Moving From Abstract to Manuscript

Abstract

Research is often presented at scientific conferences in abstract form. Unfortunately, a large percentage of abstracts are never published as a full manuscript, but having a strong department process for research can increase the number of abstracts published as manuscripts. Publishing as a full manuscript is critical to the advancement of science due to the rigor of the peer review process. This manuscript covers common reasons abstracts are not published as manuscripts, tips to overcome mistakes, how to respond to reviewer comments, and specific tips to avoid flaws in each manuscript section.

Introduction

Research is often presented at scientific conferences in abstract form, such as the annual American Association for Respiratory Care Congress.^1–3 Presenting abstracts is an important first step for early-career researchers and can be a career highlight for many clinicians.³ Unfortunately, a large percentage of abstracts presented at conferences are never published as a full manuscript, but having a strong departmental process can increase the number of abstracts published as manuscripts.⁴ This paper covers why publishing as a manuscript is important, the process of doing research that will result in manuscripts, reasons abstracts are not published as manuscripts, tips to maximize chances of acceptance, common reasons papers are not accepted for publication, the peer review process, and how to respond to reviewers.

Importance of Scientific Publishing

Publishing research in full manuscript form, including negative studies, is critical to the advancement of evidence-based medicine.^1,5 Unfortunately, results from a well-executed study that is only published in abstract form cannot be applied by other centers due to the less stringent peer review process for abstracts. Acceptance rates for abstracts may be > 80% compared to 20–25% acceptance rates for full manuscripts. The acceptance rate for manuscripts varies between journals, with higher-impact journals generally having lower acceptance rates. It is common for methodological flaws in a study to be revealed when attempting to write a full manuscript or during the peer review process. In most cases, studies published in manuscript form in peer-reviewed journals such as Respiratory Care do not have any major methodological flaws. No published study is perfect, and studies with minor flaws still contribute to the overall evidence for a given topic, but the peer review process should prevent publication of studies with major flaws.

Individual researchers or research teams who publish a large number of abstracts but few manuscripts may lack the courage of their convictions, do sloppy science, or lack the skills to translate their work into full manuscripts.⁶ In some cases, such as student projects or other learners, the goal of a project may be to write and present an abstract to learn the scientific process. For respiratory therapists performing research, there are several barriers to publishing abstracts as full manuscripts, including lack of protected time, training, and institutional support.⁷ Additional challenges noted by researchers are difficulty organizing data appropriately, difficulty writing, and challenges with journal formatting.⁸ Due to these limitations and others, it is estimated that only 5–6% of abstracts presented at the annual Open Forum are published as manuscripts.⁴

Having a Departmental Process for Research

For any research program to be successful, appropriate resources need to be available, and there needs to be a clear process in place to ensure each study is well designed and appropriately executed.^6,9,10 Resources include research infrastructure, including time dedicated for research, statistical support, and funding to present study results at scientific conferences. We evaluated our experience with a formal research committee and found an increase in abstracts published as manuscripts.⁴ This was accompanied by a decrease in the number of abstracts published each year. How to develop the process has been covered elsewhere;⁹ however, in brief, it involved monthly meetings, written proposals, and the expectation that each project be submitted as a full manuscript if possible. If this process is followed, abstracts have a higher likelihood of acceptance as a full manuscript.⁴ During the initial proposal stage, the planned authorship order and any disagreements can be resolved by the project leader or principal investigator. The final order may change, and for middle authors with similar contributions, listing them alphabetically is a reasonable practice.

Reasons Abstracts Are Not Published as Manuscripts

There are many reasons abstracts are not published as manuscripts.¹¹ The top reasons manuscripts are not published by Respiratory Care are included in Table 1. The top 2 reasons are failure to submit a manuscript and failure to resubmit after being offered a revision. Setting the expectation that each project will result in a manuscript submission can help solve both of these challenges.⁴ A team or lead author putting in enough work to submit a manuscript will likely be willing to resubmit if offered a revision. The downside is that some may choose not to pursue a project if a manuscript is expected, which may decrease the number of abstracts published by the team.⁴ Authors need to be cognizant that being offered a revision means there is a possibility, but not a guarantee, the paper will ultimately be accepted.

Table 1.

