What is the Truth about Abstracts
Over the past several months I have become more actively involved in the social media side of medicine. At first, I saw it as a way to promote my medical practice but as time went on it became much less about self promotion and much more about connecting with other spine care and musculoskeletal professionals. I have enjoyed the lively debates about various issues in pain medicine. One thing that has come to my attention in these discussions is how medical abstract references are used to prove or disprove and argument. Often times, several abstracts will be referenced to make a point. The more I observed this the more I wondered how accurate are the abstract as it relates to the entire article. It brought back memories of journal clubs in the past in which the abstract did not match the results of the paper.
I came across this article online:
The article was piece that was written about paper published in Spine this month.
Inconsistencies Between Abstracts and Manuscripts in Published Studies About Lumbar Spine Surgery
Lehmen, Jeff A. MD; Deering, Rachel M. MPH; Simpson, Andrew K. MD; Carrier, Charles S. BA; Bono, Christopher M. MD
Some of the conclusions of the paper were the following:
- Inconsistencies between the abstract and full manuscript in 75% of the studies they reviewed.
- The implied meaning of a study’s findings was found to differ between the abstract and manuscript in 15% of studies
- Important negative findings were omitted from 40% of abstracts and the primary outcome findings were clearly stated in only 23% of abstracts
- While most studies claimed randomization, in 28% of the papers it was unjustified
In order to truly get an accurate take on what the article is reporting a full read is truly necessary. The two biggest problems with that are time constraints and the lack of open source journal articles. Amazingly so, our government will fund research and then I have to pay $30 to get an article in a journal that I do not subscribe to. That is a whole other topic that needs to be discussed.