Their initial objective was to develop a list of selected titles for hospital libraries. However, the lists had great reach and influence, and medical libraries of all sizes and types relied on their guidance. Although the librarians had help and support from colleagues, publishers, and medical book wholesalers, they essentially reviewed and made the selections themselves, and they never maintained that their selection process was anything but subjective.

Each new list was published in the J Med Libr Assoc, and then the Medical Library Association (MLA) sold reprints of the lists to medical book wholesalers, who, in turn, distributed the lists for free throughout the medical library community. In 2001, the Brandon / Hill Selected Lists became available online (http://www.mssm.edu/library/brandon-hill) [1,2].

Doody provides both Content Specialists and Library Selectors with Web-based tools that allow access to the titles that appeared in the final issues of the Brandon / Hill Selected Lists; a searchable database of all in-print book and software titles in Doody's Book Review Service™; an aggregated database consisting of all information from Doody's database (including reviews and ratings; tables of contents licensed from Majors; and bibliographic data on all in-print book and software titles from the web sites of the three major medical book distributors.

According to Doody's description of its methodology, "use of the same five scoring tools and criteria for each title and selector – and averaging the scores across multiple selectors – yields a more "objective" measure than simply asking 'is this a core title?"' In addition, the greater the number of selectors scoring titles in a given specialty, the greater the degree of "objectivity" achieved. Doody states that "relying upon scoring by multiple individuals to provide objectivity is a time-tested approach," and is part of their effort to bring evidence-based methodology to collection development.

The five criteria for collection development are: Authoritativeness of Author(s) and Publisher; Scope and Coverage of the Subject Matter; Quality of Content; Usefulness and Purpose; and Value for the Money.

Edition In nearly all cases, the latest edition is listed, provided it was available for review. In a few instances, the most recent edition is listed, though not yet available, or a link is provided. This irregularity is due to efforts to publish the inaugural edition in time to be of use for library collection budgets; it should disappear with the next edition of DCT 2004.

Volume or Cover Type The goal of the DCT 2004 is to offer information about the single volume of a title that is offered in single and multivolume packages; about the hard cover version of a book offered in both hard and soft cover; and to give the most up-to-date retail pricing available.

Publisher Mergers and acquisitions have resulted in core titles changing publishers. Although the DCT 2004 strives to list the publisher which currently owns the right to the title, the way titles are listed in the two databases that supplied the information for DCT 2004 has made that difficult to do

Scoring A unique attribute of DCT 2004 is the score assigned to each title. See the section above for a complete description of the process.

Display options:

When selecting titles, features include View Mode (condensed and expanded) and various Sort By selections: Author's Last Name, Title, Price, Score, and Copyright Year. Identifying icons are Essential Core and Key Core titles. Titles can be chosen and viewed in the following ways:

• List Overview and Analysis

• Titles By Specialty

• Unique Title List

In terms of psychiatry, it is clear that most core titles are represented. These include (but are certainly not limited to):

American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders: Text Revision, 4th Edition, 2000;

Davis's Neuropsychopharmacology: The Fifth Generation of Progress: An Official Publication of the American College of Neuropsychopharmacology, 5th Edition, 2002;

Janicak's Principles and Practice of Psychopharmacotherapy, 3rd Edition, 2001;

Sadock, Kaplan and Sadock's Comprehensive Textbook of Psychiatry – 2 Volume Set, 7th Edition, 2000

Schatzberg's Manual of Clinical Psychopharmacology, 4th Edition, 2003, and The American Psychiatric Publishing Textbook of Psychopharmacology, 3rd Edition, 2004;

Stahl's Essential Psychopharmacology: Neuroscientific Basis and Practical Applications, 2nd Edition, 2000; and

Yudofsky's The American Psychiatric Publishing Textbook of Neuropsychiatry and Clinical Neurosciences, 4th Edition, 2002.

In terms of pharmacy from the associated health sciences and pharmacology from the basic sciences, core titles listed include:

ASHP's AHFS Drug Handbook, 2nd Edition, 2003;

Allen's Ansel's pharmaceutical dosage forms and drug delivery systems, 8^th ed., 2005;

Avery's drug treatment : principles and practice of clinical pharmacology and therapeutics, 3^rd ed., 1987;

Cooper, The Biochemical Basis of Neuropharmacology, 8th Edition, 2003;

DiPiro's Pharmacotherapy: A Pathophysiologic Approach, 5th Edition, 2002; Evans's Trease and Evans pharmacognosy, 15^th ed., 2002.

