Retrieval Comparison of EndNote
to Search MEDLINE (Ovid and PubMed)
versus Searching Them Directly
Carole Gall
Frances A. Brahmi
ABSTRACT. Using EndNote version 7.0, the authors tested the search
capabilities of the EndNote search engine for retrieving citations from
MEDLINE for importation into EndNote, a citation management soft-
ware package. Ovid MEDLINE and PubMed were selected for the
comparison. Several searches were performed on Ovid MEDLINE and
PubMed using EndNote as the search engine, and the same searches
were run on both Ovid and PubMed directly. Findings indicate that it is
preferable to search MEDLINE directly rather than using EndNote.
The publishers of EndNote do warn its users about the limitations of
their product as a search engine when searching external databases. In
this article, the limitations of EndNote as a search engine for searching
MEDLINE were explored as related to MeSH, non-MeSH, citation veri-
fication, and author searching.
[Article copies available for a fee from The
Haworth Document Delivery Service: 1-800-HAWORTH. E-mail address:
<docdelivery@haworthpress.com> Website: <http://www.HaworthPress.com>
© 2004 by The Haworth Press, Inc. All rights reserved.]
Carole Gall, MLS (cfgall@iupui.edu) is Development Officer & Manager Print
Journal Collection, and Frances A. Brahmi, MA, MLS, AHIP (fbrahmi@iupui.edu)
is Curriculum & Education Director, both at Ruth Lilly Medical Library, Indiana
University School of Medicine, 975 West Walnut Street, IB Rm 322, Indianapolis,
IN 46202-5121. Ms. Brahmi is also Director of Lifelong Learning, Indiana Univer
-
sity School of Medicine.
Medical Reference Services Quarterly, Vol. 23(3), Fall 2004
http://www.haworthpress.com/web/MRSQ
© 2004 by The Haworth Press, Inc. All rights reserved.
Digital Object Identifier: 10.1300/J115v23n03_03 25
KEYWORDS. EndNote, MEDLINE, Ovid, PubMed, citation manage
-
ment software
INTRODUCTION
The purpose of this article is to demonstrate the differences of out
-
put from searching MEDLINE direct and searching MEDLINE using
EndNote. The authors were interested in the pitfalls of using EndNote
as the search intermediary to retrieve citations from MEDLINE versus
searching MEDLINE direct, using either PubMed or Ovid. Because
searching via EndNote is convenient in terms of one-stop shopping, so
to speak, it might be very tempting to the untrained searcher. In addi-
tion, searching from EndNote is fairly straightforward and simple. Ex-
pert searchers avail themselves of sophisticated tools in Ovid and
PubMed, but can that kind of searching be done via EndNote? Is it ever
appropriate to use the EndNote search features for MEDLINE? How do
the search results differ?
METHODOLOGY
Four searches were conducted using PubMed via EndNote (EN),
PubMed direct, Ovid via EndNote, and Ovid direct. The four searches
were as follows: Medical Subject Heading (MeSH) combined with
free-text; MeSH combined with Author; Non-MeSH; and known cita-
tion verification. Ovid and PubMed were selected because they are the
two most commonly used versions of MEDLINE in academic and hos-
pital settings. PubMed was searched the simplest way via the National
Library of Medicine and Ovid MEDLINE from 1966 through Week 2
of November 2003. The study was limited to the evaluation of output,
given the searching capabilities of each system. EndNote was not ad
-
dressed as a citation manager, but rather as a way to search external
databases in order to build a citation library. The EndNote manual cau
-
tions its users that the EndNote search interface is meant for simple
searches, and “not to replace the advanced search options” offered by
MEDLINE information providers. Actually, the EndNote manual pro
-
vides an extensive, up-to-date table of Output Formats and Correspond
-
ing Import Options to assist its users with downloading instructions for
29 databases.
26 MEDICAL REFERENCE SERVICES QUARTERLY
RESULTS
Combined Author and MeSH Term searching is a fairly common re
-
quest. In putting the search together, four different ways were tried, as
seen in Table 1. L. H. Einhorn is a local researcher who is recognized in
-
ternationally for his work on testicular cancer. He publishes both as
L. H. and L. Einhorn. As can be seen from Table 1, under EndNote
PubMed, using Einhorn (qualified to author) and ANDing it with
testicular (qualified to Keywords), resulted in no retrieval. This is be
-
cause testicular is not a MeSH term. However, if testicular is qualified
to Any Field, EndNote PubMed retrieves 161 articles. If Einhorn, LH
(qualified to author) is ANDed to testicular (qualified to Any field), the
retrieval is 158. Looking across the rows, it is easy to discern how con-
fusing this might be for a user. Similarly, the same is true for EndNote
Ovid searching.
