Tim Allen, M.S.
United States Department of Agriculture, Agricultural Research Service,
National Agricultural Library, Animal Welfare Information Center, Beltsville, Maryland, USA
Editor's note: This paper is adapted from a talk given at the Second World Congress on Alternatives and Animal Use in the Life Sciences held in Utrecht, The Netherlands in 1996.
It has been the Animal Welfare Information Center (AWIC) staff's experience that many researchers looking for alternatives to painful procedures or the use of animals search only Medline and ignore other databases that index biomedical, biological, a n d bioengineering literature, computer hardware and software, or audiovisuals. There are, however, many comprehensive and specialty databases that should be examined, each with strengths and weaknesses as they pertain to the use of animals or alternatives in research. And an effective multidatabase searching technique and terminology, such as that used by AWIC staff to find alternatives, is also important.
While these sources will provide the user with a wealth of information, they cannot provide information that is not made available by the scientific community. The publication of negative scientific results and/or specific conditions affecting animals used in experiments, use of alternatives terminology when abstracting journal articles or assigning keywords, and standardization of indexing terms for alternatives are several areas that would greatly benefit the search for alternatives.
[Contents | References]
Information in this section was obtained from DIALOG Blue Sheets, WWW, and experience using the database. DIALOG Blue Sheets, which contain descriptive information about the databases found on DIALOG, can be found at http://library.dialog.com/bluesheets/
DIALOG is an information service that provides access to more than 450 databases covering a range of disciplines. Subscription information can be obtained from Knight-Ridder Information, Inc., 2440 El Camino Real, Mountain View, California 94040, USA; tel: (415) 858-3785; WWW: http://www.dialog.com/
AGRICOLA is produced by the U.S. Department of Agriculture's National Agricultural Library (NAL) and covers the period from 1970 to the present. One of the major strengths of this database is its inclusion of a variety of information sources. The mate r ials covered in this database consist of articles, notes, letters, or chapters from peer-reviewed journals, popular magazines, newsletters, books, theses, patents, translations, audiovisuals, software, technical reports, and Congressional documents relat e d to agricultural and animal welfare issues. Subject coverage includes agriculture in its broadest sense, alternatives to animal testing, animal behavior, animal sciences, animal welfare, laboratory animal medicine, physiology, veterinary medicine, wildl i fe, and zoology. It is especially strong in veterinary anesthesiology for farm animals, dogs, and cats, and increasingly so for laboratory animals. The CAB Thesaurus is the basis for controlled vocabulary indexing. Both AGRICOLA and CAB Abstract s index using the phrase animal testing alternatives. A search performed in AGRICOLA in October 1996 using this phrase retrieved 693 records. One complaint is lack of information concerning drugs, chemicals, enzymes, etc. used in a study. It woul d be useful to have Chemical Abstract Service (CAS) registry numbers added in a descriptor field, or at the very least, list all pertinent compounds used in a study.