Reasons Manuscripts Are Not Published by Respiratory Care

No amount of clever writing or complex statistical analysis will allow a poorly designed study to be published. Having a rigorous process and a committee/mentor that looks at all proposals to evaluate the scientific merit of individual studies should minimize, but may not eliminate, poorly designed studies.^4,6,9 For bench studies, not using the correct tools or incorrectly modeling the disease state is a mistake that cannot be overcome. For example, some lung simulators are quite complex and require extensive knowledge to operate, and incorrectly calibrating the device can result in rejection. Disease states can be modeled incorrectly by using non-physiologic values for compliance or resistance. Surveys are another poignant example and have been covered in depth elsewhere.¹² Sampling bias, lack of internal validity, or low response rates are common reasons for surveys being rejected.

Modifiable reasons for manuscript rejection are not following manuscript preparation instructions, poor writing, inadequate description of the methods, and overstating the results in the discussion. Failure to follow the instructions for Respiratory Care will not automatically result in rejection; some other journals will return the paper to the authors for reformatting before it is sent to an editor for screening or to peer reviewers. Following the instructions carefully sends the signal to the editor and peer reviewers that the authors are detailed oriented.¹³ While perhaps unfair to authors, some editors and peer reviewers view an improperly formatted manuscript as a sign that the study itself was sloppily performed. A common example is a paper that was clearly formatted for submission to another journal and the sections, references, and cover letter are clearly intended for another journal. This causes the editor to assume that the paper was rejected at another journal. Taking the time to reformat for each journal can increase your chances of acceptance.

Different types of studies have different reporting standards. Randomized controlled trials follow the CONSORT guidelines, observational studies STROBE, multivariable predictive models TRIPOD, qualitative studies COREQ, quality improvement SQUIRE, and systematic reviews and meta-analysis use PRISMA.¹⁴ Following these reporting standards, which can be accessed at https://www.equator-network.org, can increase the likelihood of success. Utilizing these resources can help authors ensure their paper has the proper sections and each section contains all the relevant information.

Poor writing is a challenge that can be overcome with practice and constructive feedback from a mentor.⁹ Writing requires focus, and authors should work to reduce distractions when writing.¹⁴ This includes scheduling time to write, being in a quiet area, turning off cell phones or email notifications, and ensuring you are focused on writing. Novice researchers should remember that writing is a skill that can be developed over time. Most successful researchers are voracious readers. When reading an article, pay close attention to the details on how each section of similar papers is worded to fully understand what the authors are attempting to convey. It is advisable to use active voice, simple language, and avoid run-on sentences. Short declarative sentences may be easier to read than sentences with long complex structures. The use of simple language allows for all readers to fully understand the methodology. Whereas it can be tempting for authors to demonstrate their knowledge by using complex words or phrases, it can make the paper less readable and may increase the likelihood of rejection, especially when words are used incorrectly. For individuals whose first language is not English, it can be challenging to write in a second language, and language editing services can be used. For some papers with sound science but poor writing, the use of an editing service or working with someone experienced in English science writing can increase the likelihood of acceptance. In most cases, it is ideal to engage the person editing the manuscript before submission to a journal.

Once the manuscript is complete, authors are usually anxious to submit their work to a journal. To avoid careless writing errors, it is a good practice to put the completed manuscript aside for several days and then reread a final time before submission. Often this fresh read reveals writing errors (typos) that otherwise might have gone unnoticed. For novice writers, it may be helpful to also have your mentor or principal investigator take a final look prior to submission.

Tips for Individual Sections

Introduction

In contrast to an abstract, individual sections of a manuscript require more detail. Differences between manuscripts and abstracts are highlighted in Table 2. The introduction or background section should provide readers with background on why the topic is important, what is known about the topic, and how the study adds to knowledge of the topic. Common mistakes include being too short or overly detailed, missing references, or presenting a biased view of the available knowledge of the topic. This section usually includes 3 paragraphs, with the final sentence clearly stating the research hypothesis or purpose of the study.¹⁵

Table 2.

Differences in Sections Between Abstracts and Manuscripts

Methods

The methods section is the most important part of a research paper because it explains how the study was performed.¹⁶ A critical aspect of the methods section is providing enough detail that another research team could replicate the results. In contrast to an abstract, in which word limits make including all the details impossible, manuscripts will reveal every flaw in the study design.³ Unfortunately, many researchers do not include an adequate description of the methodology. For the purposes of this paper, we will assume the study was well designed and competently executed.

