David E. Winchester, MD; Anthony A. Bavry, MD, MPH
Potential Conflicts of Interest: None disclosed.
Winchester D., Bavry A.; Limitations of the MEDLINE Database in Constructing Meta-analyses. Ann Intern Med. 2011;154:372. doi: 10.7326/0003-4819-154-5-201103010-00017
Download citation file:
Published: Ann Intern Med. 2011;154(5):372.
We appreciate the comments from Dr. Leclercq and colleagues about our letter on limitations of MEDLINE. The authors elucidate the central point of our letter, which is that incomplete MeSH indexing can contribute to inaccurate searches when limits are used. To clarify our search strategy, the precise search string that we used was (“Platelet Glycoprotein GPIIb-IIIa Complex”[MeSH]) OR eptifibatide OR abciximab OR tirofiban. On repeating this search with the publication date limit of 1 February 2010 and the search limit “clinical trial,” we found the same results as Dr. Leclercq and colleagues: 1003 records.
After discussion with researchers at our institution and at the National Library of Medicine, we believe that 2 factors account for the increase in the number of studies from our initial report to the present. First, although we used the publication date limit, additional trials that were not in the database at the time of our search may since have been added. Second, as each database entry is fully indexed with the appropriate MeSH terms, the “clinical trial” term will be applied and more studies will be properly identified by the search limit. This reflects the shortcoming that we illustrated in our letter.
to gain full access to the content and tools.
Learn more about subscription options.
Register Now for a free account.
Copyright © 2016 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use
This PDF is available to Subscribers Only