Gahart's 2005 Intravenous Medications: A Handbook for Nurses and Allied Health Professionals, 2005;

Hardman's Goodman & Gilman's the pharmacological basis of therapeutics, 10th ed., 2001;

Katzung's Basic and Clinical Pharmacology, 9th Edition, 2004

Gennaro's Remington's pharmaceutical sciences, 20^th ed.;

Shargel's Applied biopharmaceutics & pharmacokinetics, 2005; and

Tietze's Clinical Skills for Pharmacists: A Patient-Focused Approach, 2nd ed., 2004.

It is clear that some fundamental texts are missing from the pharmacy/pharmacology lists. For example, Block, Wilson and Gisvold's Textbook of Organic Medicinal and Pharmaceutical Chemistry, 11th Edition, 2004; Goldfrank's toxicologic emergencies, 7^th ed., 2002; and Koda-Kimble's Applied therapeutics : the clinical use of drugs, 8^th ed. 2005.

With regard to nursing theory, many core titles are listed:

Anderson's Community as Partner: Theory and Practice in Nursing, 4th Edition, 2004;

Andrews's Transcultural Concepts in Nursing Care, 3rd Edition, 1999;

Cherry's Contemporary Nursing: Issues, Trends and Management, 2nd Edition, 2002;

Chinn's Integrated Knowledge Development in Nursing, 6th Edition, 2004;

Chitty's Professional Nursing: Concepts and Challenges, 4th edition, 2004;

Fawcett's Relationship of Theory and Research, 3rd edition, 1999;

Kim's Nursing Theories: Conceptual and Philosophical Foundations, 1999;

Orem's Nursing: Concepts of Practice, 6th Edition, 2001; and

Tomey's Nursing Theorists and Their Work, 5th Edition, 2002.

Missing, however, are: Fawcett's Contemporary nursing knowledge :analysis and evaluation of nursing models and theories, 2005; and some version of Nightingale's Notes on Nursing.

Thus, the strengths and weaknesses of Doody's DCT 2004 are illustrated by an analysis of the selection and rating of items in the psychiatry, pharmacology, and pharmacy sections. In the three disciplines discussed above, the DCT 2004 is more or less complete, but the rating system has too little variation (1–3) to measure meaningful differences between items. For instance, what is the true difference between mean scores of 2.5 and 2.6? or between 2.9 and 3.0? This overly restrictive range results in a system as necessarily idiosyncratic as the Brandon-Hill lists it de facto replaces. Take, for instance, the ratings of such foundational texts as the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders or Sadock, Kaplan and Sadock's Comprehensive Textbook of Psychiatry: the former is rated 3.0 and the latter 2.9. It is hard to understand how these classics are not rated the same. Not to put to fine a point on it, but if 3.0 is the highest rating of the most important psychiatric texts (the two Essential Core texts are each given a 3.0), while the mean score for Key Core titles is about 2.75, then is this mean difference between "Essential" and "Key" a distinction without a difference?

Restriction of range problem could result in spurious relationships or the attenuation of significant relationships; that is, a scale with restricted range is simply less sensitive to meaningful differences [3,4]. For example, when the titles are distributed by score, we see that 189 titles are scored at 2.9, 158 at 2.8, 206 at 2.7, and 183 and 2.6. It's difficult, if not impossible, to decipher how these numbers are representing quality differences. As with many poorly explained phenomena modeled as bell curves, most items fall into the upper middle quartile.

Moreover, the owners and designers of DCT 2004 have a different interpretation of "objectivity" than what it commonly refers to, at least in terms of scale construction. For example, as mentioned earlier in the review, they state that "Relying upon scoring by multiple individuals to provide objectivity is a time-tested approach." In terms of psychometric test construction, this refers to reliability not to objectivity. Again from earlier in the review, "Use of the same five scoring tools and criteria for each title and selector – and averaging the scores across multiple selectors – yields a more objective measure than simply asking 'is this a core title."' This is typically a method to establish a scale's construct validity, not its objectivity. In fact, I think it's safe to say that objectivity is not considered a scale property.

• Radiologic Technology

• Respiratory Therapy

• Case Management

• Research

• Theory