UNYOC/MLA

Recently I was fortunate to attend the Systematic Review Workshop: The Nuts and Bolts for Librarians at the Health Sciences Library System (HSLS) at the University of Pittsburgh in Pittsburgh, PA with funding from the professional development grant that I received from UNYOC. This 2.5 day workshop concentrates on the theory and practice of systematic reviews and the role of the librarian. This jam-packed workshop is filled with valuable information and resources, but I’d like to focus on a couple of topics that were extremely beneficial to me.

Within the 2011 publication of Finding What Works in Health Care: Standards for Systematic Reviews by the Institute of Medicine, the IOM recommends that investigators establish teams with appropriate expertise, work with a librarian to develop a search strategy, use an independent librarian to peer review the search strategy and document the search strategy. With this recommendation investigators are more open to asking for help from librarians for their systematic reviews. Here at the University of Rochester Medical Center we are just now beginning to receive requests for librarian assistance with systematic reviews, so I was pleased to be able to attend. A particularly useful session was entitled: “Librarian as Collaborator.” One component of this session outlined talking points that a librarian could use with an investigator to increase the chance of inclusion in a systematic review. These points include maximizing recall (using multiple databases, complex search strategies, many false positives) which lessens the risk of bias in search results or in interpretation of those results, a growing body of literature on the inadequacy of many published searches, and publication standards require detailed accounting. Librarians are encouraged to think of themselves as the equivalent of a biostatistician.

Search term harvesting was another session that was enlightening. This in-depth technique is useful in identifying all relevant subject headings, textwords and phrases. To begin, the investigator supplies good articles on the topic. The librarian will use multiple parts of the supplied articles including titles, abstracts, subject headings, etc. to identify search terms. In addition, the librarian will begin preliminary searching to identify other relevant database records and other studies’ search strategies, including other systematic reviews on the topic. For each concept identified, a “quick and dirty” search should be performed to look for additional subject headings that reflect the major concept and titles and abstracts should be reviewed looking for author-generated terms such as textwords and phrases. All search terms are recorded carefully on a harvest form and shared with the investigator. Search term harvesting is complex and time-consuming, but vital for a high quality systematic review.

I highly recommend this workshop and encourage all to attend if you have the opportunity. Thanks so much to UNYOC and its members for their continued support of professional development.

Angela Dixon, BSN, MLS, AHIP
University of Rochester Medical Center