Free
Special Communication  |   January 2015
Mastering the Art of Abstracts
Author Affiliations & Notes
  • Laura Riordan, BA
    Mrs Riordan is assistant managing editor for The Journal of the American Osteopathic Association.
  •  *Address correspondence to Laura Riordan, Assistant Managing Editor, American Osteopathic Association, 142 E Ontario St, Chicago, IL 60611-2864. E-mail: jaoa@osteopathic.org
     
Article Information
Professional Issues
Special Communication   |   January 2015
Mastering the Art of Abstracts
The Journal of the American Osteopathic Association, January 2015, Vol. 115, 41-45. doi:10.7556/jaoa.2015.006
The Journal of the American Osteopathic Association, January 2015, Vol. 115, 41-45. doi:10.7556/jaoa.2015.006
Abstracts are arguably the most important part of a scientific article. Often, abstracts are the only substantive portion of an article that readers are able to view in electronic databases1 and on pay-per-view journal websites2 and thus may be the only part of the article that is read. An abstract must therefore be well written, stand on its own, accurately portray the content of the full article, and capture readers' attention.1-3 As authors are well aware, however, writing an abstract that meets all of these requirements—not to mention adheres to stringent word count limits—is easier said than done. 
In this fifth part in our series on scientific writing,4-7 I review key elements of abstract writing, including basic structure, required components, and tips for ensuring accuracy and meeting word count limits. 
General Guidelines
An abstract is a short summary of an article. For research articles, abstracts include the objective and scope of the investigation, the methods, the primary results, and the principal conclusions.3,8 Although readers usually read the abstract of an article first, authors should write this part of their manuscript last. Abstracts should contain the same information as the text of a manuscript—including methods, data, terminology, etc. Manuscripts often go through several revisions and rounds of corrections before a final draft is complete. If an abstract is written too early in the process, it may contain outdated or even incorrect information.1,2 
In How to Write and Publish a Scientific Paper, Day and Gastel state, "Usually, a good abstract is followed by a good paper; a poor abstract is a harbinger of woes to come."8 A quality manuscript describes a study, experience, or other type of finding by answering key questions about that study, experience, or finding. A quality abstract will accomplish the same goal by answering the same questions.2 Before drafting their abstract, authors should determine which type of abstract is most appropriate for their manuscript— structured or unstructured—and ensure that they are familiar with conventional guidelines for that abstract type. 
Structured Abstracts
Structured abstracts contain headings and briefly summarize the main sections of an article.8 They are typically required for original research articles (eFigure 1), systematic reviews (eFigure 2), and meta-analyses. The exact names and number of headings will vary depending on article type and journal preferences. In general, abstracts for randomized controlled trials should follow the Consolidated Standards of Reporting Trials, or CONSORT, checklist for abstracts,11 and abstracts of systematic reviews and meta-analyses should follow the Preferred Reporting Items for Systematic Reviews and Meta-Analyses, or PRISMA, guidelines.12 However, authors should adhere to abstract requirements of the journal to which they are submitting their manuscript. Word count limits for structured abstracts vary (the JAOA's limit is 350 words).13 Authors should keep in mind that some electronic databases also have word limits14; even if a journal allows a lengthy abstract, the full published abstract might not always be available to database researchers if it is truncated by the database. 
eFigure 1.
Example of a structured abstract for an original research article. Adapted from Ruhlen et al.9
eFigure 1.
Example of a structured abstract for an original research article. Adapted from Ruhlen et al.9
eFigure 2.
Example of a structured abstract for a systematic review article. Adapted from Mueller et al.10
eFigure 2.
Example of a structured abstract for a systematic review article. Adapted from Mueller et al.10
The following sections detail general guidelines for original research article abstracts, as they are the most common type of structured abstract. 
Context
The context of an abstract should answer the question, "Why is the current study important?" In other words, it should describe the rationale behind the study question and emphasize new and important areas addressed by the study.1,13 Authors should limit this section to 1 sentence. Consider the following example15: 
XYZ has been studied in A population, but its prevalence in B population has not to our knowledge been investigated. 
In the example, the context statement describes the importance of the study and why the study was initated: Research is lacking on the prevalence of XYZ in B population. 
Objective
The objective section of an abstract should contain the study's primary objective. Key secondary objectives and a priori hypotheses are also appropriate for this section but may be omitted because of space constraints. The objective should be a "To..." statement.3 For example15: 
To determine the prevalence of XYZ in population B and to compare findings with previously published data in population A. 
