The choice of the best most efficient healthcare search methodology on the Internet involves consideration of an over abundance of sites, megabit database resources, numerous and strangely different search methods and directories, formal and traditional medical topics and medical specialties, currency of information, very important information quality issues, and aggressive advertising.  One common methodology that we all know is the use of Web Indexes, Directories, and Search Engines as the gateway to information resources.  Several design aspects and issues of these concern their organizational structures (i.e., manner of listing, grouping, categories, and annotations),  and their contents (i.e., medical topics, credibility, and currency).  Other important aspects concern search strategies, statistical and computer algorithms, list ranking of responses, and output format and abstracted textual add-on blurbs, including side bar advertising. 

Web Indexes and Directories (http://www.yahoo.com is known as a directory) are often used to initiate specialized topic searches and to obtain quickly quality link references to the best sites.  They offer alternatives to using Search Engines (http://www.google.com is known as a search engine) by avoiding getting overwhelmed or overloaded by thousands and millions of references, many of which are irrelevant, repeats, and of dangerously poor medical quality.  Indexes and Directories now come in many styles and concepts and in many levels of categories and link output detail.  Many are available with no frills, no side bar advertising, and with excellent descriptive annotations.  Unfortunately, a growing number of health care lists are now available on the Web with only general consumer topics (like Travel, Entertainment, and Real Estate) and little serious attention to medical subject matter.  And the general public consumer doesn't realize the difference. 

The variations in structural styles reflect the desires of the audience or special groups of potential consumers.  Professional searchers (e.g., physicians, nurses, librarians) seem to benefit from greater organizational front-end categorical details with peer reviewed abstracts and full text articles..  Patients seem to benefit from fewer details and with focus on common disease categories and not too technical text.  Some professional directories provide a keyword search input box linking directly to the page content of referenced Web sites, which makes them almost vertical search engines.       

Traditional structures can be seen in the Web sites of the National Library of Medicine, (http://www.nlm.nih.gov/) MESH and MEDLINE, and related NLM databases; the Library of Congress Medicine Index, as well as many private contract databases useful for full text searches.  But there are many others appearing now on the Web with a wide range of organizational structures and  information.  An example of a very complex structure is the Google Directory with its gigantic list of categories and millions of references from its supporting database. 

In plain English, there are many practical details in Web directory and indexing that involve window dressing, the amount of sugar coated frosting on the cake, the ultimate responsible author for the information on the Web sites, and the guy who pays for the site.  The current medical Web has gotten more unstable and more chaotic than ever before.  And now there are many copycats in our crowded Domain Name System.  Fortunately, there really are great gems of information to be found if the Web is conducted, prepared, and searched with professional integrity and high-tech skill.   Inspite of the size and popularity of search engines, the size of their databases, their statistical algorithms, their cached information, and their great speed in returning references, there are many search strategies that they can't do in a few seconds.  Personalized human searches can still get the best results. 

The flow process used in directory development falls into what some academic people call Data Mining, or more specifically, Text Mining.  The steps of this flow are diagrammed in the following Figure 2.  Length and depth is dependent on the style of the index or directory chosen, and also on the size of the database that support the categories of information, including site addresses, site titles, annotations, and links to cross-references and related resource sites. 

Several examples of directory categories and their format structure are in various portions of our book, “Medical Internet Directory, 5th Edition, Updated July 2005, by Dr. Mike Gainey and Heuston.  Included are four, (1) a simple list of URLs and titles, (2) a more extensive Hierarchical List of Annotated Unique Sites, (3) a Cross Referenced Index, and (4) an Addendum.  The format structure in Chapter 4, was patterned after the traditional Domain Name System and added descriptive information about many sites, annotations not then available by the major search engines.  You can inspect this structure in our Web site section called "Content."  

We no longer publish hard cover books, but instead are trying online and on demand publishing.  You can review a useful group of topical categories and their lists of selected quality medical site links in our online free http://www.med-list.com, by Dr. Gainey. This new online directory uses 9 major categories and about 60 or more subcategories.  These link to medical sites on the Web. The design of med-list keeps the search function as simple as possible for general consumers who may not have much experience in medical library searching.  The PDF version of the 5th Edition referenced in this home page provides internal search capabilities using MS Word Edit and "Find" or "Search."  A new 6th Edition, PDF, is being developed that will present significantly improved annotated directory structure with a simplified format and internal search methods. 
    

As mentioned, our goal is to find better ways to search the healthcare Web, and to provide consultation to individuals, companies, medical organizations, libraries, and government agencies.  Our 10 years of advanced technology research experience have helped establish the research tools and background for further work and to report our findings through our books and consultations.   We can help you on Directory Design, Local and National Healthcare Lists, Simplified and Complex Listings of Specialized Subjects, Hierarchical Category Structures, Specialized Medical Keywords, Tree Diagrams, Filters, Networks, Cross Referencing, Convoluted Connectivity, Algorithm Models, and other aspect requirements.

Our research often involves quantitative and comparative analyses of medical Web site references presented by the typical search engines and big directories.   An example of the results of one of our earlier test comparing Google, Yahoo, and Profusion is shown below.  In this case, of course, Profusion no longer exists.  With reference to the column titled Directories, notice the number of site references listed by the various levels of categories.  Our interest has been in the number of site references claimed by the search engine, the number of references that can be opened, and the quality of the medical information on each site referenced, particularly in the first 10, 20, and 30 listed.  We also back-track some of the search engine references to check the validity of their two or three lines of abstracted descriptions.  
     

We also provide assistance for Cost Effective Search Strategies, Decision Support Analysis, Data Mining, Statistical Record Keeping and Analytical Inferences, and Database Development.