Objectives: Our objective was to design simplified search strategies for three completed
SRs in ZPH. Their original search strategies were comprehensive. Reviews
included: the zoonotic potential of Mycobacterium avium ssp. paratuberculosis
(MAP); effectiveness of probiotics for reducing E. coli O157 in
domestic weaned ruminants (EC); and a review of the microbial load and
antimicrobial resistance in organic food animal production systems
(Organics). Methods: We elicited the opinions of SR methodology/ZPH
experts using a survey designed to identify relevant databases and search
terms. Two brief search strategies (BSS) were formulated from responses.
BSS1 included original search terms and the top three ranked databases.
BSS2 used the same top databases and the top ranked search terms.
Retrieved citations were downloaded and compared with the list of
relevant articles included in the original SRs. Original search verification
strategies were utilized to evaluate if missing articles could be captured,
and to investigate the reasons for missing articles. Results: The sensitivity
of BSS1 and BSS2 were 94% and 89% for MAP, 95% and 70% for
Organics, and 50% and 50% for EC, respectively. Search verification,
which was different for each review, found all missing articles for MAP
except one (99%), whereas the other two SRs’ verification failed to yield
any additional citations. Two major reasons for missing articles was an
insufficient number of search terms and indexed articles not being
captured by the database for no apparent reason. Conclusions: The
results from this study show that effective search strategies for SRs in ZPH
should include comprehensive search terms and algorithms, while the
number of databases can be reduced to at least three major databases e.g.
MEDLINE, selected based on the topic. Searching the literature is a critical
stage in any systematic review and more research is needed to develop a
protocol for creating effective search strategies in ZPH.