Common mistakes include not clearly describing the population studied or modeled, what interventions were done, the equipment used, how data were collected, and what statistical analysis was performed. Not collecting adequate demographic information, such as illness severity scores or pre-intervention lung mechanics, is also reason for rejection. It is important to clearly describe the inclusion and exclusion criteria. For bench models or physiologic studies, not providing detailed descriptions of the equipment used and how measurements were made can also result in rejection. Unclear description of complex statistical methods, using the wrong statistical tests, incorrect or overly optimistic sample size calculations, and inadequate justification for the statistical models used will also increase the likelihood of rejection.

Results

The results section is critically important to report the major findings of the study. In most cases, the data are primarily reported in tables and figures. In contrast to abstracts, the results section of a manuscript should report all the findings from the study, whereas an abstract can only report selected outcomes due to character limits. The text should include the major findings but not repeat what is in the tables. It can be challenging to decide what to put in text versus tables, but it is generally reasonable to note statistically significant differences in demographics in the first paragraph, primary outcome in the second, and any other outcomes in the following paragraphs, with multivariable or multivariate analysis in the final paragraph. Subgroup, secondary, or post hoc analysis can be reported in subheadings after the main study results.

Avoid long paragraphs as these are difficult for readers, especially when there are many results presented sequentially. Tables and figures should be clearly called out within the text and should stand on their own. Common mistakes in this section include interpretation of results, stating P values near statistical significance constitute a trend, not reporting all the results, or including information that should be in the methods. Some authors do not report enough data within this section, which increases the likelihood of rejection.¹¹

Tables and Figures

Tables and figures should be used to summarize the results. A reasonable structure is to have Table 1 report the demographic and baseline data, Table 2 primary and secondary outcomes, and Table 3 the results of multivariable analysis. Additional tables can be added for different types of studies, especially when there are many variables being analyzed. Figures summarize data visually. In most clinical studies, the first figure is a flow diagram of subjects in the study. This is not required for all studies but is helpful when many subjects were excluded for different reasons, allowing the reader to assess the internal and external validity of the study. Data presented in figures should not be repeated within the text.

Table 3.

Tips to Maximize Chances of Acceptance

Discussion

The primary purpose of the discussion is to explain the results.¹⁷ Many authors overstate the results in the discussion section of the paper.¹⁷ While it is understandable that authors are enthusiastic about their work, it is important to not overstate the results. For example, a bench study may discuss the clinical implications of the work but should refrain from suggesting the results would result in improved patient outcomes. Observational and physiologic studies often make mistakes related to causation and correlation. Even the highest-quality observational and physiologic studies are hypothesis generating.⁵ Classic examples of physiologic outcomes not resulting in improved outcomes include high-frequency oscillatory ventilation and inhaled nitric oxide for adult ARDS; both improved oxygenation but ultimately had no impact on mortality.^18,19 A contemporary example is driving pressure, which has been noted to be a potential driver of mortality in ARDS^20,21 but has yet to be confirmed in a randomized clinical trial.

Additional problems with the discussion section include missing important citations or presenting only data that support the authors' biases. A complete literature review is important for any project,²² and it is common for peer reviewers or editors to note missing citations during the peer review process. It is important for authors to compare their results to other studies so the reader can understand how their study fits within the knowledge on the topic. This may not result in rejection; however, it can be challenging for authors to provide a balanced view of the literature when they carry conscious or unconscious bias favoring the intervention being studied. Lastly, some authors are overly self-referential and use terms like “novel” or “we are the first to report,” which is off-putting for reviewers and editors. These self-congratulatory statements will not result in rejection but may negatively affect the reviewers' assessment of the paper.

Limitations

Every study, no matter how well executed, includes limitations. The limitations section is often part of the discussion section but may also be its own subsection. The authors should list all the noted limitations of the study. It is generally best to be open about any limitations instead of having them pointed out in an editorial or letter to the editor. Some authors attempt to justify each limitation or include strengths within this section; however, it is usually advisable to simply report the limitations. Providing strengths to balance limitations gives the appearance of an attempt to dilute the limitations.