The same author and MeSH term were searched in Ovid MEDLINE,
and the three different search strategies appear in Table 2. The first three
sets in Table 2 show the author and MeSH term ANDed together to
complete one search strategy. Sets four and five demonstrate additional
strategies, each complete in themselves. Set four uses a “$” as a wild-
card for the initial of the author, and uses “.mp.” to map the term to a ti-
tle, abstract, Chemical Abstracts Service Registry number or Enzyme
Commission number, and MeSH heading.
The entry terms “lyme disease AND children” were searched as fol-
lows: “lyme disease” was searched both as a MeSH term and as text
word, and “children” was searched as a text word and as the MeSH term
“child.” Table 3 shows the varied results again using EndNote PubMed,
EndNote Ovid MEDLINE, PubMed, and Ovid MEDLINE.
The strategy for searching PubMed directly is included in Table 3.
The strategy for searching Ovid MEDLINE directly is in Table 4.
Non-Medical Subject Heading (MeSH)
The non-MeSH term, GATA-3, a zinc finger protein that is preferen
-
tially expressed in T cells in adult animals, was selected because it is al
-
pha-numeric and includes a hyphen, traditionally a challenging format
for most database systems. This topic was searched as GATA-3, GATA
3, and GATA3.
As shown in Table 5, GATA-3 retrieved identical results (245 hits)
when searching PubMed directly or via EndNote (qualified to Any
Carole Gall and Frances A. Brahmi 27
28 MEDICAL REFERENCE SERVICES QUARTERLY
TABLE 1. Combined Author and MeSH Term
Combined Author
and Subject Search
(MeSH) 12/10/03
Einhorn
AND
testicular
Lh Einhorn
AND testicular-
neoplasms
Einhorn AND
testicular-
neoplasms
Einhorn lh*
AND testicular
EndNote PubMed
Author & Keywords 0 149 151 0
Author & Any Field 161 149 151 158
EndNote MEDLINE Ovid
Author & Keywords 151 136 151 136
Author & Any Field 161 136 151 143
PubMed 181 3 168 142
(no comma)
Ovid MEDLINE–
See below for strategies
136 136 151 79
(no comma)
*The comma after Einhorn was either included or not depending on the requirements of the system tested.
TABLE 2. Ovid MEDLINE 1966 to November Week 2 2003
#
Search History Results
1 Testicular Neoplasms/ 14871
2 Einhorn lh.au. 297
3 1 and 2 136
4 Einhorn $ (au) and testicular neoplasms.mp.** 151
5 Einhorn lh (au) and testicular.mp.** 79
**Mp = map to title, abstract, CAS registry/ec number, MeSH.
TABLE 3. Combined MeSH and Term and Free Text
Combined Subject Search (MeSH) AND free-text
Lyme Disease
AND Children
EndNote PubMed
Keyword & Any Field 863
Any Field & Any Field 901
EndNote Ovid MEDLINE
Keyword & Any Field 232
Any Field & Any Field 253
PubMed direct–(lyme disease AND children) 921
Ovid MEDLINE direct–see Table 4 for strategy 781
Field). The same search in Ovid MEDLINE via EndNote retrieved zero,
while searching Ovid MEDLINE directly yielded 297 hits.
GATA 3 (without the hyphen) yielded zero or mostly false drops,
with the exception of the direct Ovid MEDLINE search, where the term
was searched as “GATA 3.mp.” and yielded 315 hits. Using GATA3
(no space) yielded 58 articles in PubMed via EndNote (qualified to Any
Field) or directly, whereas Ovid MEDLINE via EndNote and Ovid
MEDLINE directly yielded 58 and 59 articles, respectively.
Citation Verification
A common way of using a citation management system is to search
for and save a known reference. The authors chose to include a known-
reference search in the comparison, and used a reference that was five
months old at the time of the search.
Houghton, J.; George, W.D.; Cuzick, J.; Duggan, C.; Fentiman,
I.S.; Spittle, M. Radiotherapy and tamoxifen in women with com
-
pletely excised ductal carcinoma in situ of the breast in the UK,
Australia, and New Zealand: randomised controlled trial. Lancet.
362(9378):95-102, 2003 Jul 12.
Carole Gall and Frances A. Brahmi 29
TABLE 4. Ovid MEDLINE Direct 1966 to November Week 2 2003
# Search History
Results
1 Lyme Disease/ 5662
2 Child/ (entry word: children) 870850
3 1 and 2 781
TABLE 5. GATA-3, Non-MeSH Search Term with Two Variants
Non-MeSH Search
GATA-3 GATA 3 Variant GATA3 Variant
EndNote PubMed
Any Field 245 2,591 58
Keywords 0 0 0
EndNote Ovid MEDLINE
Any Field 0 0 58
Keywords 0 0 0
PubMed direct 245 1,007 58
Ovid MEDLINE direct 297 315 59
