How to write an abstract

Abstract

Abstract

Since the abstracts of original papers are one of their most frequently read and most easily accessible elements, they should be as informative and accurate as possible. It is therefore worrying that 18–68% of 264 abstracts from six major general medical journals, were shown to contain data that were either inconsistent with or absent from the main body of the article. This paper provides an overview of published structures for writing an abstract of an original study or review and quality criteria to assess such abstracts. Guidelines for structured abstracts were first proposed in 1987 and have been under continuous review since. Today, nearly all journals request authors to prepare a structured abstract before peer review, preferably in accordance with the IMRAD format (i.e. Introduction, Methods, Results And Discussion) or the ‘eight-heading’ format (i.e. Objective, Design, Setting, Patients and participants, Intervention(s), Main outcome measures, Results, and Conclusions). In addition, guidelines for structured abstracts for review studies are available. Quality criteria to assess abstracts have been available since 1993. Their use has been associated with decreased discrepancy between the abstract and the main report. Editorial boards of journals are recommended to provide instructions on the components of structured abstracts. Authors should provide accurate data, including the notion if these are preliminary or final. Reviewers should pay increased attention to the quality of the abstract, while readers must retain a cautious and critical reading attitude at all times.