Methods
The methods section of an abstract should state the study's design, setting, participants, intervention(s), and main outcome measures. For large or complex studies, it is often appropriate to break the methods into several sections (eg, participants, intervention, main outcome measures). It is unlikely that authors will be able to include every aspect of their study's methodology in the abstract. To keep the abstract concise, authors should list only key study criteria, essential features of the intervention, and primary outcome measures3,15: 
Participants (aged ≥ 18 y) from population B in the B region were recruited from March 2011 through December 2013. Participants were excluded if they had a history of ABC. The authors gathered data on XYZ from population B and compared them with data from population A. 
Results
The results should contain the main outcomes of the study and their statistical or clinical significance.1 Findings should be in the form of raw data (not just percentages) and be accompanied by relevant statistical information (eg, P values, CIs).3 Authors should ensure that findings are included for all outcome measures described in the methods. Likewise, the results section should not contain findings for outcomes that were not already described in the methods section. Abstracts for survey-based studies should include response rates.3,15 
Of the XX participants recruited, XX met the inclusion criteria and completed the study. In this population, XX participants (XX%) had XYZ, compared with XX (XX%) in population A. 
Conclusion
According to the AMA Manual of Style,3 authors should "[p]rovide only conclusions of the study directly supported by the results." For example15: 
XYZ was found to occur in XX% of our population. These findings are consistent with previously published data in population A. 
Supplemental information and in-depth evaluations of the findings should be reserved for the text of the manuscript,16 but authors should note important limitations.1 In addition, authors should take care to not overinterpret findings and refrain from recommending vast changes to clinical practice if additional research is needed. If applicable, clinical implications should be noted. Clinical trial registration numbers and registry names should appear at the end of the abstract (eg, ClinicalTrials .gov number 1234). 
Unstructured Abstracts
Unstructured abstracts are usually appropriate for manuscripts that do not involve original research, such as case reports and narrative reviews. Authors should always check the requirements of the journal to which they are submitting; abstracts are not typically required for opinion pieces, essays, poems, and letters to the editor.3 The word limit for unstructured abstracts is typically lower than that for structured abstracts. The JAOA requires unstructured abstracts to be 150 words or less.13 
Although specific guidelines vary depending on the type of manuscript (eg, the CARE Guidelines17 are helpful for case reports), unstructured abstracts should generally describe the context, findings or observations, conclusion, and implications of the information in the article. 
Context
Unstructured abstracts should start with a brief, 1- to 2-sentence statement that describes why the topic is important and timely. For review articles, authors should include a clear objective statement.13 For case reports, authors should describe why their case is unique and of interest to readers. 
XYZ is common in the US population, but to the authors' knowledge, no cases of XYZ in a patient presenting with ABC have been described. 
Findings
The authors should describe their findings or observations. For case reports, authors should include the resolution of the case. 
A 32-year-old man presented to the emergency department with ABC. Examination findings revealed XYZ. After management of XYZ, the patient's symptoms resolved and he was discharged to home. 
Conclusion
As with structured abstracts, unstructured abstracts should include conclusions directly supported by the authors' findings or observations.3 
As shown in the present case, XYZ is a potential underlying cause of ABC. 
Implications
Unstructured abstracts should end with the clinical or other implications of the authors' findings. 
Physicians should consider XYZ in patients who present with ABC. Early detection and management can improve outcomes for these patients. 
Using the above examples, the abstract for this case report would read as follows: 
XYZ is common in the US population, but to the authors' knowledge, no cases of XYZ in a patient presenting with ABC have been described. A 32-year-old man presented to the emergency department with ABC. Examination findings revealed XYZ. After management of XYZ, the patient's symptoms resolved and he was discharged to home. As shown in the present case, XYZ is a potential underlying cause of ABC. Physicians should consider XYZ in patients who present with ABC. Early detection and management can improve outcomes for these patients. 
Additional Considerations
After the abstract is drafted, authors should ensure that all components of the abstract are consistent with those included in the text and graphic elements of the manuscript.3 Terminology, presentation and rounding of data, and chronology of events in the body and abstract should match. The abstract should not contain any information that does not already appear in the text of the manuscript. 
Conventional guidelines for abstracts differ from those for the main body of the manuscript in a few key areas. For example, abstracts often contain phrases rather than complete sentences for brevity. They should not cite references, tables, or figures.3,16 Authors should avoid the use of acronyms and abbreviations unless a long term appears several times in the abstract.8,16 In general, abstracts should be able to stand on their own without reference to the text or any other components of the manuscript. 
Most guidelines for abstracts, however, are the same as those for the main body of the manuscript. It should have appropriate information, complete and accurate data, concise language, and good grammar. Authors should use nonproprietary drug names3 and follow appropriate style guidelines when reporting tests used, units of measure, and statistical findings. 
Authors commonly include extra, unnecessary information in their abstracts.8 By adhering to abstract guidelines and including only required information, authors can keep their abstract's word count in check. For additional tips on keeping abstracts within a certain word count, see the Table. 
Table.
