The purpose of this study was to carry out a comparison between references to articles published in the Journal of Music Therapy before December 2002 identified through searching PubMed/MEDLINE, the lngenta database and those actually published in the journal. To carry out this comparison we have created a new database, called the Music Therapy World Journal Index, which includes complete indexes of selected music therapy journals. All articles published in the Journal of Music Therapy since its commencement in 1964 are referenced in this database. We found a large number of articles that are not identified by searching PubMed/MEDLINE. There is an under-representation of both the amount of articles published in this journal and in the range of areas for clinical application and research in music therapy. The Ingenta database is comprehensive but limited in its coverage of the early years. To carry out a comprehensive research of the literature, it is necessary to consult an expert database, such as the Music Therapy World Journal Index, that guarantees an unequivocal level of completeness.
MEDLINE is a database of bibliographical references to journal articles produced by the US National Library of Medicine (http://www.nlm.nih.gov) and contains indices of articles published since 1966 in approximately 4,500 selected journals. The Journals cover "the fields of medicine, nursing, dentistry, veterinary medicine, the health care system, the preclinical sciences, and some other areas of the life sciences" (MEDLARS, 2001). The journal indices are also produced in printed form as Index Medicus. MEDLINE is also available on CD-Rom.
It has been possible to search the MEDLINE database online, via Internet, since 1971. The U.S. National Center for Biotechnology Information has developed an additional service called PubMed that provides free access to MEDLINE (http://www.ncbi.nlm.nih. gov/pubmed/). Access to MEDLINE is also provided by commercial organizations and the use of these services requires a fixed-fee.
To the present day the PubMed/MEDLINE database has included indexes of only one music therapy journal, the Journal of Music Therapy. No other music therapy journal has been considered to fulfil the criteria for inclusion in PubMed/MEDLINE by the National Library of Medicine.
When searching for literature it is important to be able to rely on comprehensive databases. We need to know if all articles published in the journal are referenced in the database. Otherwise relevant literature may not be identified, and we will not be able to judge the efficiency of our search strategy. More significantly, if we are not informed about existing relevant material, the value and relevance of both our clinical and research endeavors is jeopardized. A comprehensive search of the literature is also the foundation of a structured review. We must, therefore, be fully informed about how comprehensive a database is and how accurate the contents are.
There have been various assessments of the coverage and comprehensiveness of the MEDLINE database. Dickersin, Scherer, and Lefebvre (1994) found only 77% of all known trials in a search of MEDLINE for randomized controlled trials in the area of ophthalmology. The Cochrane Collaboration (Clarke & Oxman, 2001) stated that "only 30-80% of all known published randomized controlled trials are identifiable using MEDLINE (depending on the area or specific question)" (p. 28). Greenhalgh (2001) emphasizes the situation in which "according to one estimate, 40% of material which should be listed by MEDLINE can, in reality, only be accessed by looking through all the journals again, by hand" (p. 34). It is of interest to assess PubMed/MEDLINE in the light of music therapy literature. Our suspicion of missing references developed during early stages of our Structured Review Project.
We have carried out a large number of searches of PubMed/MEDLINE as a part of our Structured Review Project that began in April 2002. …