Editors of this issue:

Gail Hendler, Ehrman Medical Library, New York University Medical School, 550 First Avenue, New York, NY 10016, S-10, E-mail: hendlerg@yahoo.com,

and

William Self, The Medical Library Center of New York, 5 East 102nd St., 7th Floor, New York, NY 10029 S-1, Phone: 212-427-1630, Fax: 212-860-3496, E-mail: wself@mlcny.org.

Bachmann LM, Coray R, Estermann P, Riet GT. Identifying diagnostic studies in MEDLINE - reducing the number needed to read. JAMIA; preprint http://www.jamia.org/cgi/reprint/M1124v1.pdf

A welcome article to librarians, physicians and others interested in improving search retrieval in evidence-based medicine searches. The study focuses on the research conducted by Haynes et al ¹ and currently used as part of PubMed's clinical queries feature. The authors note that the precision of the clinical query of diagnostic studies is much lower than that of the clinical query on therapeutics. The authors sought to develop a more robust and precise search statement for diagnostic studies. They used the gold standard of a hand-search of medical journals for three separate years and then identified appropriate search words through a frequency analysis. They found that using the truncated terms (diagno*, predict* and accura*) in combination with the MeSH term "sensitive and specificity" produced high sensitivity and precision rates. The authors also coined the term, "number needed to read" to describe the number or irrelevant references someone must read in order to find one of relevance. The authors do not mention plans to contact either Dr. Haynes or the National Library of Medicine to implement suggested changes within the Clinical Queries feature but for librarians working in the area of evidence-based medicine, this publication provides insight and a useful approach to obtaining diagnostic studies.

1. Haynes RB, Wilczynski N, McKibbon KA, Walker CJ, Sinclair JC. Developing optimal search strategies for detecting clinically sound studies in MEDLINE. J Am Med Infom Assoc 1994;1:447-458.

Florance V, Moller MT. Better_Health@here.now. [Internet]. Association of American Medical Colleges. Available from http://www.aamc.org/programs/betterhealth/betterhealthbook.pdf.

If you haven't yet read this report, now is the time to do so. A quick read at 32 pages, the report is a series of recommendations made by the Association of American Medical Colleges' (AAMC) Better Health 2010 Advisory Board. The article updates, in part, the 1982 report (Matheson NW, Cooper JA. Academic information in the academic health sciences center. Roles for the library in information management. J Med Educ 1982;57(10 Pt 2):1-93) that influenced the National Library of Medicine's Integrated Advanced Information Management Systems (IAIMS) program. (The other report just out from AAMC addresses IAIMS specifically. It is the Integrated Advanced Information Management Systems IAIMS): The Next Generation (TNG). The complete report is available in PDF format at http://www.aamc.org/programs/betterhealth/start.htm#iaims.)

The report made recommendations in the following areas:

• Optimize the health and health care of individuals and populations through best-practice information management.
• Enable continuous, life-long, performance-based learning.
• Create tools and resources to support discovery, innovation and dissemination of research results.
• Build and operate a robust information environment that simultaneously enables health care, fosters learning and advances science.

Llewellyn RD, Pellack LJ, Shonrock DD. The use of electronic-only journals in scientific research. Issues in Science and Technology Librarianship 2002; Available from http://www.istl.org/02-summer/refereed.html.

This piece is an examination of the state of e-only journals in scientific research. The crux of their survey is in assessing the presence of e-only journals in traditional library access points: cataloged in library catalogs, indexed by indexing services and cited by scholars in their own work. The authors also include the results of a brief survey of indexing services as to why they did or did not index an e-only journal. The authors compiled a list of e-only journals (defined by them as journals that exist only electronically and not as an electronic counterpart to a print journal) in the sciences. Of the 144 identified titles, 45 were in biology/medicine (27 identified as "clearly medicine" and the other 14 were more biology in nature). Most of the e-only journals were free but a few (22/144) required paid subscriptions. MEDLINE indexed 7 (50%) of the e-only journals. Ninety-seven percent of the e-only journals were cataloged in OCLC and 66% of the titles were cited in other publications (Web of Science search). Overall, the authors found that the age of the e-only journal as well as its presence in indexing services and OCLC were indicators of use. Two interesting anomalies were the journals Psyche and J.USC: the Journal of Universal Computer Science, which were both heavily cited but not indexed in any service.

Solomon David J. Talking past each other: making sense of the debate over electronic publication. First Monday 7(8). http://firstmonday.org/issues/issue7_8/solomon/index.html

The article unfortunately does not live up to its enticing title. While First Monday articles are usually quite informative and entertaining reads, in this instance, the article seems more of an opinion piece than it does a well-argued and rational case for the wide, electronic dissemination of scholarly literature. But the reader seeking insight on the debate over electronic publication would do better to consult the references the author cites as dissemination models. BioMed Central in the biomedical sciences is a familiar one, but the author also references H-NET in the humanities and the Public Knowledge Project in education. One would expect a more rigorous approach from an editor and founder of an online publication: in this case, Medical Education Online (http://www.med-ed-online.org).