Tips for Cutting Abstract Word Counts
Tip Example Revised
Eliminate redundant phrases period of time,3 aged XX years old,18 all of18 period, aged XX years, all
Delete unnecessary adjectives and adverbs Students were divided into small groups of 3 to 5. Students were divided into groups of 3 to 5.
BUT, use adjectives and adverbs to replace lengthy phrases Treatment sessions occurred every other week for 3 months. Biweekly treatment sessions occurred for 3 months.
Instead of using conjunctions, use 2 separate sentences Inclusion criteria were ABC, and exclusion criteria were XYZ. Inclusion criteria were ABC. Exclusion criteria were XYZ.
Use symbols (ensuring the appropriate use of style) Patients aged 18 years or older were recruited. Patients (aged ≥ 18 y) were recruited.
Report statistical findings parenthetically Mean findings were XX for group A and XX for group B, with statistically significant differences noted. Mean findings were higher for group A (XX) than for group B (XX) (P<.05).
Use lists creatively Mean findings were XX for group A, XX for group B, XX for group C, and XX for group D. Mean findings by group were as follows: A, XX; B, XX; C, XX; and D, XX.
Avoid expendable words this report describes, the authors investigated Omit; such phrases add nothing.3 Instead, state what is described or investigated.
Do not repeat information Number of patients is listed in both methods and results. Remove number from methods.
Table.
Tips for Cutting Abstract Word Counts
Tip Example Revised
Eliminate redundant phrases period of time,3 aged XX years old,18 all of18 period, aged XX years, all
Delete unnecessary adjectives and adverbs Students were divided into small groups of 3 to 5. Students were divided into groups of 3 to 5.
BUT, use adjectives and adverbs to replace lengthy phrases Treatment sessions occurred every other week for 3 months. Biweekly treatment sessions occurred for 3 months.
Instead of using conjunctions, use 2 separate sentences Inclusion criteria were ABC, and exclusion criteria were XYZ. Inclusion criteria were ABC. Exclusion criteria were XYZ.
Use symbols (ensuring the appropriate use of style) Patients aged 18 years or older were recruited. Patients (aged ≥ 18 y) were recruited.
Report statistical findings parenthetically Mean findings were XX for group A and XX for group B, with statistically significant differences noted. Mean findings were higher for group A (XX) than for group B (XX) (P<.05).
Use lists creatively Mean findings were XX for group A, XX for group B, XX for group C, and XX for group D. Mean findings by group were as follows: A, XX; B, XX; C, XX; and D, XX.
Avoid expendable words this report describes, the authors investigated Omit; such phrases add nothing.3 Instead, state what is described or investigated.
Do not repeat information Number of patients is listed in both methods and results. Remove number from methods.
×
Conclusion
Authors should select the abstract type appropriate for their manuscript and ensure it follows basic structure and includes essential information. By adhering to conventional guidelines, authors can ensure that their abstracts are meaningful, concise, and representative of their full-length manuscript. (doi:10.7556/jaoa.2015.006) 
References
International Committee of Medical Journal Editors. Uniform requirements for manuscripts submitted to biomedical journals. International Committee of Medical Journal Editors website. http://www.icmje.org/recommendations/browse/manuscript-preparation/preparing-for-submission.html. Accessed October 29, 2014.
Annesley TM. The abstract and the elevator talk: a tale of two summaries. Clin Chem. 2010; 56(4): 521-524. http://www.clinchem.org/content/56/4/521.full. [CrossRef] [PubMed]
Manuscript preparation: abstract. In: AMA Manual of Style: A Guide for Authors and Editors. 10th ed. New York, NY: Oxford University Press; 2007:20-24.
Riordan L. Modern-day considerations for references in scientific writing. J Am Osteopath Assoc. 2012; 112(8): 567-569. [PubMed]
Riordan L. Enhancing your manuscript with graphic elements, part 1: tables. J Am Osteopath Assoc. 2013; 113(1): 54-57. [PubMed]
Riordan L. Enhancing your manuscript with graphic elements, part 2: figures. J Am Osteopath Assoc. 2013; 113(5): 424-431. [PubMed]
Lo IY. Elements of manuscript submission. J Am Osteopath Assoc. 2013; 113(11): 849-852. doi:10.7556/jaoa.2013.060. [CrossRef] [PubMed]
How to prepare the abstract. In: Day RA, Gastel B. How to Write and Publish a Scientific Paper. 7th ed. Santa Barbara, CA: Greenwood; 2011.
