Why Conduct Literature Searches for Alternatives?
This article, first published in the ASLAP Newsletter, 1997, 30(3):19-23), has been updated (January 2020) to reflect current information.
USDA. NAL. Animal Welfare Information Center
"What is the literature search for alternatives and why do I have to do it?" is a frequent question asked by primary investigators who contact the Animal Welfare Information Center (AWIC). When the USDA/ Animal and Plant Health Inspection Service (APHIS)/Animal Care (AC) first published the Animal Care Policies (USDA, May, 1997), the questions about the alternatives literature search were a hot topic.
Although literature searches have been used to address the alternatives question since the concept of alternatives was introduced in the 1985 amendments to the Animal Welfare Act (Improved Standards for Laboratory Animals Act), there are always researchers entering the field and new IACUC members so the issue stays fresh. Some find the idea confusing or are unclear about what is required by law and what the benefits may be. While there are resources available that discuss how to do an alternatives search (Smith 1994, Stokes and Jensen 1995), this article will discuss the questions about what the search is and why it is conducted. Below are answers to typical questions confronted by AWIC staff about the alternatives requirement:
1. What are alternatives?
The concept of alternatives was conceived as the three Rs by W.M.S. Russell and R.L. Burch (Russell and Burch 1959) in their book The Principles of Humane Experimental Technique. The three Rs are Reduction in number of animals used, Refinement of methods to minimize pain and distress to the animals, and Replacement of the animal model with insentient material. In doing the literature search, it is important to recognize that alternatives are not just replacing the animal with a computer simulation or an in vitro procedure. An improved method of restraining the animal which involves positive reinforcement and minimizes the distress involved in capture and restraint is a refinement alternative. A thorough literature search of articles similar to the study proposed may help determine appropriate animal numbers (i.e. reduction alternative).
2. Where does the legislation say that I have to do a literature search?
The Animal Welfare Act (Title 7, U.S. Code), as written and approved by Congress, emphasizes minimizing pain and distress, but does not mention how alternatives consideration should be documented. It states in Section 13(a)(3)(B):
- "that the principal investigator consider alternatives to any procedure likely to produce pain or distress in an experimental animal;"
Title 9 of the Code of Federal Regulations (9CFR, Part 2, Sec. 2.31 (d)(1)(ii)) gives the USDA regulation on how consideration of alternatives should be accomplished. It mentions AWIC, which relies on multiple database literature searching, as a resource:
- "[The] IACUC shall determine that... The principal investigator has considered alternatives to procedures that may cause more than momentary or slight pain or distress to the animals, and has provided a written narrative description of the methods and sources, e.g., the Animal Welfare Information Center, used to determine that alternatives were not available;"
The Federal Register (Vol. 54, No. 168, Thursday, August 31, 1989) gives the USDA rationale for making the alternatives consideration a written requirement and suggests a series of databases that can be searched to document whether or not alternatives are available:
- "The principal investigator must provide a written narrative of the sources, such as biological abstracts, MEDLINE, the Current Research Information Service (CRIS), and the Animal Welfare Information Center that is operated by the National Agricultural Library. We believe that in fulfilling this requirement Committee members will discuss these efforts with the principal investigator in reviewing the proposed activity. We also believe that considerations of alternatives will be discussed during Committee meetings where proposed activities are presented for approval, and made part of the meeting minutes..."
The legislation indicates that the investigator must provide a written narrative which demonstrates to the IACUC that alternatives, useful or not, were at least considered in the experimental design. The literature search is suggested as the best way to demonstrate this. IACUC members, including the nonaffiliated member, a visiting AC inspector, or a member of the public can follow a printed search strategy, view the list of databases and keywords, and verify that the investigator has made a good faith effort to demonstrate whether or not alternatives exist and why he/she will or will not adopt them. The literature search if far less questionable than a check-off box or a sentence or two saying there are no alternatives written on the protocol form.
3. How should the protocol form be modified to reflect the literature search?
Every institution or IACUC develops its own protocol form so there is no universal standard or template. Some institutions do, however, contain a section which asks the investigator to describe efforts to determine if alternatives are available. At a minimum, the investigator must provide the search strategy used and list the databases searched. Some IACUCs ask investigators to attach a copy of the search while others suggest the investigator keep it on file for future reference.
4. Since many of our facility's records can be accessed by the public, won't inclusion of literature searches open us up to additional public scrutiny -- such as why wasn't this database searched or those keywords used?
