A good literature search is one that helps you:
Has your proposed experiment already been done by another researcher?
A good non-duplication literature search will help you find out and potentially avoid any unnecessary duplication.
Is there a way to answer your research question without using vertebrates? Can you answer it without using live animals at all?
A good literature search for alternatives will help you identify any options for replacing the vertebrate animals in your experiments.
Is there a way to improve the overall well-being of the animals in your experiments? Are there less invasive procedures you can do that will still help you answer your research question? Are there any appropriate methods for pain management you can implement? Is there any way to provide a more habitable and less stressful environment for the animals?
A good literature search for alternatives will help you identify any options for refining your experiments and procedures to limit or avoid pain and stress.
What is the lowest number of animals you can use in your experiments while still ensuring you have enough to determine statistical or biological significance?
A good literature search for alternatives will help you find relevant standards or guidance, or comparable studies you can use as a basis for a power analysis.
As part of your protocol application, you will need to include several important pieces of information about your search:
Therefore, before you begin searching for alternatives or to ensure non-duplication, you need to decide how you will manage and document your search. Keeping track of the entire search process, including the keywords you use, the databases you search, and the results you find, is a critical part of the protocol application.
The literature search should not be treated as "just another bureaucratic requirement". It should not be the last thing to do on your application checklist. A good literature search is one that is conducted throughout the planning stages of your research. It should inform your research design.
Federal law requires all researchers working with covered species (basically, all vertebrates, whether in the lab or in the field) to perform and document appropriate literature searches. Specifically, the Animal Welfare Act charges the principal investigator with considering "alternatives to any procedure likely to produce pain to or distress in an experimental animal" and providing a written narrative describing the methods and sources used to determine whether alternatives were available. It also charges PIs with providing "written assurance that the activities do not unnecessarily duplicate previous experiments".
Through literature searches, you can maybe find that there's another laboratory that has already researched what you're interested in, thereby saving you the time and expense of repeating it. You can also potentially find that the procedure you were planning on using is outdated or inefficient, and find a better procedure to use.