Getting a scientific paper published in Epilepsia: An editor's perspective


  • Philip A. Schwartzkroin

    Corresponding author
    1. Department of Neurological Surgery, University of California, Davis, California, U.S.A
    • Address correspondence to Philip A. Schwartzkroin, Department of Neurological Surgery, University of California, Davis, CA, U.S.A. E-mail:

    Search for more papers by this author


Getting a paper published in Epilepsia depends first and foremost on the quality of the work reported, and on the clarity and convincingness of the presentation. Papers should focus on important and interesting topics with clearly stated objectives and goals. The observations and findings are of greatest interest when they are novel and change our views on the mechanisms and/or treatment of an epileptic disease. Studies should be carefully designed to include adequate sample size, comparison groups, and statistical analyses. Critically, the data must be clearly presented and appropriately interpreted. If followed, these recommendations will improve an author's chances of having his/her paper accepted in a high quality journal like Epilepsia.

During my tenure as co-editor-in-chief of Epilepsia, we received many inquiries about the type of article the Journal would publish–and a good number of protests from authors who were unhappy about our decision to reject their manuscripts. Every once in a while, we received a letter similar to that published in this issue of the Journal (Hu et al., 2013), that reflects what all the other questioners really wanted to know: “What do I have to do to get a paper accepted in your journal?” Sometimes this question arose out of frustration (I've done everything the reviewers suggested, but you've still rejected my manuscript). Sometimes the question stemmed from a genuine confusion about what it takes to get a manuscript accepted in a high quality international journal. Often the question came from authors (or their representatives) in parts of the world that are emerging as major scientific/research contributors. It is especially for this latter category of author that I offer the following suggestions and guidelines. Many of the general guidelines presented briefly below have been discussed in considerably more detail in previous publications (e.g., Iverson et al., 1998; Lang, 2010).

Do High Quality Work

The starting point for any successful manuscript is to have something worthwhile to say, to have generated something valuable that others will want to read about–and to have generated the report in a rigorous manner so that the presented data are believable and interpretable. That is, the manuscript should reflect “high quality” work. Although there is no single accepted definition of “high quality,” authors should remember that the editors of most journals are themselves research investigators and/or practicing clinicians (and often leaders in their fields). For that reason, these editors are likely to have acute insights into the key questions in their fields, and will certainly have rather definite opinions about what constitutes excellence in a paper submitted to their journal. Here are some questions that most editors and reviewers will ask when evaluating a manuscript:

  • 1.Does the report focus on an interesting and topical question?

If you want to publish in a well-respected journal, then you need to focus your efforts on a topic that is of current or emerging interest in your field. It is not enough to do competent work. Your study must grab the attention of the editors (and reviewers and readers). Consider why you decided to do the study in the first place. Presumably, you were drawn to a problem of current interest, to an issue that was the subject of debate among your colleagues, or to a problem that remains unsettled. Or perhaps you have made observations or gathered data that are inconsistent with current thinking about this problem. If the issue is interesting to you, it is likely to be interesting to others in your field; your job is to make sure that your report reflects the feature of your work that makes it interesting/important.

  • 2.Are the goals of the study clearly and explicitly stated?

To communicate clearly with your reader, and to help the reader understand the motivation for your research, it helps to identify your goals. Whether your research focuses on understanding the mechanisms of action of a disorder, the efficacy of a treatment, or the nature of a clinical presentation, your report should begin with telling the reader what you hope to convey in your report. Developing well-focused hypotheses and goals, and stating them clearly and succinctly, not only helps the reader understand what you're trying to convey but also helps you (as an investigator, a clinician, an author) do your work in a well-focused manner.

  • 3.Are the observations novel and insightful?

A significant part of any manuscript's appeal is in its presentation of new information. There is generally little interest is a study that shows what has been previously demonstrated or described–even if there is some minimal variation (e.g., a different drug, an untested population). Confirmation of previously reported results is certainly important, and I don't mean to discourage research aimed at such confirmation, especially if the issue is controversial. But confirmatory studies do not have the same excitement as reports of new observations. Studies that provide novel or unexpected insights–whether they are mechanistic or phenomenological–are viewed with much greater interest than studies that offer incremental data but no new way of looking at a problem. If you offer a novel way of thinking about a problem, based on new data or unexpected observations, then you not only gain the readers' attention but also potentially make significant contributions to understanding the issue on which your research focuses.

  • 4.Does the manuscript describe appropriate study groups?

In all experimental studies, a “test” group is assessed against a “comparison” group or a “control” group. It is important to identify clearly the comparisons to be made and explain the role and/or significance of each comparison group. Ideally, there is a “control” group that has not received any treatment and is “normal.” It is not always possible, of course, to include a “normal” population–especially in some clinical studies where normal or untreated individuals cannot be included for ethical reasons. Given this limitation, it is important to draw conclusions based only on the groups involved in the study.

  • 5.Is the sample number adequate, and have appropriate statistical analyses been applied?

In experimental studies, the significance of the results depends on appropriate statistical testing–which in turn revolves around (among other things) the number of subjects or samples from which the data are generated. Small sample numbers–and large variability–make it difficult to draw strong conclusions (Button et al., 2013). Statistical significance depends on whether the sample number provides adequate statistical “power.” Choice of statistical tests is determined by (among other things) the normal–or not–distribution of the data, the number of comparisons to be made, and the type of data being analyzed. All of these issues are described in more detail in any number of statistical texts and articles (e.g., Lang & Secic, 2006).

