THE FELLOWS’ CORNER

Manuscript writing Manuscript writing is something that we almost certainly do not receive any formal training in during fellowship. To adequately showcase our research findings, however, publication of a well-written manuscript is essential. This month in the Fellows’ Corner, Dr Hashem El-Serag provides some helpful tips in the manuscript writing process. Jonathan M. Buscaglia, MD Fellows’ Corner Editor Therapeutic Endoscopy Fellow Johns Hopkins Hospital Baltimore, Maryland, USA Scientific writing is essential to disseminate research findings. Publications are also a measure of academic productivity that is assessed for promotion and competitive grant application. Importantly, you will really get to know a field only if you contribute to it.

HOW TO WRITE A MANUSCRIPT Overview It is recommended to assemble all vital study materials, including protocols, final analyses, and references before proceeding with writing. It is possible to get a head start and reduce the ‘‘blank page panic’’ by starting to write the Methods and possibly the Introduction sections; however, the final analysis should precede writing the Results and Discussion sections. The stop (for more analysis) and go (write a couple of more lines) style of writing is inefficient, as well as frustrating. Begin by constructing tables and figures, because these will be useful for developing an outline

Key Points d

OBSTACLES TO WRITING AND POTENTIAL SOLUTIONS It is estimated that it takes a ‘‘quick’’ clinical study an average of 2 years from conception to publication. You therefore should plan your goals accordingly. Several barriers interfere with accomplishing the goals set for scientific writing. External barriers include teaching responsibilities, committee assignments, personnel disputes, and grant deadlines. Intrinsic barriers include procrastination, perfectionism, and/or insecurity; these manifest as an inability to start or complete manuscripts. It is important to develop the ability for spontaneous writing, creating/adhering to self-imposed timelines, and to write efficiently. Regular, preferably daily, writing is likely to teach the ability to write without feeling ‘‘ready’’ or awaiting inspiration. One way is to make writing contingent on doing a recurrent daily activity. For example, schedule writing time before recurrent activities such as eating, taking a shower, or going home to have dinner with the family. Efficient writers typically use several methods of conserving sentences for future use, including working from existing templates of previous manuscripts, focusing on closely related areas of content and methods, or dictating to generate a speedy first draft.

d

d

d

Assemble all vital study materials before you begin the writing process. The number and order of authors should be agreed upon as early as possible. Make an effort to try to write regularly, preferably daily. Follow the basic format of abstract, background, methods, results, and discussion.

with major and minor points in each section. Then, write a first draft, which is crucial in moving the writing process forward, circulate it to coauthors, revise it, and read it again. Revise for content first and for fluency, clarity, accuracy, and grace next. See Table 1 for a general outline of manuscript sections.

Authorship The number and order of authors should be discussed and agreed upon as early as possible. Ideally, this should be discussed before writing begins, but, at a minimum, the author list should be included on the front page of the first draft.

Title Copyright ª 2008 by the American Society for Gastrointestinal Endoscopy 0016-5107/$32.00 doi:10.1016/j.gie.2007.07.023

The title should be informative and specific to the study, and should contain key elements that advertise the paper’s contents. Declarative titles (eg, Racial disparities exist in . . .)

www.giejournal.org

Volume 67, No. 2 : 2008 GASTROINTESTINAL ENDOSCOPY 311

Manuscript writing

TABLE 1. Outline of sections in a scientific paper 1. Title page 2. Conflict of interest notification page 3. Abstract and key words

El-Serag

process, and methods used for analyses. Although few readers are interested in this section, it is the most thoroughly inspected section by reviewers. For specific types of clinical and epidemiologic studies, following these guidelines is helpful in providing an ordered and complete description of the Methods. The technical aspects should be written or at least edited by those with expertise.

4. Introduction 5. Methods a. Selection and description of participants b. Technical information c. Statistics 6. Results 7. Discussion 8. References

Results This section is the heart of the paper. The text explains the tables and is aided by the figures. You should provide only enough interpretation to lead the reader from one analysis to the next. Avoid describing lengthy, irrelevant analyses and comparing study findings to the work of others. Make every attempt to create ‘‘stand alone’’ tables and figures by including details of the study sample and methods in the legends.

a. General considerations related to references b. Reference style and format 9. Tables 10. Illustrations (figures) 11. Legends for illustrations (figures) 12. Units of measurement 13. Abbreviations and symbols

are preferred to inquisitive titles (eg, Is there a race-based difference in . . . ?).

Abstract The abstract section is typically written last when it is easiest to summarize all aspects of the study. Although the abstract may not be crucial for the initial editorial decision to review the manuscript, it is the only part of the manuscript read by the vast majority of readers. The Methods and Results sections of the abstract should provide facts about the number of participants, the duration of followup, and major findings. The Conclusions section of the abstract should answer the study aims presented in the Background/Aims section.

Discussion This section should provide a critical argument that covers the following questions: Why did you start? What did you do? What answer did you get? What is your finding, evidence, counterevidence, and what does it mean anyway? I recommend starting the discussion by summarizing the study and its findings. Next, explain how the findings relate to existing literature. Avoid sweeping statements that describe your study as ‘‘first, only, best’’; it is unlikely to be completely true, and it is likely to alienate reviewers and editors. Be thoughtful and tactful in explaining differences between your study and previous studies; appropriate terms include ‘‘conflicting,’’ ‘‘inconsistent,’’ and having ‘‘limited generalizability,’’ as possibilities for differences.

References They should be cited to support methods and claims. Be generous and thorough, but not exhaustive, in acknowledging previous work. References are one of the first sections read by editors and reviewers. The use of specialized reference software, such as Refman and End Note (Thompson Corp, Carlsbad, Calif ) is recommended.

Background The major function of this section is to explain the rationale for the study. The introduction should briefly cover the magnitude/importance of the problem and highlight gaps in knowledge, which will be addressed by the current study. This section should not read like a comprehensive review article and thus probably should not exceed 2 doublespaced pages.

DISCLOSURE The authors report that there are no disclosures relevant to this publication.

This section should contain details of the selection and description of study participants, the data collection

Hashem B. El-Serag, MD, MPH Associate Professor of Medicine Michael E. DeBakey VA Medical Center Baylor College of Medicine Houston, Texas, USA

312 GASTROINTESTINAL ENDOSCOPY Volume 67, No. 2 : 2008

www.giejournal.org

Methods