At the present time, AGRICOLA indexes more than 1,400 journals including many types of gray literature (pamphlets, conference reports, etc.) that are difficult to locate. AGRICOLA is available from DIALOG as File 10. AGRICOLA can also be searched on the WWW at http://agricola.nal.usda.gov/. Information about AGRICOLA can be found at http://www.nal.usda.gov/general_info/agricola/agricola.html
BIOSIS Previews is a comprehensive biological and biomedical database containing more than 12 million citations As with AGRICOLA, this database also includes a variety of information types such as journals, meeting abstracts, reviews, books, notes, le t ters, institutional and government reports, and research communications. Subject coverage includes all the life sciences such as agriculture, behavior, biotechnology, cell biology, pharmacology, physiology, radiation biology, toxicology, veterinary scien c e, etc. According to BIOSIS, there is no thesaurus or controlled vocabulary used (personal communication, 1996). Several years ago, BIOSIS considered either developing a separate database on alternatives or devising a new indexing system to make searchin g for alternatives more practical. It was finally decided that there was not enough interest in the topic within their user community to warrant a separate database and that the indexing system in place was adequate to address searching for alternatives. I n spite of those decisions, BIOSIS remains a tremendous resource for those looking to implement alternatives in their studies. BIOSIS has also added another database--Methods Finder--to its list of services. It is available at http://www.methodsfinder.org/home.html
At the present time, BIOSIS indexes almost 10,000 journals and monographs each year. It is available as File 5 (1969-present) or File 55 (1985-present) on DIALOG. The BIOSIS web site can be found at http://www.biosis.or g
CAB Abstracts is probably the world's most comprehensive database for agriculture, animal health, veterinary medicine, and increasingly, laboratory animal medicine, husbandry, and welfare. This database indexes more than 11,000 journals, as well as bo o ks, serial monographs, reports, newsletters, theses, symposium and conference proceedings, bibliographies, and translations. Most of the records contain abstracts. The CAB Thesaurus is the basis for controlled vocabulary indexing. It should be n o ted that AGRICOLA does not usually index materials that can be found in CAB Abstracts. Subject coverage includes agriculture in its broadest sense, animal health, animal production, animal sciences, animal testing alternatives, animal welfare, laborator y animal medicine and husbandry, veterinary medicine and science, and related topics. Unlike AGRICOLA, CAB provides very specific information with each record. CAS registry numbers makes it very simple to locate records pertaining to specific chemical com p ounds, and organism descriptors make it easy to search for information by species, breed, variety, etc. Although animal testing alternatives is used as an indexing term, a quick search of the database done in October 1996 using this phrase retri e ved only 30 records. The reason for the lack of records will be discussed later.
At the present time, CAB Abstracts contains more than 3 million records. It is available on DIALOG as File 50 (1972-present) and is also available as a CD-ROM. The CAB Abstracts Database web site can be found at http://www.cabi.org/home.asp
EMBASE is produced by Elsevier Science Publishers in the Netherlands and covers the period from 1974 to the present. It is an important database to use when looking for alternatives to animal research because it indexes articles and notes from journal s , conferences, symposia, and meetings. Subject coverage includes all aspects of human medicine and in vivo and in vitro biomedical research on topics including but not limited to anesthesiology, cancer, cardiovascular disorders, drug abuse, neurology, op h thalmology, pharmacology, physiology, psychiatry, surgery, toxicology, etc. In general, there is about a 10 percent to 30 percent overlap with Medline on materials indexed, depending on the subject area. EMBASE also provides extensive documentation of d r ugs and/or chemicals used in experiments. The July 1995 issue of Profile: the Excerpta Medicine Newsletter (Profile 1995) had an article on searching EMBASE for alternatives to animal testing. The article advises using their EMTREE term "animal welfare" a long with terms such as "animal experiment," "animal model," or "animal testing alternative." But using the term "animal welfare" retrieves only about 250 records from the entire database. Even so, the AWIC staff has found EMBASE to be a major source of i nformation on alternatives and always includes this database in any literature search.
EMBASE currently indexes more than 3,500 journals from over 110 countries and adds almost 400,000 new records annually to the 7 million records already indexed. It is available from DIALOG as File 72 (1974-present) or File 73 (1985-present). The Else v ier Science web site search utility can be found at http://www.elsevier.com/inca/search/
The website offers free searching of the tables of contents of all journals indexed by Elsevier.
MEDLINE is produced by the U. S. National Library of Medicine and covers the period from 1966 to the present. Medline is an exceptional database in that it provides comprehensive coverage of human medicine, animal-based and in vitro biomedical researc h , and veterinary medicine and science for both farm and laboratory animals. Unlike the databases already discussed, MEDLINE includes only information from peer reviewed journals. Indexing uses a controlled vocabulary known as MeSH (MEdical Subject Headin g s). The database also provides extensive information on drugs and chemicals used in experiments. This information can be found using trade names, chemical names, or CAS registry numbers. MEDLINE indexes articles using the phrases "animal testing alternat i ves" and "animal welfare." Unfortunately, a search performed in October 1996 found that "animal testing alternatives" retrieves only 404 records, while "animal welfare" finds only 2,006. However, it is an easy database to search using free text terms. < / P>
MEDLINE currently contains about 9 million records. It is available from DIALOG as File 154 (1985-present) or File 155 (1966-present). MEDLINE can also be searched using on the Internet at http://www.nlm.nih.gov/databases/databases_medline.html
PASCAL is produced by the French National Research Council's Institut de l'Information Scientifique et Technique and covers the period from 1973 to the present. This is a major multidisciplinary database that provides coverage of chemistry, biology, m e dicine, biomedical research, neurosciences, biotechnology, zoology (especially invertebrates), and the agricultural sciences. It does not cover animal husbandry or veterinary pathology. PASCAL also indexes materials from a variety of sources including jo u rnals, theses, conference proceedings, reports, books, and patents. Indexing is done with a controlled vocabulary of more than 80,000 terms. However, free text searching is very easy to perform.