Ruhlen RL, Singh VKPhD; Pazdernik VK, et al. Changes in rat spinal cord gene expression after inflammatory hyperalgesia of the joint and manual therapy. J Am Osteopath Assoc. 2014; 114(10): 768-776. doi:10.7556/jaoa.2014.151. [CrossRef] [PubMed]
Mueller A, Franke H, Resch KL, Fryer G. Effectiveness of osteopathic manipulative therapy for managing symptoms of irritable bowel syndrome: a systematic review. J Am Osteopath Assoc. 2014; 114(6): 470-479. doi:10.7556/jaoa.2014.098. [CrossRef] [PubMed]
Using the CONSORT for abstracts checklist: some examples. CONSORT website. http://www.consort-statement.org/Media/Default/Downloads/Extensions/CONSORT%20Extension%20for%20Abstracts%20checklist%20examples.pdf. Accessed October 29, 2014.
PRISMA 2009 checklist. PRISMA website. http://www.prisma-statement.org/2.1.2%20-%20PRISMA%202009%20Checklist.pdf. Accessed November 13, 2014.
JAOA information for authors. The Journal of the American Osteopathic Association website. http://www.jaoa.osteopathic.org/site/misc/ifora.xhtml. Accessed October 29, 2014.
Journal style and format. Scientific Style and Format: The CSE Manual for Authors, Editors, and Publishers. 8th ed. Chicago, IL: University of Chicago Press; 2006:511.
Sample outline of student research project. The Journal of the American Osteopathic Association website. http://www.jaoa.osteopathic.org/site/misc/template.rtf. Accessed October 29, 2014.
Coghill AM, Garson LR, eds. The ACS Style Guide: Effective Communication of Scientific Information. 3rd ed. New York, NY: Oxford University Press; 2014.
CARE checklist. CARE website. http://www.care-statement.org/care-checklist.html. Accessed November 13, 2014.
The Chicago Manual of Style. 16th ed. Chicago, IL: The University of Chicago Press; 2010.
eFigure 1.
Example of a structured abstract for an original research article. Adapted from Ruhlen et al.9
eFigure 1.
Example of a structured abstract for an original research article. Adapted from Ruhlen et al.9
eFigure 2.
Example of a structured abstract for a systematic review article. Adapted from Mueller et al.10
eFigure 2.
Example of a structured abstract for a systematic review article. Adapted from Mueller et al.10
Table.
Tips for Cutting Abstract Word Counts
Tip Example Revised
Eliminate redundant phrases period of time,3 aged XX years old,18 all of18 period, aged XX years, all
Delete unnecessary adjectives and adverbs Students were divided into small groups of 3 to 5. Students were divided into groups of 3 to 5.
BUT, use adjectives and adverbs to replace lengthy phrases Treatment sessions occurred every other week for 3 months. Biweekly treatment sessions occurred for 3 months.
Instead of using conjunctions, use 2 separate sentences Inclusion criteria were ABC, and exclusion criteria were XYZ. Inclusion criteria were ABC. Exclusion criteria were XYZ.
Use symbols (ensuring the appropriate use of style) Patients aged 18 years or older were recruited. Patients (aged ≥ 18 y) were recruited.
Report statistical findings parenthetically Mean findings were XX for group A and XX for group B, with statistically significant differences noted. Mean findings were higher for group A (XX) than for group B (XX) (P<.05).
Use lists creatively Mean findings were XX for group A, XX for group B, XX for group C, and XX for group D. Mean findings by group were as follows: A, XX; B, XX; C, XX; and D, XX.
Avoid expendable words this report describes, the authors investigated Omit; such phrases add nothing.3 Instead, state what is described or investigated.
Do not repeat information Number of patients is listed in both methods and results. Remove number from methods.
Table.
Tips for Cutting Abstract Word Counts
Tip Example Revised
Eliminate redundant phrases period of time,3 aged XX years old,18 all of18 period, aged XX years, all
Delete unnecessary adjectives and adverbs Students were divided into small groups of 3 to 5. Students were divided into groups of 3 to 5.
BUT, use adjectives and adverbs to replace lengthy phrases Treatment sessions occurred every other week for 3 months. Biweekly treatment sessions occurred for 3 months.
Instead of using conjunctions, use 2 separate sentences Inclusion criteria were ABC, and exclusion criteria were XYZ. Inclusion criteria were ABC. Exclusion criteria were XYZ.
Use symbols (ensuring the appropriate use of style) Patients aged 18 years or older were recruited. Patients (aged ≥ 18 y) were recruited.
Report statistical findings parenthetically Mean findings were XX for group A and XX for group B, with statistically significant differences noted. Mean findings were higher for group A (XX) than for group B (XX) (P<.05).
Use lists creatively Mean findings were XX for group A, XX for group B, XX for group C, and XX for group D. Mean findings by group were as follows: A, XX; B, XX; C, XX; and D, XX.
Avoid expendable words this report describes, the authors investigated Omit; such phrases add nothing.3 Instead, state what is described or investigated.
Do not repeat information Number of patients is listed in both methods and results. Remove number from methods.
×