The literature search provides a good faith effort on the part of the researcher and reflects well for the facility. Alternatives may be found that can lead to adoption of experimental methods that are less painful, use fewer animals, and make better scientific and economic sense. If alternatives are not found, what could be more positive than having documentation that assures that there is no other way and no more humane way to do the research than what is proposed? A simple checkbox saying a search was run or a statement relying on the experience of the investigator is far more open to scrutiny.
5. Is the literature search time-consuming?
Not necessarily. Searches with more specific keywords (i.e. chemical compound names, known alternatives, species being proposed), take less time than very broad generic online searches. Searching time can be reduced by preplanning the search strategy on paper or in free online databases. Keeping search sentences brief and later combining them also helps. With the help of a librarian, information specialist, or AWIC staff member, a search strategy can be designed that minimizes time online. It is up to the investigator, however, to find and review appropriate articles from the resulting citation list to determine if alternatives to the proposed study are found.
6. Is the search expensive?
Whether it is the institution or the investigator, somebody pays the cost for searching. Many university libraries give students and staff free access to online databases. Costs also depend on which databases are accessed. MEDLINE and AGRICOLA are available free of charge online. Some database systems like OVID or EBSCOHost offer unlimited use of a set of databases for one subscription fee (check with your institution's library to learn about what multi-database search systems are available for use). Others, like ProQuest Dialog, have a yearly fee which gives access to hundreds of databases, but there are additional fees associated with using each database every time you log on. Some facilities have included literature search costs as part of their operating budget or the costs are built into grant or study proposals.
7. Is the alternatives search different from searching for appropriate animal numbers?
Appropriate animal numbers can be justified by consulting with a statistician. Reduction of animal numbers, however, is an alternative. By running an alternatives search, citations are retrieved that cover studies similar to the one proposed. Animal numbers are often found in the materials and methods section of the articles.
8. What is the minimum number of databases that must be searched?
The literature search is a performance, not an engineering, standard. Although there is no minimum number, no one database reviews all the literature in all research fields. Databases do overlap somewhat in the journals they index and the subject areas they cover, but they also complement each other. Testing a new medical device, for example, might involve searching biomedical, engineering, and even computer sciences databases. The objectives of the search are to demonstrate whether or not alternatives are available and, if so, why or why can they not be used. A thorough search usually requires more than one database.
9. How do I run an alternatives search if I am doing toxicology testing and don't know what compounds are being tested?
If the type of compound is known, the search is easily run. If not, it is important to remember that alternatives means more than simply searching for a replacement technique. The investigator can search for a method which uses fewer animals, where mortality is not the endpoint, or techniques that minimize pain or distress. Even environmental enrichment can be considered an alternative.
10. The legislation specifically mentions alternatives to the painful procedure. Why do I need to search for any more than that specific part of the study?
The painful procedure must be examined in the context of the entire study. This information is often buried within the paper. Databases generally keyword search for words or phrases in the title, abstract, or descriptor. The painful procedures are sometimes, but not often, mentioned in those categories. Therefore, a broader view is needed to see if the study's ultimate objectives can be met with alternative methods.
Beside the legal aspect, there are many other benefits of the alternatives search. Many researchers run literature searches when designing a study. This helps determine if the research is original or "unnecessary duplication" (which must be documented). Such searches can easily be tailored to address alternatives. The search shows the investigator will stand behind his/her work and that it is as humane as possible. Sometimes, the adoption of an alternative method is more economical (i.e. using an in vitro model vs. housing a colony of animals). It may provide more meaningful data by avoiding confounding factors in the experimental design such as distress in the animal model. Hopefully, with a little patience and understanding, researchers will see the alternatives search as a valuable tool for improving the quality of research and not a dreaded Federal mandate.
Federal Register (Thur., Aug. 31, 1989), Vol. 54, No. 168, Part IV, Animal Welfare; Final rules.
Russell, W.M.S. and R.L. Burch (1959, reprinted 1992). The Principles of Humane Experimental Technique. Universities Federation for Animal Welfare: Herts, England.
Smith, C. (1994). AWIC tips for searching for alternatives to animal research and testing. Lab Animal March: 46-48.
Stokes, W.S. and D.J.B. Jensen (1995). Guidelines for Institutional Animal Care and Use Committees: consideration of alternatives. Contemporary Topics 34(3):51-60.
Title 7, U.S. Code (1997). Animal Welfare Act as Amended, Sec. 2131 et. seq.
Title 9, Code of Federal Regulations (1996). Chapter 1, Subchapter A, Animal Welfare.