At the present time, PASCAL indexes more than 8,500 journals. This accounts for 93 percent of the database of 11 million records; 7 percent comes from gray literature. According to the PASCAL web site, 65 percent of the literature indexed covers the m e dical and biological sciences. It is available from DIALOG as File 144 (1973-present). The English version WWW address is http://services.inist.fr/public/eng/conslt.htm
The homepage of the French National Research Council's Institut de l'Information Scientifique et Technique can be found at http://www.inist.fr/
TOXLINE is produced by the U.S. National Library of Medicine and covers the period from pre-1950 to the present. The types of publications indexed are journals, books, reports, theses, letters, meetings, project summaries, and unpublished materials. S u bject coverage includes adverse drug reactions, carcinogenesis, drug evaluation, mutagenesis, pollution, pesticides, herbicides, radiation, teratogenesis, and all other aspects of toxicology. TOXLINE uses the MESH terms discussed in MEDLINE as its contro l led vocabulary but again it is an easy database to search free text. Useful terms include "animal welfare" and "animal testing alternatives" although these terms alone are not sufficient to ensure a thorough search for alternative methods. Information on drugs or chemical compounds is easy to find using CAS registry numbers, chemical names, or tradenames.
At the present time, TOXLINE contains more than 2 million records. It is available from DIALOG as File 156. It can also be searched via Internet Grateful Med at the address given above.
Below is a comprehensive listing of databases, including some not mentioned in this discussion.
[Contents | References]
[Contents | References]
Several papers have discussed, in detail, strategies for retrieving information on alternatives from databases (Shevell and James 1995, Smith 1994, Snow 1990). Those interested in an in-depth look at developing search strategies are encouraged to read these articles. The first step in conducting a search is to have a clear understanding of the objectives and methods of the proposed study. Too often investigators ask for alternatives to very specific procedures without putting the procedure in the cont e xt of an experiment. To properly look for alternatives you have to know why the procedure is being performed and what the expected outcome is.
Once all pertinent information is at hand, the literature search strategy can be developed. It is convenient to conduct a search using the 3Rs as a guide. The first part of the search will examine the literature closely related to the proposed study f o r refinements to the proposed methods, methods or models that reduce the number of animals used, and to see if the proposed work duplicates previously published experiments (this is a requirement of the U.S. Animal Welfare Act). The terminology used in t h is part of the search will come from the area of study. Depending on the type of research, it might also be important to look for appropriate anesthetics, analgesics, methods of restraint, etc. Also remember to include both American and European spelling of words--for example, anesthesia, anaesthesia, anasthesia. It is also useful to determine that any anesthetics that are going to be administered do not interfere with any of the physiological variables that are being measured (for example, when methoxyf l urane is metabolized it produces fluoride ions, that may cause renal damage (Flecknell 1987)).
In the second part of the strategy, the remaining R--replacement--is considered. There may be some overlap with the first part of the search, in that alternative animal models may already be in hand. If not, then alternative mammalian and nonmammalian models should be considered. Below is a short list of useful terms that AWIC staff use. The ? is a truncation code used by DIALOG. For a more complete listing, see Searching Agricola for Animal Welfare (Clingerman et al. 1990) and Animal Wel f are Information Center Scope Notes (Swanson 1991). Both are available from AWIC.
animal testing alternative(s)
Alternative(s)-use with caution!
vitro (method, model, technique)
culture (cell, tissue, organ)
isolated (cell, tissue, organ)
simulat? (simulation(s), simulator(s))
virtual (surgery, reality)
video? (disc, display)
anesthe?, anasthe?, anaesthe?
analges?, sedative, anxiolytic
housing, faciliit?, caging
train?, educat?, teach?
welfare, pain, stress, distress
assay?, technique?, method?, proced?
environ? enrich?, toy, toys, play?
