- W W Rosser,
- C Starkey and
- R Shaughnessy
Abstract
OBJECTIVE To explain differences in the results of literature searches in British general practice and North American family practice or family medicine.
DESIGN Comparative literature search.
SETTING The Department of Family and Community Medicine at the University of Toronto in Ontario.
METHOD Literature searches on MEDLINE demonstrated that certain search strategies ignored certain key words, depending on the search engine and the search terms chosen. Literature searches using the key words "general practice," "family practice," and "family medicine" combined with the topics "depression" and then "otitis media" were conducted in MEDLINE using four different Web-based search engines: Ovid, HealthGate, PubMed, and Internet Grateful Med.
MAIN OUTCOME MEASURES The number of MEDLINE references retrieved for both topics when searched with each of the three key words, "general practice," "family practice," and "family medicine" using each of the four search engines.
RESULTS For each topic, each search yielded very different articles. Some search engines did a better job of matching the term "general practice" to the terms "family medicine" and "family practice," and thus improved retrieval. The problem of language use extends to the variable use of terminology and differences in spelling between British and American English.
CONCLUSION We need to heighten awareness of literature search problems and the potential for duplication of research effort when some of the literature is ignored, and to suggest ways to overcome the deficiencies of the various search engines.
