What’s an abstract, and why does it matter?
A recent report linking longer breastfeeding with more agressive breast cancer got me thinking the way researchers report their data. It’s a process that typically starts with a preliminary presentation at a scientific meeting, followed by a peer-reviewed article. The media can report on a story at either point, and they don’t always distinguish clearly between the two. The end result is that a lot of incomplete science gets published as fact.
Most researchers start out by presenting at scientific meetings. To present, the researcher prepares a short summary, called an abstract. These abstracts are usually no more than 300 words, and undergo a more-or-less cursory review by other researchers.
Abstracts that look really interesting to the reviewers are presented as short “oral presentations,” which are lectures with questions that last about 10 minutes. Lesser abstracts are presented as posters — essentially a science fair presentation by an adult researchers, where we stand next to a 3 x 6 foot poster summarizing out work. Some of these presentations and posters are later published in leading journals; others don’t ever get published.
To be published in a journal, studies undergo a much more extensive process called peer review. In this process, a researcher sends a manuscript to a medical journal, and two or three experts provide detailed, line-by-line feedback and critiques of the study design and the statistical methods. The author may either be offered a chance to respond and improve the manuscript, or may be told that it won’t be published in that journal. Top journals publish only a small percentage of submitted manuscripts. What this means is that if a study appears in a peer-reviewed journal, several experts have read it carefully, and the author has responded to those critiques. Reviewers will critique any parts of the article that exagerate the findings or suggest conclusions that are beyond the scope of the science.
Finally, there’s the story that appears in the media. Reporters may write about exciting or controversial studies that are presented at meetings, and they will also report on studies published in journals.
In both cases, the researcher works with the PR team for the scientific meeting or the journal, and/or their institution’s media relations department, to put together a press release about the findings. These releases typically include a “news peg” to generate publicity for the people involved. News reporters may or may not have access to the full scientific article. For presentations, such as the recent breast cancer study, all they have to go on is the abstract, which is just a few hundred words.
So what’s the bottom line? First, abstracts, or meeting presentations, are preliminary results that haven’t been rigorously reviewed by experts. Many clinicians do not change their practice based on meeting presentations, waiting instead for “peer review” and the chance to look at the study themselves.
Second, media reports may or may not be an accurate reflection of what the study actually shows. To get good information, it’s optimal to find the original article and not rely on a second hand version of the story.
Alison Stuebe, MD, MSc, is a maternal-fetal medicine physician and assistant professor of Obstetrics and Gynecology at the University of North Carolina School of Medicine.