  • 6.If the study is descriptive (i.e., does not compare experimental groups), have the authors made a case for the significance of such description?

Many studies are primarily descriptive–for example, case reports. If you are preparing such a report for publication, the description should be compelling. Such reports should emphasize not only the novelty of the description, but also features that provide insight into the disorder (or problem) in question–for example, providing hints about underlying mechanisms or providing a window to better diagnosis or treatment. Certainly, some studies are “experimental” but not hypothesis-driven. These studies are often referred to today as “exploratory.” It is particularly important that the author of such a manuscript makes a strong case for the general problem area of investigation. If there's no hypothesis, then it's critical to be clear about what this new observation can add to our understanding of the problem/field.

  • 7.Are the conclusions clearly based on (tied to) results of the analysis?

Overinterpretation is a common feature of journal articles. The best papers not only discuss the significance of the data but also recognize their limitations. If you start with a clear set of hypotheses, and your experiments are designed to address those hypotheses, then you are in a position to draw strong conclusions. It is bad practice (and misleading) to conclude that there are “significant differences” between groups if the statistical analysis doesn't support that conclusion. Don't extrapolate data based on limited populations to broad applicability. Be conservative.

Present the Material Clearly and Cogently

It has often been said that if you don't get the study published, you might just as well have not done the work. Given that perspective, what are the secrets of writing a manuscript and getting it published? There are many volumes devoted to manuscript writing (e.g., Davis, 2004; Cargill & O'Connor, 2009; Katz, 2009; Day & Gastel, 2011), and there are many approaches to the process. Manuscript writing is an art (or at least a craft) that can be refined with attention and practice. From the point of view of a journal editor, there are a few important features that make a manuscript particularly attractive:

  • 1.The message is clear and easy to understand.

Keep your message simple. Readers–and reviewers and editors–need to be able to understand the rationale for the study and the goals of the work. Provide the relevant background and context, but don't confuse your message with a lot of extraneous information. In setting the context for your work, try to articulate the problem clearly, and indicate why/how your approach to the problem will help provide better understanding. Above all, make it easy for your readers.

  • 2.The paper is well organized.

Structure your paper so that your experiments and observations flow logically. Readers should be able to follow the rationale behind your methods, easily relate the text to the figures and tables, and appreciate how the conclusions follow from the results. Remember, you do studies/experiments to test hypotheses and make “objective” observations–not to confirm your own beliefs/guesses about the answer to a problem. Your conclusions, therefore, should reflect the actual experimental data (study results), not what you want the study to confirm.

  • 3.The manuscript uses the standard format for the journal to which you've submitted.

Pay attention to the instructions that every journal provides to its authors. If your study does not fit the journal's format, then you've probably chosen the wrong journal. Choose a manuscript length that will allow you to present the relevant data (more is not necessarily better). Use the referencing format described in the journal's instructions. Provide the requested information about authors and their affiliations.

  • 4.Figures and tables are designed so as to present the study results in an easy-to-understand manner.

Figures and tables are often the best way to present the details of your results and to summarize your data. But figures and tables can also be abused, leading to confusion (and neglect). Many books have been written about how best to present data in figures and tables (e.g., Lang, 2010). Try to keep them relatively simple. Complex figures and tables are annoying and often ignored. Take advantage of color (if the journal publishes in color), and use visual cues to help your reader keep track of experimental groups or treatments.

  • 5.The authors conform to high ethical standards.

Effective communication of study results depends, in large measure, on trust. The reader trusts that the authors adhere to ethical practice, both in the execution of their study and in the reporting of their results. In return, the authors' obligations include a pledge to ethical practices, the most basic of which include: presenting real (not fabricated) data; presenting all the relevant results (not selecting data to “prove” one's point); and giving appropriate credit to others for previously-published results and ideas (i.e., not plagiarizing from others). Ethical publication also includes a number of less obvious practices, most of which have been described at length in previous publications (e.g., Graf et al., 2007) and on various websites (e.g., With pressures to publish often intense, and with competition a real part of the modern research environment, adherence to high standards of ethical publication is more important than ever.

Make Sure You Use Proper, Easy-To-Understand English

While it may not be “fair,” the international language for science publication is currently English. That being the case, presenting material in proper, easy-to-understand English is a factor that cannot be ignored if an author wants to have his/her paper accepted for publication. English is not an easy language. Indeed, even many native English speakers do not convey their thoughts effectively in written English. For them, and for those who do not use English routinely for communication, here are a couple of suggestions.

  • 1.Have your manuscript edited by a native English speaker.

While it is true that not all native English speakers can communicate well in written English, they can often hear whether something “sounds” right. Unfortunately, in trying to communicate scientific/clinical concepts in English, non-English speakers will need the help of someone who not only speaks (and writes) English well but also has some knowledge of the relevant area of research. Constructing a proper English sentence that also conveys the specialized concept that you want to express is more difficult than simply using acceptable English.

  • 2.Consider paying for consulting services.

Some journals provide translation and English editorial services, but most do not. Virtually all journals can recommend translation and/or editorial services that provide consultation for a fee. As an author, you may need to determine how much you are willing to spend to publish in a given journal. Are you willing to pay someone if he/she can edit your paper in proper English?

Conflict of Interest

The author offers English editing services, which may be construed as a conflict of interest with respect to a message expressed in this commentary. The author confirms that he has read the Journal's position on issues involved in ethical publication and affirms that this commentary is consistent with those guidelines.