Note: Refinement alternatives are found using terms relevant to the area of study.
[Contents | References]
As noted, the phrase animal testing alternatives is used as an indexing term by AGRICOLA, MEDLINE, and other databases but fails to retrieve much useful information. Why? If the author of a scientific paper has not made it excruciatingly clea r that the paper discusses an alternative technique or model, the indexer usually does not have the leeway to add the alternatives tag as a descriptor, keyword, or MESH term. Similarly, in looking for animal models, many authors fail to mention the specie s or strain of animal in the abstract or keywords. Abstracts often don't mention anesthetics, analgesics, or sex of animals used. So this information don't make it into the database. Many authors do not even mention this type of information in the body of the paper. Papers often don't mention the husbandry and environmental conditions under which the animals are kept nor spell out problems encountered during the course of an experiment. This is the type of information that other scientists can use to avoi d the same mistakes and thus refine a procedure or find an alternative model. A very good accounting of how scientists can improve scientific writing and add to the body of literature on alternatives can be found in Morton (1992). To ensure that articles a re properly indexed, authors should include relevant alternatives terminology in the title, abstract, and keywords.
Another aspect of scientific publishing is the publishing of negative results or experiments that fail to confirm a hypothesis. While it may not seem useful at the time, the chances are very good that someone else is going to encounter the same proble m . Make it known! Similarly, laboratories conduct small pilot studies or experiments to determine things such as the effects on receptor function or density of using carbon dioxide as a pre-decapitation anesthetic. These studies may not warrant a peer-rev i ewed article but may be useful as a technical note or newsletter item. The point is to make that information available. The Animal Welfare Information Center Newsletter welcomes articles detailing this type of information.
Finally, one of the major problems in retrieving information on alternatives is the lack of standard indexing terminology. While proprietary indexing systems is important to the identity and profitability of commercial databases, it would enhance info r mation retrieval if databases could develop controlled vocabulary for alternatives terminology.
In an era when there is instant access to literally millions of scientific documents, taking the time to properly write up research will ensure that it is more easily retrieved and more widely cited.
[Contents | References]
Profile: The Excerpta Medica Newsletter (1995). Humane alternatives to animal testing. 12(1): 5.
Clingerman, K., C. Dowling, and J. Swanson (1990). Searching AGRICOLA for Animal Welfare. Search Tip Series:STS-03, National Agricultural Library, AWIC: Beltsville, Maryland, 20 pp.
Flecknell, P.A. (1987). Laboratory Animal Anesthesia. Academic Press:London, New York, p.28-29.
Morton, D. (1992). A fair press for animals. New Scientist 134(1816): 28-30.
Personal communication with Allan Clark (October 1996), Director of Product Development, Biosis.
Shevell, J.L. and M.L. James (1995). Search for animal alternatives and the role of the Information Specialist. Contemporary Topics 34(3): 65-68.
Smith, C. (1994). AWIC tips for searching for alternatives to animal research and testing. Lab Animal 23(3): 46-48.
Snow, B. (July, 1990). Online searching for alternatives to animal testing. Online p.94-97.
Swanson, J. (1991). Animal Welfare Information Center Scope Notes. AWIC Series # 6. National Agricultural Library, AWIC: Beltsville, Maryland, 11 pp.
Contents, Animal Welfare Information Center Bulletin
Top of Document
The Animal Welfare Information Center
U.S. Department of Agriculture
Agricultural Research Service
National Agricultural Library
10301 Baltimore Ave.
Beltsville, MD 20705-2351
Phone: (301) 504-6212
FAX: (301) 504-7125
Contact us: http://www.nal.usda.gov/awic/contact.php
Policies and Links