Other Reasons for Rejection

Less common reasons for rejection include not getting institutional review board approval for human subject research, no approval from the institutional animal care and use committee, unreported conflicts of interest, failure to register a clinical trial, and not registering a systematic review. It is not unethical to have relationships with industry; however, it is unethical to fail to disclose them. It is up to the reader to determine if the relationship(s) biased the study. Many respiratory care studies are funded by industry and present a potential source of bias. The role of the company should be clearly stated within the text. Respiratory Care allows industry representatives to be coauthors; however, the lead author and principal investigator should not work for industry, and the data analysis should be done independently of the company. Failure to register clinical trials will result in rejection, and systematic reviews may also be rejected if not registered. Deceiving subjects to gain consent or coercing subjects to enroll will also result in rejection.

Tips to Maximize Chances of Acceptance

Some tips to maximize the chances of acceptance are summarized in Table 3. A cover letter to the editor can help explain why your study is important and what it adds to the body of knowledge. Editing and feedback during the writing process are important considerations. Many novice or early-career researchers take constructive feedback personally or become overly defensive when their work is criticized. It is usually a good idea to take a few days to allow emotions to cool before making the revisions. Most manuscripts require multiple revisions, even for experienced researchers. Each draft of the paper should result in improved clarity of each individual section, including the tables and figures. I have had papers go through over 10 revisions prior to submission, and my first manuscript²³ took over 20 revisions prior to submission! Once the lead and senior author/mentor/principal investigator agree, the final draft can be circulated to the team before submission.

The Peer Review Process

After submission, manuscripts are screened by the editorial office. Once all the correct forms (these vary from journal to journal) are submitted, it will be forwarded to the editor for review. In some cases, manuscripts are rejected at this stage for a variety of reasons. Once ready for peer review, the paper is sent to multiple experts on the topic. Once all the reviews are returned, the editor makes the decision. Often, the editor making the decision will have additional comments to the authors to be addressed in a revision.

Revising the Manuscript

It is highly unusual to have a paper accepted without revision. Failing to resubmit after being offered a revision is a common reason for rejection. At Respiratory Care, we allow 6 months after decision for resubmission. Generally, it is ideal to resubmit the manuscript as soon as possible so it is still fresh in the editor and reviewers' minds.

Once offered a revision, authors need to revise the manuscript and respond to reviewer and editor comments. A good strategy is to cut and paste the editor and reviewer comments into word processing software and respond to them one by one. Every comment should have a response, even if the authors feel as though they do not need to make the suggested changes. Some authors may respond in a numbered format or use bullet points; the format is less important than clearly noting the comment and reviewer response. It is a good strategy to include passages revised based on the reviewers' feedback within the response to decision section so reviewers can clearly see if the changes they suggested were made. Often, these comments provide valuable insight into sections that were not clearly written and can identify missing citations and misinterpretation of the papers in question.

Usually, the first author will make the revisions and send to the team for review. In some cases, the revisions will be minimal; but even in those cases, at minimum the first and senior authors should agree with the revisions and reviewer responses. If more extensive revisions are needed, the revision may go through several drafts before being resubmitted. At Respiratory Care, we request 2 documents, one clean and one clearly marked where the text was changed. This allows reviewers and editors to quickly evaluate the changes that were made. The marked document can be made using the Track Changes function in Microsoft Word, highlighting the changed text with a different background color or changing the font to red where edits were made.

It is critical for authors to be polite, professional, and appreciative when responding to reviewer comments. A combative, defiant response or a condescending tone can result in rejection, even after multiple revisions. Avoid any personal attacks or angry verbiage. While not every change suggested by reviewers needs to be made, in most cases the reviewers' goals are to improve the manuscript to ensure the paper can be clearly understood by readers. There should be clear justification for why any changes were made or not made. Some reviewers may be very focused on copyediting, grammar, and other minor issues. In these situations, it is best to be polite and make the changes as requested.

Lastly, references are often formatted incorrectly during the first submission. At Respiratory Care, we do not reject papers if the references are not formatted correctly in the initial submission but may if they are not fixed in subsequent revisions. There are several software programs available to help with reference formatting, including EndNote and Zotero, which both have the format for Respiratory Care. In addition, instructions for authors are included on our web site and attached to the decision letter when a revision is offered.

Summary

It can be challenging for novice researchers to make the leap from abstract to manuscript. Having a clear process, expecting projects to result in manuscript submission, proper formatting, avoiding overinterpretation, and professional responses to reviewer comments will increase the likelihood of acceptance.