauthor returns the proofs.

For society journals such as ours, the relationship withthe publisher can enhance the role and profitability of thejournal. Because of its extensive resources and distribu-tion networks, the publisher works with the society tomarket the journal to new subscribers, to sell advertising,and to develop new initiatives, such as publishing thejournal on line. The publisher also produces reprints,back issues, bound volumes, and supplements.

The peer review and publication processes aredesigned to ensure the accuracy, reliability, and acces-sibility of scientific information. Although publicationof a submitted article is never guaranteed, following theguidelines outlined herein will improve your chances.

Address for reprints: Pamela W. Fried, Managing Editor,The Journal of Thoracic and Cardiovascular Surgery, MCPHahnemann University, 245 North 15th St, Mail Stop 496,Philadelphia, PA 19102-1192.

S E L E C T E D R E A D I N G SEthical Issues

1. Guide for the care and use of laboratory animals. Washington,DC: National Academy Press; 1996 (available online atwww.nap.edu/catalog/5140.html).

S6 Fried and Wechsler The Journal of Thoracic andCardiovascular Surgery

April 2001

2. Nuremberg Code (available online at http://ecco.bsee.swin.edu.au/studes/ethics/Nuremberg.html).

3. World Medical Association Declaration of Helsinki. Re-commendations guiding physicians in biomedical researchinvolving human subjects. JAMA 1997;277:925-6 (availableonline at www.vitreoussociety.org/journal/instruct/helsinki.htm).

Peer Review1. Evans AT, McNutt RA, Fletcher SW, Fletcher RH. The charac-

teristics of peer reviewers who produce good-quality reviews. JGen Intern Med 1993;8:422-8.

2. International Congress on Peer Review in Biomedical Publication(available online at http://www.ama-assn.org/public/peer/peer-home.htm).

3. Lock S, Smith J. What do peer reviewers do? JAMA 1990;263:1341-3.

4. Yankauer A. Who are the peer reviewers and how much do theyreview? JAMA 1990;263:1338-40.

Manuscript Preparation1. Gopen GD, Swan JA. The science of scientific writing. Am Sci

78;1990:550-8.2. Haynes RB, Mulrow CD, Huth EJ, Altman DG, Gardner MJ.

More informative abstracts revisited. Ann Intern Med 1990;113:69-76.

3. Huth EJ. How to write and publish papers in the medical sci-ences. 2nd ed. Baltimore: Williams & Wilkins; 1990.

4. Instructions to Authors in the Health Sciences (available onlineat: http://www.mco.edu/lib/instr/libinsta.html). Links to websitesthat provide instructions to authors for more than 3000 journalsin the health and life sciences. All links are to primary sources,that is, to publishers or organizations with editorial responsibili-ties for the titles.

5. Iverson C, Flanagan A, Fontanarosa PB, Glass RM, Glitman P,Lantz JC, et al, editors. American Medical Association manual ofstyle. 9th ed. Baltimore: Williams & Wilkins; 1998.

6. Style Manual Committee, Council of Biology Editors. Scientificstyle and format: the CBE manual for authors, editors, and pub-lishers. 6th ed. New York: Cambridge University Press; 1994.

7. Tufte ER. The visual display of quantitative information.Cheshire, CT: Graphics Press; 1983.

8. Uniform requirements for manuscripts submitted to biomedicaljournals. Ann Intern Med 1997;126:36-47 (available online atwww.acponline.org/journals/resource/unifreqr.htm).

9. Welch HG. Preparing manuscripts for submission to medicaljournals: the paper trail. Eff Clin Pract 1999;2:131-7.

10. World Association of Medical Editors (available online athttp://www.wame.org). Extensive list of references and otherwebsites related to publication issues.

Statistics, Study Design1. Altman DG, Dore CJ. Randomisation and baseline comparisons

in clinical trials. Lancet 1990;335:149-53.2. Bailar JC III, Mosteller F. Guidelines for statistical reporting in

articles for medical journals. Ann Intern Med 1988;108:266-73.3. Begg C, Cho M, Eastwood S, Horton R, Moher D, Olkin I, et al.

Improving the quality of reporting of randomized controlled tri-als: the CONSORT statement. JAMA 1996;276:637-9.

4. Bland JM, Altman DG. Multiple significance tests: theBonferroni method. BMJ 1995;310:170.

Page 5: How to get your paper published

research: principles and quantitative methods. New York: JohnWiley & Sons; 1982.

13. Meinert CL. Clinical trials dictionary: terminology and usagerecommendations. Baltimore: Johns Hopkins University; 1996.

14. Piantadosi S. Clinical trials: a methodologic perspective. NewYork: John Wiley & Sons; 1997.

15. Sackett DL, Haynes RB, Tugwell P. Clinical epidemiology: abasic science for clinical medicine. 2nd ed. Philadelphia:Lippincott-Raven: 1991.

16. Salsburg DS. The religion of statistics as practiced in medicaljournals. Am Stat 1985;39:220-3.

17. Simon R. Confidence intervals for reporting results of clinical tri-als. Ann Intern Med 1986;105:429-35.

18. Wilkinson L. Graphical displays. Stat Methods Med Res 1992;1:3-25.

19. Yusuf S, Wittes J, Probstfield J, Tyroler HA. Analysis and inter-pretation of treatment effects in subgroups of patients in random-ized clinical trials. JAMA 1991;266:93-98.

The Journal of Thoracic andCardiovascular SurgeryVolume 121, Number 4

Fried and Wechsler S7

5. Braitman LE. Confidence intervals extract clinically useful infor-mation from data. Ann Intern Med 1988;108:296-8.

6. Braitman LE. Confidence intervals assess both clinical signifi-cance and statistical significance. Ann Intern Med 1991;114:515-7.

7. Braitman LE, Davidoff F. Predicting clinical states in individualpatients. Ann Intern Med 1996;125:406-12.

8. Concato J, Feinstein AR, Holford TR. The risk of determining riskwith multivariable models. Ann Intern Med 1993;118:201-10.

9. Everitt BS. The Cambridge dictionary of statistics in the medicalsciences. Cambridge, UK: Cambridge University Press; 1995.

10. Gardner MJ, Altman DG. Confidence intervals rather than P val-ues: estimation rather than hypothesis testing. BMJ 1986;292:746-50.

11. Gardner MJ, Machin D, Campbell MJ. Use of check lists inassessing the statistical content of medical studies. BMJ1986;292:810-2.

12. Kleinbaum DG, Kupper LL, Morgenstern H. Epidemiologic