Bibliographic System in Health Care
Classifying
knowledge in books and other documents is in the domain of library and
information sciences. Books and other documents are considered “physical
objects” that can be classified in a number of ways; however, subject
classification is considered the most significant characteristic.
Whereas
scientists in a field identify the knowledge of the field, bibliographic
classifiers organize the knowledge produced by the scientists ( Landgridge ,
1992). The classification system is used to index the literature and thus
serves a purpose of location and retrieval of the indexed documents.
When the
classification system is accompanied by an alphabetical list of terms with
cross-references, it is called a thesaurus ( Landgridge , 1992).
The major
bibliographic classification: schemes dealing with the nursing literature are
implemented in computerized bibliographic database retrieval systems for
nursing and medicine. Computerized bibliographic databases based on
specialty-subject thesauri are available for many other reference disciplines
of psychology, education, sociology, and so forth.
Access To Bibliographical Retrieving System
Access
to bibliographic databases is either through a search service offered by the
primary developer of the classification system or is licensed for use by
bibliographic retrieval services that provide access to multiple bibliographic
databases. Fees for such services vary considerably.
Bibliographical Retrieving System As A Distinct
A
bibliographic retrieval system is a special type of information retrieval
system. The information that is stored (and retrieved) provides citations of
documents represented in the system.
Citations commonly include the article
author, title, and the exact location of the article (the title of the journal
in which it is published, journal volume and issue number, and pages).
Other
document types (books, videos, etc.), if incorporated in the system, have
descriptors appropriate for that document type. Other data that help to locate
a specific document-for example, accession number and author address-will be
added to the database by the producer of the system. Abstracts are usually
included.
Computerized Bibliographical Retrieving System And Components
Computerized
bibliographic retrieval systems have three components:
(a) the classification
system for the field of knowledge (subject headings, thesaurus, controlled
vocabulary).
(b) a database of documents indexed with the controlled vocabulary
of the classification system.
(c) the retrieval system search engine
(software).
The quality of retrieval is a function of all three elements. The
controlled vocabulary must adequately represent the literature in the field.
Terms from the controlled vocabulary must be accurately assigned to the
documents in the field.
The search software logic with which searches are done
facilitates certain types of searches and hinders others, thereby affecting the
quality of the retrieval.
Data Base Of Nursing Bibliographical Retrieving System
Nursing
has long been dissatisfied with bibliographic databases that index the nursing
literature. In part, this is because the vocabulary used by major systems has
not satisfactorily reflected nursing terminology.
Systems oriented towards
nursing literature overcome some of this difficulty by classifying things of
importance in nursing but not in medicine, such as nursing theoretical
frameworks.
Another
long-standing disappointment in the profession has been the inability to locate
nursing research by variables studied. This is because variable names are
sometimes so far out on the classification tree that they are usually not
suitable for subject headings.
This makes sense because variables usually
represent the new nomenclature in a field. These new terms are frequently
renamed or incorporated porated into another term or they may disappear
altogether.
Vocabularies need more stability than is characteristic of research
variable names. Because variable names are not always included in bibliographic
classifications, articles are not indexed by the names of variables studied in
the research.
The
results section of research articles, where the variable names reside, is
rarely used for assigning index terms (Horowitz, RS, & Fuller, 1982). The
identification of variable names as keywords by researchers is of little use.
Currently, there is no way to tell whether an author identified term or a
classifier-as-signed subject heading is a research variable name or just
another topic the article is “about.” It is fair to say that
“aboutness” indexing has a serious impact on retrievals of interest
to researchers (Weinberg, B. H.. 1987).
Nursing Research And Bibliographical Retrieving System
Research
document representation in nursing-related databases is a problem for several
reasons. First, if a controlled vocabulary is inadequate for any reason,
indexers cannot assign terms to adequately represent documents. Second,
research by nurses that is published outside the field may be in journals that
are not indexed by the database developers in the domain.
Third, bibliographic
data-bases are limited to the published research literature. Frequently, the
published literature fails to adequately reflect the knowledge being generated
in a field.
Cost of publishing and availability of reviewers limit the number
of articles that can be published. Publication bias against small studies with
nonsignificant findings and perhaps of parochial interest works against
publication of clinical research in nursing.
With more focus on statistical
meta-analysis strategies, these studies might be combined and thus yield
valuable new knowledge. The consequence of large amounts of fugitive research
lies not just in the invisibility of knowledge to the discipline but results in
a significant waste of resources to duplicate work that has already been done
or to identify work that needs to be done.
The strategy or “logic”
that software uses to search databases determines how documents can be
retrieved and how accurately the document set of interest can be retrieved.
Although other search strategies are upcoming available, the primary search
strategy used by bibliographic retrieval systems in nursing and related fields
is based on Boolean logic.
The searcher must fully understand the logic used by
the search system and how it is implemented in the database of interest.
Boolean Login And Nursing Bibliographical Retrieving System
Boolean
logic is based on set theory, which is a way of combining sets of things in
this case, search terms in documents. The operators, called Boolean operators,
dictate how the documents containing the terms will be combined.
The operator
and causes all the documents containing one term, x, and all documents
containing another search term, y, to be combined into the set of documents
that contain both x and y. This set of documents is called the search result.
Operator Result Of System
The
operator or results in a set of documents that have either the term x or y. It
includes the set of documents that have both x and y. Other common operators
are not, adjacent, includes, excludes, begins with.
operators a system makes available, the more accurate the search that can be
performed.
Operating Accuracy
Accuracy
is a generic term that refers to the concepts of sensitivity and specificity of
the search result. In Boolean search systems of bibliographic databases,
sensitivity and specificity are inversely related: the search either results in
many documents that are not relevant but includes most that are relevant or it
results.
In
addition to Boolean operators, common bibliographic database retrieval systems
will have tags that identify other salient features of documents in the field;
For example, the language the article is written in, document type, and whether
the article is about humans or other animals. These characteristics can be used
to further delimit a search.
Researchers
are interested in scientific findings, not documents (Doyle, 1986; Weinberg,
BH, 1987). All that can be obtained from a bibliographic database search is a
list of citations of documents or perhaps the full text of some documents that
may or may not contain research findings.
The accuracy of these searches can be
extremely low, depending on the complexity of the search.The
scientific knowledge is the research result or findings; however, bibliographic
classification is done to organize the scientific knowledge produced after it
has been embodied in documents ( Landgridge , 1992).
When viewed this way,
perhaps the results of research should not be part of a literature
classification system because the results are the knowledge, not the document
with the knowledge. Or perhaps this was and is the only legitimate method
available to library and information scientists when approaching the literature
of all disciplines.
Nonetheless,
research knowledge can be indexed by its variables (Graves, 1997; Weiner,
Stowe, Shirley, & Gilman, 1981) and linked to its source (the researchers);
if published, the dissemination history of the study (bibliographic citations)
can be provided.
The Virginia Henderson International Nursing Library makes the
nursing research that is in the Registry of Nursing Research accessible by
directly indexing the studies by variable names as well as by researcher and by
subject headings (see “The Virginia Henderson International Nursing Library”).